persistent_range_query: add layer_map_test

Move same generic parameters to associated types.
Work around private fields of SumResult<T>

.github/PULL_REQUEST_TEMPLATE/release-pr.md

@@ -10,7 +10,7 @@
 <!-- List everything that should be done **before** release, any issues / setting changes / etc -->
 
 ### Checklist after release
-- [ ] Based on the merged commits write release notes and open a PR into `website` repo ([example](https://github.com/neondatabase/website/pull/120/files))
+- [ ] Based on the merged commits write release notes and open a PR into `website` repo ([example](https://github.com/neondatabase/website/pull/219/files))
 - [ ] Check [#dev-production-stream](https://neondb.slack.com/archives/C03F5SM1N02) Slack channel
 - [ ] Check [stuck projects page](https://console.neon.tech/admin/projects?sort=last_active&order=desc&stuck=true)
 - [ ] Check [recent operation failures](https://console.neon.tech/admin/operations?action=create_timeline%2Cstart_compute%2Cstop_compute%2Csuspend_compute%2Capply_config%2Cdelete_timeline%2Cdelete_tenant%2Ccreate_branch%2Ccheck_availability&sort=updated_at&order=desc&had_retries=some)

.github/actions/allure-report/action.yml

@@ -47,7 +47,7 @@ runs:
         else
           key=branch-$(echo ${GITHUB_REF#refs/heads/} | tr -c "[:alnum:]._-" "-")
         fi
-        echo "::set-output name=KEY::${key}"
+        echo "KEY=${key}" >> $GITHUB_OUTPUT
 
     - uses: actions/setup-java@v3
       if: ${{ inputs.action == 'generate' }}
@@ -186,7 +186,7 @@ runs:
         aws s3 cp --only-show-errors ./index.html "s3://${BUCKET}/${REPORT_PREFIX}/latest/index.html"
 
         echo "[Allure Report](${REPORT_URL})" >> ${GITHUB_STEP_SUMMARY}
-        echo "::set-output name=report-url::${REPORT_URL}"
+        echo "report-url=${REPORT_URL}" >> $GITHUB_OUTPUT
 
     - name: Release Allure lock
       if: ${{ inputs.action == 'generate' && always() }}

.github/actions/download/action.yml

@@ -34,7 +34,7 @@ runs:
         S3_KEY=$(aws s3api list-objects-v2 --bucket ${BUCKET} --prefix ${PREFIX%$GITHUB_RUN_ATTEMPT} | jq -r '.Contents[].Key' | grep ${FILENAME} | sort --version-sort | tail -1 || true)
         if [ -z "${S3_KEY}" ]; then
           if [ "${SKIP_IF_DOES_NOT_EXIST}" = "true" ]; then
-            echo '::set-output name=SKIPPED::true'
+            echo 'SKIPPED=true' >> $GITHUB_OUTPUT
             exit 0
           else
             echo 2>&1 "Neither s3://${BUCKET}/${PREFIX}/${FILENAME} nor its version from previous attempts exist"
@@ -42,7 +42,7 @@ runs:
           fi
         fi
 
-        echo '::set-output name=SKIPPED::false'
+        echo 'SKIPPED=false' >> $GITHUB_OUTPUT
 
         mkdir -p $(dirname $ARCHIVE)
         time aws s3 cp --only-show-errors s3://${BUCKET}/${S3_KEY} ${ARCHIVE}

.github/actions/neon-project-create/action.yml

@@ -41,8 +41,8 @@ runs:
             ;;
         esac
 
-        echo "::set-output name=api_host::${API_HOST}"
-        echo "::set-output name=region_id::${REGION_ID}"
+        echo "api_host=${API_HOST}" >> $GITHUB_OUTPUT
+        echo "region_id=${REGION_ID}" >> $GITHUB_OUTPUT
       env:
         ENVIRONMENT: ${{ inputs.environment }}
         REGION_ID: ${{ inputs.region_id }}
@@ -72,10 +72,10 @@ runs:
 
         dsn=$(echo $project | jq --raw-output '.roles[] | select(.name != "web_access") | .dsn')/main
         echo "::add-mask::${dsn}"
-        echo "::set-output name=dsn::${dsn}"
+        echo "dsn=${dsn}" >> $GITHUB_OUTPUT
 
         project_id=$(echo $project | jq --raw-output '.id')
-        echo "::set-output name=project_id::${project_id}"
+        echo "project_id=${project_id}" >> $GITHUB_OUTPUT
       env:
         API_KEY: ${{ inputs.api_key }}
         API_HOST: ${{ steps.parse-input.outputs.api_host }}

.github/actions/neon-project-delete/action.yml

@@ -32,7 +32,7 @@ runs:
             ;;
         esac
 
-        echo "::set-output name=api_host::${API_HOST}"
+        echo "api_host=${API_HOST}" >> $GITHUB_OUTPUT
       env:
         ENVIRONMENT: ${{ inputs.environment }}
 

.github/actions/run-python-test-set/action.yml

@@ -55,6 +55,22 @@ runs:
         name: neon-${{ runner.os }}-${{ inputs.build_type }}-artifact
         path: /tmp/neon
 
+    - name: Download Neon binaries for the previous release
+      if: inputs.build_type != 'remote'
+      uses: ./.github/actions/download
+      with:
+        name: neon-${{ runner.os }}-${{ inputs.build_type }}-artifact
+        path: /tmp/neon-previous
+        prefix: latest
+
+    - name: Download compatibility snapshot for Postgres 14
+      if: inputs.build_type != 'remote'
+      uses: ./.github/actions/download
+      with:
+        name: compatibility-snapshot-${{ inputs.build_type }}-pg14
+        path: /tmp/compatibility_snapshot_pg14
+        prefix: latest
+
     - name: Checkout
       if: inputs.needs_postgres_source == 'true'
       uses: actions/checkout@v3
@@ -76,10 +92,15 @@ runs:
     - name: Run pytest
       env:
         NEON_BIN: /tmp/neon/bin
+        COMPATIBILITY_NEON_BIN: /tmp/neon-previous/bin
+        COMPATIBILITY_POSTGRES_DISTRIB_DIR: /tmp/neon-previous/pg_install
         TEST_OUTPUT: /tmp/test_output
         BUILD_TYPE: ${{ inputs.build_type }}
         AWS_ACCESS_KEY_ID: ${{ inputs.real_s3_access_key_id }}
         AWS_SECRET_ACCESS_KEY: ${{ inputs.real_s3_secret_access_key }}
+        COMPATIBILITY_SNAPSHOT_DIR: /tmp/compatibility_snapshot_pg14
+        ALLOW_BACKWARD_COMPATIBILITY_BREAKAGE: contains(github.event.pull_request.labels.*.name, 'backward compatibility breakage')
+        ALLOW_FORWARD_COMPATIBILITY_BREAKAGE: contains(github.event.pull_request.labels.*.name, 'forward compatibility breakage')
       shell: bash -euxo pipefail {0}
       run: |
         # PLATFORM will be embedded in the perf test report
@@ -102,7 +123,12 @@ runs:
           exit 1
         fi
         if [[ "${{ inputs.run_in_parallel }}" == "true" ]]; then
+          # -n4 uses four processes to run tests via pytest-xdist
           EXTRA_PARAMS="-n4 $EXTRA_PARAMS"
+
+          # --dist=loadgroup points tests marked with @pytest.mark.xdist_group
+          # to the same worker to make @pytest.mark.order work with xdist
+          EXTRA_PARAMS="--dist=loadgroup $EXTRA_PARAMS"
         fi
 
         if [[ "${{ inputs.run_with_real_s3 }}" == "true" ]]; then
@@ -137,9 +163,9 @@ runs:
         # --verbose prints name of each test (helpful when there are
         # multiple tests in one file)
         # -rA prints summary in the end
-        # -n4 uses four processes to run tests via pytest-xdist
         # -s is not used to prevent pytest from capturing output, because tests are running
         # in parallel and logs are mixed between different tests
+        #
         mkdir -p $TEST_OUTPUT/allure/results
         "${cov_prefix[@]}" ./scripts/pytest \
           --junitxml=$TEST_OUTPUT/junit.xml \
@@ -154,8 +180,17 @@ runs:
           scripts/generate_and_push_perf_report.sh
         fi
 
+    - name: Upload compatibility snapshot for Postgres 14
+      if: github.ref_name == 'release'
+      uses: ./.github/actions/upload
+      with:
+        name: compatibility-snapshot-${{ inputs.build_type }}-pg14-${{ github.run_id }}
+        # The path includes a test name (test_create_snapshot) and directory that the test creates (compatibility_snapshot_pg14), keep the path in sync with the test
+        path: /tmp/test_output/test_create_snapshot/compatibility_snapshot_pg14/
+        prefix: latest
+
     - name: Create Allure report
-      if: always()
+      if: success() || failure()
       uses: ./.github/actions/allure-report
       with:
         action: store

.github/ansible/.gitignore

@@ -1,4 +1,5 @@
-zenith_install.tar.gz
-.zenith_current_version
 neon_install.tar.gz
 .neon_current_version
+
+collections/*
+!collections/.keep

.github/ansible/ansible.cfg

@@ -3,6 +3,7 @@
 localhost_warning = False
 host_key_checking = False
 timeout = 30
+collections_paths = ./collections
 
 [ssh_connection]
 ssh_args   = -F ./ansible.ssh.cfg

.github/ansible/deploy.yaml

@@ -1,7 +1,7 @@
 - name: Upload Neon binaries
   hosts: storage
   gather_facts: False
-  remote_user: admin
+  remote_user: "{{ remote_user }}"
 
   tasks:
 
@@ -14,7 +14,8 @@
       - safekeeper
 
     - name: inform about versions
-      debug: msg="Version to deploy - {{ current_version }}"
+      debug:
+        msg: "Version to deploy - {{ current_version }}"
       tags:
       - pageserver
       - safekeeper
@@ -35,7 +36,7 @@
 - name: Deploy pageserver
   hosts: pageservers
   gather_facts: False
-  remote_user: admin
+  remote_user: "{{ remote_user }}"
 
   tasks:
 
@@ -63,15 +64,29 @@
       tags:
       - pageserver
 
-    - name: update remote storage (s3) config
-      lineinfile:
-        path: /storage/pageserver/data/pageserver.toml
-        line: "{{ item }}"
-      loop:
-        - "[remote_storage]"
-        - "bucket_name = '{{ bucket_name }}'"
-        - "bucket_region = '{{ bucket_region }}'"
-        - "prefix_in_bucket = '{{ inventory_hostname }}'"
+    - name: read the existing remote pageserver config
+      ansible.builtin.slurp:
+        src: /storage/pageserver/data/pageserver.toml
+      register: _remote_ps_config
+      tags:
+      - pageserver
+
+    - name: parse the existing pageserver configuration
+      ansible.builtin.set_fact:
+        _existing_ps_config: "{{ _remote_ps_config['content'] | b64decode | sivel.toiletwater.from_toml }}"
+      tags:
+      - pageserver
+
+    - name: construct the final pageserver configuration dict
+      ansible.builtin.set_fact:
+        pageserver_config: "{{ pageserver_config_stub | combine({'id': _existing_ps_config.id }) }}"
+      tags:
+      - pageserver
+
+    - name: template the pageserver config
+      template:
+        src: templates/pageserver.toml.j2
+        dest: /storage/pageserver/data/pageserver.toml
       become: true
       tags:
       - pageserver
@@ -109,7 +124,7 @@
 - name: Deploy safekeeper
   hosts: safekeepers
   gather_facts: False
-  remote_user: admin
+  remote_user: "{{ remote_user }}"
 
   tasks:
 

.github/ansible/get_binaries.sh

@@ -23,6 +23,7 @@ docker cp ${ID}:/data/postgres_install.tar.gz .
 tar -xzf postgres_install.tar.gz -C neon_install
 mkdir neon_install/bin/
 docker cp ${ID}:/usr/local/bin/pageserver neon_install/bin/
+docker cp ${ID}:/usr/local/bin/pageserver_binutils neon_install/bin/
 docker cp ${ID}:/usr/local/bin/safekeeper neon_install/bin/
 docker cp ${ID}:/usr/local/bin/proxy neon_install/bin/
 docker cp ${ID}:/usr/local/v14/bin/ neon_install/v14/bin/

.github/ansible/neon-stress.hosts

@@ -1,20 +0,0 @@
-[pageservers]
-neon-stress-ps-1 console_region_id=1
-neon-stress-ps-2 console_region_id=1
-
-[safekeepers]
-neon-stress-sk-1 console_region_id=1
-neon-stress-sk-2 console_region_id=1
-neon-stress-sk-3 console_region_id=1
-
-[storage:children]
-pageservers
-safekeepers
-
-[storage:vars]
-env_name = neon-stress
-console_mgmt_base_url = http://neon-stress-console.local
-bucket_name           = neon-storage-ireland
-bucket_region         = eu-west-1
-etcd_endpoints        = etcd-stress.local:2379
-safekeeper_enable_s3_offload = false

.github/ansible/neon-stress.hosts.yaml

@@ -0,0 +1,31 @@
+storage:
+  vars:
+    bucket_name: neon-storage-ireland
+    bucket_region: eu-west-1
+    console_mgmt_base_url: http://neon-stress-console.local
+    etcd_endpoints: neon-stress-etcd.local:2379
+    safekeeper_enable_s3_offload: 'false'
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "{{ inventory_hostname }}"
+    safekeeper_s3_prefix: neon-stress/wal
+    hostname_suffix: ".local"
+    remote_user: admin
+  children:
+    pageservers:
+      hosts:
+        neon-stress-ps-1:
+          console_region_id: aws-eu-west-1
+        neon-stress-ps-2:
+          console_region_id: aws-eu-west-1
+    safekeepers:
+      hosts:
+        neon-stress-sk-1:
+          console_region_id: aws-eu-west-1
+        neon-stress-sk-2:
+          console_region_id: aws-eu-west-1
+        neon-stress-sk-3:
+          console_region_id: aws-eu-west-1

.github/ansible/prod.ap-southeast-1.hosts.yaml

@@ -0,0 +1,35 @@
+storage:
+  vars:
+    bucket_name: neon-prod-storage-ap-southeast-1
+    bucket_region: ap-southeast-1
+    console_mgmt_base_url: http://console-release.local
+    etcd_endpoints: etcd-0.ap-southeast-1.aws.neon.tech:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "pageserver/v1"
+    safekeeper_s3_prefix: safekeeper/v1/wal
+    hostname_suffix: ""
+    remote_user: ssm-user
+    ansible_aws_ssm_region: ap-southeast-1
+    ansible_aws_ssm_bucket_name: neon-prod-storage-ap-southeast-1
+    console_region_id: aws-ap-southeast-1
+
+  children:
+    pageservers:
+      hosts:
+        pageserver-0.ap-southeast-1.aws.neon.tech:
+          ansible_host:  i-064de8ea28bdb495b
+        pageserver-1.ap-southeast-1.aws.neon.tech:
+          ansible_host:  i-0b180defcaeeb6b93
+
+    safekeepers:
+      hosts:
+        safekeeper-0.ap-southeast-1.aws.neon.tech:
+          ansible_host:  i-0d6f1dc5161eef894
+        safekeeper-1.ap-southeast-1.aws.neon.tech:
+          ansible_host:  i-0e338adda8eb2d19f
+        safekeeper-2.ap-southeast-1.aws.neon.tech:
+          ansible_host:  i-04fb63634e4679eb9

.github/ansible/prod.eu-central-1.hosts.yaml

@@ -0,0 +1,35 @@
+storage:
+  vars:
+    bucket_name: neon-prod-storage-eu-central-1
+    bucket_region: eu-central-1
+    console_mgmt_base_url: http://console-release.local
+    etcd_endpoints: etcd-0.eu-central-1.aws.neon.tech:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "pageserver/v1"
+    safekeeper_s3_prefix: safekeeper/v1/wal
+    hostname_suffix: ""
+    remote_user: ssm-user
+    ansible_aws_ssm_region: eu-central-1
+    ansible_aws_ssm_bucket_name: neon-prod-storage-eu-central-1
+    console_region_id: aws-eu-central-1
+
+  children:
+    pageservers:
+      hosts:
+        pageserver-0.eu-central-1.aws.neon.tech:
+          ansible_host:  i-0cd8d316ecbb715be
+        pageserver-1.eu-central-1.aws.neon.tech:
+          ansible_host:  i-090044ed3d383fef0
+
+    safekeepers:
+      hosts:
+        safekeeper-0.eu-central-1.aws.neon.tech:
+          ansible_host:  i-0b238612d2318a050
+        safekeeper-1.eu-central-1.aws.neon.tech:
+          ansible_host:  i-07b9c45e5c2637cd4
+        safekeeper-2.eu-central-1.aws.neon.tech:
+          ansible_host:  i-020257302c3c93d88

.github/ansible/prod.us-east-2.hosts.yaml

@@ -0,0 +1,36 @@
+storage:
+  vars:
+    bucket_name: neon-prod-storage-us-east-2
+    bucket_region: us-east-2
+    console_mgmt_base_url: http://console-release.local
+    etcd_endpoints: etcd-0.us-east-2.aws.neon.tech:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "pageserver/v1"
+    safekeeper_s3_prefix: safekeeper/v1/wal
+    hostname_suffix: ""
+    remote_user: ssm-user
+    ansible_aws_ssm_region: us-east-2
+    ansible_aws_ssm_bucket_name: neon-prod-storage-us-east-2
+    console_region_id: aws-us-east-2
+
+  children:
+    pageservers:
+      hosts:
+        pageserver-0.us-east-2.aws.neon.tech:
+          ansible_host:  i-062227ba7f119eb8c
+        pageserver-1.us-east-2.aws.neon.tech:
+          ansible_host:  i-0b3ec0afab5968938
+
+    safekeepers:
+      hosts:
+        safekeeper-0.us-east-2.aws.neon.tech:
+          ansible_host:  i-0e94224750c57d346
+        safekeeper-1.us-east-2.aws.neon.tech:
+          ansible_host:  i-06d113fb73bfddeb0
+        safekeeper-2.us-east-2.aws.neon.tech:
+          ansible_host:  i-09f66c8e04afff2e8
+          

.github/ansible/production.hosts

@@ -1,20 +0,0 @@
-[pageservers]
-#zenith-1-ps-1 console_region_id=1
-zenith-1-ps-2 console_region_id=1
-zenith-1-ps-3 console_region_id=1
-
-[safekeepers]
-zenith-1-sk-1 console_region_id=1
-zenith-1-sk-2 console_region_id=1
-zenith-1-sk-3 console_region_id=1
-
-[storage:children]
-pageservers
-safekeepers
-
-[storage:vars]
-env_name = prod-1
-console_mgmt_base_url = http://console-release.local
-bucket_name           = zenith-storage-oregon
-bucket_region         = us-west-2
-etcd_endpoints        = zenith-1-etcd.local:2379

.github/ansible/production.hosts.yaml

@@ -0,0 +1,37 @@
+---
+storage:
+  vars:
+    console_mgmt_base_url: http://console-release.local
+    bucket_name: zenith-storage-oregon
+    bucket_region: us-west-2
+    etcd_endpoints: zenith-1-etcd.local:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "{{ inventory_hostname }}"
+    safekeeper_s3_prefix: prod-1/wal
+    hostname_suffix: ".local"
+    remote_user: admin
+
+  children:
+    pageservers:
+      hosts:
+        zenith-1-ps-2:
+          console_region_id: aws-us-west-2
+        zenith-1-ps-3:
+          console_region_id: aws-us-west-2
+        zenith-1-ps-4:
+          console_region_id: aws-us-west-2
+        zenith-1-ps-5:
+          console_region_id: aws-us-west-2
+
+    safekeepers:
+      hosts:
+        zenith-1-sk-1:
+          console_region_id: aws-us-west-2
+        zenith-1-sk-2:
+          console_region_id: aws-us-west-2
+        zenith-1-sk-3:
+          console_region_id: aws-us-west-2

.github/ansible/scripts/init_pageserver.sh

@@ -12,18 +12,19 @@ cat <<EOF | tee /tmp/payload
   "version": 1,
   "host": "${HOST}",
   "port": 6400,
-  "region_id": {{ console_region_id }},
+  "region_id": "{{ console_region_id }}",
   "instance_id": "${INSTANCE_ID}",
   "http_host": "${HOST}",
-  "http_port": 9898
+  "http_port": 9898,
+  "active": false
 }
 EOF
 
 # check if pageserver already registered or not
-if ! curl -sf -X PATCH -d '{}' {{ console_mgmt_base_url }}/api/v1/pageservers/${INSTANCE_ID} -o /dev/null; then
+if ! curl -sf -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" {{ console_mgmt_base_url }}/management/api/v2/pageservers/${INSTANCE_ID} -o /dev/null; then
 
     # not registered, so register it now
-    ID=$(curl -sf -X POST {{ console_mgmt_base_url }}/api/v1/pageservers -d@/tmp/payload | jq -r '.ID')
+    ID=$(curl -sf -X POST -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" {{ console_mgmt_base_url }}/management/api/v2/pageservers -d@/tmp/payload | jq -r '.id')
 
     # init pageserver
     sudo -u pageserver /usr/local/bin/pageserver -c "id=${ID}" -c "pg_distrib_dir='/usr/local'" --init -D /storage/pageserver/data

.github/ansible/scripts/init_safekeeper.sh

@@ -14,18 +14,18 @@ cat <<EOF | tee /tmp/payload
   "host": "${HOST}",
   "port": 6500,
   "http_port": 7676,
-  "region_id": {{ console_region_id }},
+  "region_id": "{{ console_region_id }}",
   "instance_id": "${INSTANCE_ID}",
-  "availability_zone_id": "${AZ_ID}"
+  "availability_zone_id": "${AZ_ID}",
+  "active": false
 }
 EOF
 
 # check if safekeeper already registered or not
-if ! curl -sf -X PATCH -d '{}' {{ console_mgmt_base_url }}/api/v1/safekeepers/${INSTANCE_ID} -o /dev/null; then
+if ! curl -sf -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" {{ console_mgmt_base_url }}/management/api/v2/safekeepers/${INSTANCE_ID} -o /dev/null; then
 
     # not registered, so register it now
-    ID=$(curl -sf -X POST {{ console_mgmt_base_url }}/api/v1/safekeepers -d@/tmp/payload | jq -r '.ID')
-
+    ID=$(curl -sf -X POST -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" {{ console_mgmt_base_url }}/management/api/v2/safekeepers -d@/tmp/payload | jq -r '.id')
     # init safekeeper
     sudo -u safekeeper /usr/local/bin/safekeeper --id ${ID} --init -D /storage/safekeeper/data
 fi

.github/ansible/ssm_config

@@ -0,0 +1,2 @@
+ansible_connection: aws_ssm
+ansible_python_interpreter: /usr/bin/python3

.github/ansible/staging.eu-west-1.hosts.yaml

@@ -0,0 +1,33 @@
+storage:
+  vars:
+    bucket_name: neon-dev-storage-eu-west-1
+    bucket_region: eu-west-1
+    console_mgmt_base_url: http://console-staging.local
+    etcd_endpoints: etcd-0.eu-west-1.aws.neon.build:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "pageserver/v1"
+    safekeeper_s3_prefix: safekeeper/v1/wal
+    hostname_suffix: ""
+    remote_user: ssm-user
+    ansible_aws_ssm_region: eu-west-1
+    ansible_aws_ssm_bucket_name: neon-dev-storage-eu-west-1
+    console_region_id: aws-eu-west-1
+
+  children:
+    pageservers:
+      hosts:
+        pageserver-0.eu-west-1.aws.neon.build:
+          ansible_host: i-01d496c5041c7f34c
+
+    safekeepers:
+      hosts:
+        safekeeper-0.eu-west-1.aws.neon.build:
+          ansible_host: i-05226ef85722831bf
+        safekeeper-1.eu-west-1.aws.neon.build:
+          ansible_host: i-06969ee1bf2958bfc
+        safekeeper-2.eu-west-1.aws.neon.build:
+          ansible_host: i-087892e9625984a0b

.github/ansible/staging.hosts

@@ -1,25 +0,0 @@
-[pageservers]
-#zenith-us-stage-ps-1 console_region_id=27
-zenith-us-stage-ps-2 console_region_id=27
-zenith-us-stage-ps-3 console_region_id=27
-zenith-us-stage-ps-4 console_region_id=27
-zenith-us-stage-test-ps-1 console_region_id=28
-
-[safekeepers]
-zenith-us-stage-sk-4 console_region_id=27
-zenith-us-stage-sk-5 console_region_id=27
-zenith-us-stage-sk-6 console_region_id=27
-zenith-us-stage-test-sk-1 console_region_id=28
-zenith-us-stage-test-sk-2 console_region_id=28
-zenith-us-stage-test-sk-3 console_region_id=28
-
-[storage:children]
-pageservers
-safekeepers
-
-[storage:vars]
-env_name = us-stage
-console_mgmt_base_url = http://console-staging.local
-bucket_name           = zenith-staging-storage-us-east-1
-bucket_region         = us-east-1
-etcd_endpoints        = zenith-us-stage-etcd.local:2379

.github/ansible/staging.hosts.yaml

@@ -0,0 +1,34 @@
+storage:
+  vars:
+    bucket_name: zenith-staging-storage-us-east-1
+    bucket_region: us-east-1
+    console_mgmt_base_url: http://console-staging.local
+    etcd_endpoints: etcd-0.us-east-2.aws.neon.build:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "{{ inventory_hostname }}"
+    safekeeper_s3_prefix: us-stage/wal
+    hostname_suffix: ".local"
+    remote_user: admin
+
+  children:
+    pageservers:
+      hosts:
+        zenith-us-stage-ps-2:
+          console_region_id: aws-us-east-1
+        zenith-us-stage-ps-3:
+          console_region_id: aws-us-east-1
+        zenith-us-stage-ps-4:
+          console_region_id: aws-us-east-1
+
+    safekeepers:
+      hosts:
+        zenith-us-stage-sk-4:
+          console_region_id: aws-us-east-1
+        zenith-us-stage-sk-5:
+          console_region_id: aws-us-east-1
+        zenith-us-stage-sk-6:
+          console_region_id: aws-us-east-1

.github/ansible/staging.us-east-2.hosts.yaml

@@ -0,0 +1,35 @@
+storage:
+  vars:
+    bucket_name: neon-staging-storage-us-east-2
+    bucket_region: us-east-2
+    console_mgmt_base_url: http://console-staging.local
+    etcd_endpoints: etcd-0.us-east-2.aws.neon.build:2379
+    pageserver_config_stub:
+      pg_distrib_dir: /usr/local
+      remote_storage:
+        bucket_name: "{{ bucket_name }}"
+        bucket_region: "{{ bucket_region }}"
+        prefix_in_bucket: "pageserver/v1"
+    safekeeper_s3_prefix: safekeeper/v1/wal
+    hostname_suffix: ""
+    remote_user: ssm-user
+    ansible_aws_ssm_region: us-east-2
+    ansible_aws_ssm_bucket_name: neon-staging-storage-us-east-2
+    console_region_id: aws-us-east-2
+
+  children:
+    pageservers:
+      hosts:
+        pageserver-0.us-east-2.aws.neon.build:
+          ansible_host: i-0c3e70929edb5d691
+        pageserver-1.us-east-2.aws.neon.build:
+          ansible_host: i-0565a8b4008aa3f40
+
+    safekeepers:
+      hosts:
+        safekeeper-0.us-east-2.aws.neon.build:
+          ansible_host: i-027662bd552bf5db0
+        safekeeper-1.us-east-2.aws.neon.build:
+          ansible_host: i-0171efc3604a7b907
+        safekeeper-2.us-east-2.aws.neon.build:
+          ansible_host: i-0de0b03a51676a6ce

.github/ansible/systemd/pageserver.service

@@ -1,5 +1,5 @@
 [Unit]
-Description=Zenith pageserver
+Description=Neon pageserver
 After=network.target auditd.service
 
 [Service]

.github/ansible/systemd/safekeeper.service

@@ -1,12 +1,12 @@
 [Unit]
-Description=Zenith safekeeper
+Description=Neon safekeeper
 After=network.target auditd.service
 
 [Service]
 Type=simple
 User=safekeeper
 Environment=RUST_BACKTRACE=1 NEON_REPO_DIR=/storage/safekeeper/data LD_LIBRARY_PATH=/usr/local/v14/lib
-ExecStart=/usr/local/bin/safekeeper -l {{ inventory_hostname }}.local:6500 --listen-http {{ inventory_hostname }}.local:7676 -D /storage/safekeeper/data --broker-endpoints={{ etcd_endpoints }} --remote-storage='{bucket_name="{{bucket_name}}", bucket_region="{{bucket_region}}", prefix_in_bucket="{{ env_name }}/wal"}'
+ExecStart=/usr/local/bin/safekeeper -l {{ inventory_hostname }}{{ hostname_suffix }}:6500 --listen-http {{ inventory_hostname }}{{ hostname_suffix }}:7676 -D /storage/safekeeper/data --broker-endpoints={{ etcd_endpoints }} --remote-storage='{bucket_name="{{bucket_name}}", bucket_region="{{bucket_region}}", prefix_in_bucket="{{ safekeeper_s3_prefix }}"}'
 ExecReload=/bin/kill -HUP $MAINPID
 KillMode=mixed
 KillSignal=SIGINT

.github/ansible/templates/pageserver.toml.j2

@@ -0,0 +1 @@
+{{ pageserver_config | sivel.toiletwater.to_toml }}

.github/helm-values/dev-eu-west-1-zeta.neon-proxy-scram.yaml

@@ -0,0 +1,31 @@
+# Helm chart values for neon-proxy-scram.
+# This is a YAML-formatted file.
+
+image:
+  repository: neondatabase/neon
+
+settings:
+  authBackend: "console"
+  authEndpoint: "http://console-staging.local/management/api/v2"
+  domain: "*.eu-west-1.aws.neon.build"
+
+# -- Additional labels for neon-proxy pods
+podLabels:
+  zenith_service: proxy-scram
+  zenith_env: dev
+  zenith_region: eu-west-1
+  zenith_region_slug: eu-west-1
+
+exposedService:
+  annotations:
+    service.beta.kubernetes.io/aws-load-balancer-type: external
+    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
+    service.beta.kubernetes.io/aws-load-balancer-scheme: internet-facing
+    external-dns.alpha.kubernetes.io/hostname: eu-west-1.aws.neon.build
+
+#metrics:
+#  enabled: true
+#  serviceMonitor:
+#    enabled: true
+#    selector:
+#      release: kube-prometheus-stack

.github/helm-values/dev-us-east-2-beta.neon-proxy-scram.yaml

@@ -0,0 +1,31 @@
+# Helm chart values for neon-proxy-scram.
+# This is a YAML-formatted file.
+
+image:
+  repository: neondatabase/neon
+
+settings:
+  authBackend: "console"
+  authEndpoint: "http://console-staging.local/management/api/v2"
+  domain: "*.us-east-2.aws.neon.build"
+
+# -- Additional labels for neon-proxy pods
+podLabels:
+  zenith_service: proxy-scram
+  zenith_env: dev
+  zenith_region: us-east-2
+  zenith_region_slug: us-east-2
+
+exposedService:
+  annotations:
+    service.beta.kubernetes.io/aws-load-balancer-type: external
+    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
+    service.beta.kubernetes.io/aws-load-balancer-scheme: internet-facing
+    external-dns.alpha.kubernetes.io/hostname: us-east-2.aws.neon.build
+
+#metrics:
+#  enabled: true
+#  serviceMonitor:
+#    enabled: true
+#    selector:
+#      release: kube-prometheus-stack

.github/helm-values/prod-ap-southeast-1-epsilon.neon-proxy-scram.yaml

@@ -0,0 +1,31 @@
+# Helm chart values for neon-proxy-scram.
+# This is a YAML-formatted file.
+
+image:
+  repository: neondatabase/neon
+
+settings:
+  authBackend: "console"
+  authEndpoint: "http://console-release.local/management/api/v2"
+  domain: "*.ap-southeast-1.aws.neon.tech"
+
+# -- Additional labels for neon-proxy pods
+podLabels:
+  zenith_service: proxy-scram
+  zenith_env: prod
+  zenith_region: ap-southeast-1
+  zenith_region_slug: ap-southeast-1
+
+exposedService:
+  annotations:
+    service.beta.kubernetes.io/aws-load-balancer-type: external
+    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
+    service.beta.kubernetes.io/aws-load-balancer-scheme: internet-facing
+    external-dns.alpha.kubernetes.io/hostname: ap-southeast-1.aws.neon.tech
+
+#metrics:
+#  enabled: true
+#  serviceMonitor:
+#    enabled: true
+#    selector:
+#      release: kube-prometheus-stack

.github/helm-values/prod-eu-central-1-gamma.neon-proxy-scram.yaml

@@ -0,0 +1,31 @@
+# Helm chart values for neon-proxy-scram.
+# This is a YAML-formatted file.
+
+image:
+  repository: neondatabase/neon
+
+settings:
+  authBackend: "console"
+  authEndpoint: "http://console-release.local/management/api/v2"
+  domain: "*.eu-central-1.aws.neon.tech"
+
+# -- Additional labels for neon-proxy pods
+podLabels:
+  zenith_service: proxy-scram
+  zenith_env: prod
+  zenith_region: eu-central-1
+  zenith_region_slug: eu-central-1
+
+exposedService:
+  annotations:
+    service.beta.kubernetes.io/aws-load-balancer-type: external
+    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
+    service.beta.kubernetes.io/aws-load-balancer-scheme: internet-facing
+    external-dns.alpha.kubernetes.io/hostname: eu-central-1.aws.neon.tech
+
+#metrics:
+#  enabled: true
+#  serviceMonitor:
+#    enabled: true
+#    selector:
+#      release: kube-prometheus-stack

.github/helm-values/prod-us-east-2-delta.neon-proxy-scram.yaml

@@ -0,0 +1,31 @@
+# Helm chart values for neon-proxy-scram.
+# This is a YAML-formatted file.
+
+image:
+  repository: neondatabase/neon
+
+settings:
+  authBackend: "console"
+  authEndpoint: "http://console-release.local/management/api/v2"
+  domain: "*.us-east-2.aws.neon.tech"
+
+# -- Additional labels for neon-proxy pods
+podLabels:
+  zenith_service: proxy-scram
+  zenith_env: prod
+  zenith_region: us-east-2
+  zenith_region_slug: us-east-2
+
+exposedService:
+  annotations:
+    service.beta.kubernetes.io/aws-load-balancer-type: external
+    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
+    service.beta.kubernetes.io/aws-load-balancer-scheme: internet-facing
+    external-dns.alpha.kubernetes.io/hostname: us-east-2.aws.neon.tech
+
+#metrics:
+#  enabled: true
+#  serviceMonitor:
+#    enabled: true
+#    selector:
+#      release: kube-prometheus-stack

.github/workflows/benchmarking.yml

@@ -46,7 +46,7 @@ jobs:
     runs-on: [self-hosted, zenith-benchmarker]
 
     env:
-      POSTGRES_DISTRIB_DIR: /tmp/pg_install
+      POSTGRES_DISTRIB_DIR: /usr/pgsql
       DEFAULT_PG_VERSION: 14
 
     steps:
@@ -138,24 +138,35 @@ jobs:
         SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
   pgbench-compare:
-    env:
-      TEST_PG_BENCH_DURATIONS_MATRIX: "60m"
-      TEST_PG_BENCH_SCALES_MATRIX: "10gb"
-      POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: 14
-      TEST_OUTPUT: /tmp/test_output
-      BUILD_TYPE: remote
-      SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref == 'refs/heads/main' ) }}
-
     strategy:
       fail-fast: false
       matrix:
         # neon-captest-new: Run pgbench in a freshly created project
         # neon-captest-reuse: Same, but reusing existing project
         # neon-captest-prefetch: Same, with prefetching enabled (new project)
-        platform: [ neon-captest-new, neon-captest-reuse, neon-captest-prefetch, rds-aurora ]
+        # rds-aurora: Aurora Postgres Serverless v2 with autoscaling from 0.5 to 2 ACUs
+        # rds-postgres: RDS Postgres db.m5.large instance (2 vCPU, 8 GiB) with gp3 EBS storage
+        platform: [ neon-captest-new, neon-captest-reuse, neon-captest-prefetch, rds-postgres ]
+        db_size: [ 10gb ]
+        include:
+          - platform: neon-captest-new
+            db_size: 50gb
+          - platform: neon-captest-prefetch
+            db_size: 50gb
+          - platform: rds-aurora
+            db_size: 50gb
 
-    runs-on: dev
+    env:
+      TEST_PG_BENCH_DURATIONS_MATRIX: "60m"
+      TEST_PG_BENCH_SCALES_MATRIX: ${{ matrix.db_size }}
+      POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
+      DEFAULT_PG_VERSION: 14
+      TEST_OUTPUT: /tmp/test_output
+      BUILD_TYPE: remote
+      SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref == 'refs/heads/main' ) }}
+      PLATFORM: ${{ matrix.platform }}
+
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rustlegacy:pinned
       options: --init
@@ -178,7 +189,7 @@ jobs:
         echo "${POSTGRES_DISTRIB_DIR}/v${DEFAULT_PG_VERSION}/bin" >> $GITHUB_PATH
 
     - name: Create Neon Project
-      if: matrix.platform != 'neon-captest-reuse'
+      if: contains(fromJson('["neon-captest-new", "neon-captest-prefetch"]'), matrix.platform)
       id: create-neon-project
       uses: ./.github/actions/neon-project-create
       with:
@@ -198,17 +209,18 @@ jobs:
           rds-aurora)
             CONNSTR=${{ secrets.BENCHMARK_RDS_CONNSTR }}
             ;;
+          rds-postgres)
+            CONNSTR=${{ secrets.BENCHMARK_RDS_POSTGRES_CONNSTR }}
+            ;;
           *)
-            echo 2>&1 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neon-captest-reuse', 'neon-captest-new', 'neon-captest-prefetch' or 'rds-aurora'"
+            echo 2>&1 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neon-captest-reuse', 'neon-captest-new', 'neon-captest-prefetch', 'rds-aurora', or 'rds-postgres'"
             exit 1
             ;;
         esac
 
-        echo "::set-output name=connstr::${CONNSTR}"
+        echo "connstr=${CONNSTR}" >> $GITHUB_OUTPUT
 
         psql ${CONNSTR} -c "SELECT version();"
-      env:
-        PLATFORM: ${{ matrix.platform }}
 
     - name: Set database options
       if: matrix.platform == 'neon-captest-prefetch'
@@ -227,7 +239,6 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_init
       env:
-        PLATFORM: ${{ matrix.platform }}
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -241,7 +252,6 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_simple_update
       env:
-        PLATFORM: ${{ matrix.platform }}
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -255,20 +265,19 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_select_only
       env:
-        PLATFORM: ${{ matrix.platform }}
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
 
     - name: Create Allure report
-      if: always()
+      if: success() || failure()
       uses: ./.github/actions/allure-report
       with:
         action: generate
         build_type: ${{ env.BUILD_TYPE }}
 
     - name: Delete Neon Project
-      if: ${{ matrix.platform != 'neon-captest-reuse' && always() }}
+      if: ${{ steps.create-neon-project.outputs.project_id && always() }}
       uses: ./.github/actions/neon-project-delete
       with:
         environment: dev

.github/workflows/build_and_test.yml

@@ -18,8 +18,8 @@ env:
 
 jobs:
   tag:
-    runs-on: dev
-    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/base:latest
+    runs-on: [ self-hosted, dev, x64 ]
+    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/base:pinned
     outputs:
       build-tag: ${{steps.build-tag.outputs.tag}}
 
@@ -35,18 +35,18 @@ jobs:
           echo ref:$GITHUB_REF_NAME
           echo rev:$(git rev-list --count HEAD)
           if [[ "$GITHUB_REF_NAME" == "main" ]]; then
-            echo "::set-output name=tag::$(git rev-list --count HEAD)"
+            echo "tag=$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
           elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
-            echo "::set-output name=tag::release-$(git rev-list --count HEAD)"
+            echo "tag=release-$(git rev-list --count HEAD)" >> $GITHUB_OUTPUT
           else
             echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
-            echo "::set-output name=tag::$GITHUB_RUN_ID"
+            echo "tag=$GITHUB_RUN_ID" >> $GITHUB_OUTPUT
           fi
         shell: bash
         id: build-tag
 
   build-neon:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
       options: --init
@@ -78,12 +78,12 @@ jobs:
 
       - name: Set pg 14 revision for caching
         id: pg_v14_rev
-        run: echo ::set-output name=pg_rev::$(git rev-parse HEAD:vendor/postgres-v14)
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v14) >> $GITHUB_OUTPUT
         shell: bash -euxo pipefail {0}
 
       - name: Set pg 15 revision for caching
         id: pg_v15_rev
-        run: echo ::set-output name=pg_rev::$(git rev-parse HEAD:vendor/postgres-v15)
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v15) >> $GITHUB_OUTPUT
         shell: bash -euxo pipefail {0}
 
       # Set some environment variables used by all the steps.
@@ -127,8 +127,8 @@ jobs:
             target/
           # Fall back to older versions of the key, if no cache for current Cargo.lock was found
           key: |
-            v9-${{ runner.os }}-${{ matrix.build_type }}-cargo-${{ hashFiles('Cargo.lock') }}
-            v9-${{ runner.os }}-${{ matrix.build_type }}-cargo-
+            v10-${{ runner.os }}-${{ matrix.build_type }}-cargo-${{ hashFiles('Cargo.lock') }}
+            v10-${{ runner.os }}-${{ matrix.build_type }}-cargo-
 
       - name: Cache postgres v14 build
         id: cache_pg_14
@@ -236,7 +236,7 @@ jobs:
         uses: ./.github/actions/save-coverage-data
 
   regress-tests:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
       options: --init
@@ -268,34 +268,8 @@ jobs:
         if: matrix.build_type == 'debug'
         uses: ./.github/actions/save-coverage-data
 
-  upload-latest-artifacts:
-    runs-on: dev
-    container:
-      image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
-      options: --init
-    needs: [ regress-tests ]
-    if: github.ref_name == 'main'
-    steps:
-      - name: Copy Neon artifact to the latest directory
-        shell: bash -euxo pipefail {0}
-        env:
-          BUCKET: neon-github-public-dev
-          PREFIX: artifacts/${{ github.run_id }}
-        run: |
-          for build_type in debug release; do
-            FILENAME=neon-${{ runner.os }}-${build_type}-artifact.tar.zst
-
-            S3_KEY=$(aws s3api list-objects-v2 --bucket ${BUCKET} --prefix ${PREFIX} | jq -r '.Contents[].Key' | grep ${FILENAME} | sort --version-sort | tail -1 || true)
-            if [ -z "${S3_KEY}" ]; then
-              echo 2>&1 "Neither s3://${BUCKET}/${PREFIX}/${FILENAME} nor its version from previous attempts exist"
-              exit 1
-            fi
-
-            time aws s3 cp --only-show-errors s3://${BUCKET}/${S3_KEY} s3://${BUCKET}/artifacts/latest/${FILENAME}
-          done
-
   benchmarks:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
       options: --init
@@ -326,12 +300,12 @@ jobs:
       # while coverage is currently collected for the debug ones
 
   merge-allure-report:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
       options: --init
     needs: [ regress-tests, benchmarks ]
-    if: always()
+    if: success() || failure()
     strategy:
       fail-fast: false
       matrix:
@@ -364,7 +338,7 @@ jobs:
           DATABASE_URL="$TEST_RESULT_CONNSTR" poetry run python3 scripts/ingest_regress_test_result.py --revision ${SHA} --reference ${GITHUB_REF} --build-type ${BUILD_TYPE} --ingest suites.json
 
   coverage-report:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
       options: --init
@@ -389,7 +363,7 @@ jobs:
             !~/.cargo/registry/src
             ~/.cargo/git/
             target/
-          key: v9-${{ runner.os }}-${{ matrix.build_type }}-cargo-${{ hashFiles('Cargo.lock') }}
+          key: v10-${{ runner.os }}-${{ matrix.build_type }}-cargo-${{ hashFiles('Cargo.lock') }}
 
       - name: Get Neon artifact
         uses: ./.github/actions/download
@@ -441,15 +415,19 @@ jobs:
         shell: bash -euxo pipefail {0}
 
   trigger-e2e-tests:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/base:pinned
       options: --init
-    needs: [ build-neon ]
+    needs: [ push-docker-hub, tag ]
     steps:
       - name: Set PR's status to pending and request a remote CI test
         run: |
+          # For pull requests, GH Actions set "github.sha" variable to point at a fake merge commit
+          # but we need to use a real sha of a latest commit in the PR's branch for the e2e job,
+          # to place a job run status update later.
           COMMIT_SHA=${{ github.event.pull_request.head.sha }}
+          # For non-PR kinds of runs, the above will produce an empty variable, pick the original sha value for those
           COMMIT_SHA=${COMMIT_SHA:-${{ github.sha }}}
 
           REMOTE_REPO="${{ github.repository_owner }}/cloud"
@@ -475,12 +453,15 @@ jobs:
               \"inputs\": {
                 \"ci_job_name\": \"neon-cloud-e2e\",
                 \"commit_hash\": \"$COMMIT_SHA\",
-                \"remote_repo\": \"${{ github.repository }}\"
+                \"remote_repo\": \"${{ github.repository }}\",
+                \"storage_image_tag\": \"${{ needs.tag.outputs.build-tag }}\",
+                \"compute_image_tag\": \"${{ needs.tag.outputs.build-tag }}\"
               }
             }"
 
   neon-image:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
+    needs: [ tag ]
     container: gcr.io/kaniko-project/executor:v1.9.0-debug
 
     steps:
@@ -494,10 +475,11 @@ jobs:
         run: echo "{\"credsStore\":\"ecr-login\"}" > /kaniko/.docker/config.json
 
       - name: Kaniko build neon
-        run: /kaniko/executor --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --snapshotMode=redo --context . --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:$GITHUB_RUN_ID
+        run: /kaniko/executor --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --snapshotMode=redo --context . --build-arg GIT_VERSION=${{ github.sha }} --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}}
 
   compute-tools-image:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
+    needs: [ tag ]
     container: gcr.io/kaniko-project/executor:v1.9.0-debug
 
     steps:
@@ -508,30 +490,12 @@ jobs:
         run: echo "{\"credsStore\":\"ecr-login\"}" > /kaniko/.docker/config.json
 
       - name: Kaniko build compute tools
-        run: /kaniko/executor --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --snapshotMode=redo --context . --dockerfile Dockerfile.compute-tools --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:$GITHUB_RUN_ID
-
-  compute-node-image:
-    runs-on: dev
-    container: gcr.io/kaniko-project/executor:v1.9.0-debug
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v1 # v3 won't work with kaniko
-        with:
-          submodules: true
-          fetch-depth: 0
-
-      - name: Configure ECR login
-        run: echo "{\"credsStore\":\"ecr-login\"}" > /kaniko/.docker/config.json
-
-        # compute-node uses postgres 14, which is default now
-        # cloud repo depends on this image name, thus duplicating it
-        # remove compute-node when cloud repo is updated
-      - name: Kaniko build compute node with extensions v14 (compatibility)
-        run: /kaniko/executor --skip-unused-stages --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --snapshotMode=redo --context . --dockerfile Dockerfile.compute-node-v14 --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node:$GITHUB_RUN_ID
+        run: /kaniko/executor --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --snapshotMode=redo --context . --build-arg GIT_VERSION=${{ github.sha }} --dockerfile Dockerfile.compute-tools --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{needs.tag.outputs.build-tag}}
 
   compute-node-image-v14:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container: gcr.io/kaniko-project/executor:v1.9.0-debug
+    needs: [ tag ]
     steps:
       - name: Checkout
         uses: actions/checkout@v1 # v3 won't work with kaniko
@@ -543,12 +507,12 @@ jobs:
         run: echo "{\"credsStore\":\"ecr-login\"}" > /kaniko/.docker/config.json
 
       - name: Kaniko build compute node with extensions v14
-        run: /kaniko/executor --skip-unused-stages  --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache  --context . --dockerfile Dockerfile.compute-node-v14 --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:$GITHUB_RUN_ID
-
+        run: /kaniko/executor --skip-unused-stages  --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache  --context . --build-arg GIT_VERSION=${{ github.sha }} --dockerfile Dockerfile.compute-node-v14 --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:${{needs.tag.outputs.build-tag}}
 
   compute-node-image-v15:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container: gcr.io/kaniko-project/executor:v1.9.0-debug
+    needs: [ tag ]
     steps:
       - name: Checkout
         uses: actions/checkout@v1 # v3 won't work with kaniko
@@ -560,28 +524,69 @@ jobs:
         run: echo "{\"credsStore\":\"ecr-login\"}" > /kaniko/.docker/config.json
 
       - name: Kaniko build compute node with extensions v15
-        run: /kaniko/executor --skip-unused-stages --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --context . --dockerfile Dockerfile.compute-node-v15 --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:$GITHUB_RUN_ID
+        run: /kaniko/executor --skip-unused-stages --snapshotMode=redo --cache=true --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache --context . --build-arg GIT_VERSION=${{ github.sha }} --dockerfile Dockerfile.compute-node-v15 --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:${{needs.tag.outputs.build-tag}}
+
+  test-images:
+    needs: [ tag, neon-image, compute-node-image-v14, compute-node-image-v15, compute-tools-image ]
+    runs-on: [ self-hosted, dev, x64 ]
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0
+
+      # `neondatabase/neon` contains multiple binaries, all of them use the same input for the version into the same version formatting library.
+      # Pick pageserver as currently the only binary with extra "version" features printed in the string to verify.
+      # Regular pageserver version string looks like
+      #   Neon page server git-env:32d14403bd6ab4f4520a94cbfd81a6acef7a526c failpoints: true, features: []
+      # Bad versions might loop like:
+      #   Neon page server git-env:local failpoints: true, features: ["testing"]
+      # Ensure that we don't have bad versions.
+      - name: Verify image versions
+        shell: bash # ensure no set -e for better error messages
+        run: |
+          pageserver_version=$(docker run --rm 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}} "/bin/sh" "-c" "/usr/local/bin/pageserver --version")
+
+          echo "Pageserver version string: $pageserver_version"
+
+          if ! echo "$pageserver_version" | grep -qv 'git-env:local' ; then
+            echo "Pageserver version should not be the default Dockerfile one"
+            exit 1
+          fi
+
+          if ! echo "$pageserver_version" | grep -qv '"testing"' ; then
+            echo "Pageserver version should have no testing feature enabled"
+            exit 1
+          fi
+
+      - name: Verify docker-compose example
+        run: env REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com TAG=${{needs.tag.outputs.build-tag}} ./docker-compose/docker_compose_test.sh
+
+      - name: Print logs and clean up
+        if: always()
+        run: |
+          docker compose -f ./docker-compose/docker-compose.yml logs || 0
+          docker compose -f ./docker-compose/docker-compose.yml down
 
   promote-images:
-    runs-on: dev
-    needs: [ neon-image, compute-node-image, compute-node-image-v14, compute-tools-image ]
+    runs-on: [ self-hosted, dev, x64 ]
+    needs: [ tag, test-images ]
     if: github.event_name != 'workflow_dispatch'
     container: amazon/aws-cli
     strategy:
       fail-fast: false
       matrix:
-        # compute-node uses postgres 14, which is default now
-        # cloud repo depends on this image name, thus duplicating it
-        # remove compute-node when cloud repo is updated
-        name: [ neon, compute-node, compute-node-v14, compute-tools ]
+        name: [ neon, compute-node-v14, compute-node-v15, compute-tools ]
 
     steps:
       - name: Promote image to latest
-        run:
-          MANIFEST=$(aws ecr batch-get-image --repository-name ${{ matrix.name }} --image-ids imageTag=$GITHUB_RUN_ID --query 'images[].imageManifest' --output text) && aws ecr put-image --repository-name ${{ matrix.name }} --image-tag latest --image-manifest "$MANIFEST"
+        run: |
+          export MANIFEST=$(aws ecr batch-get-image --repository-name ${{ matrix.name }} --image-ids imageTag=${{needs.tag.outputs.build-tag}} --query 'images[].imageManifest' --output text)
+          aws ecr put-image --repository-name ${{ matrix.name }} --image-tag latest --image-manifest "$MANIFEST"
 
   push-docker-hub:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     needs: [ promote-images, tag ]
     container: golang:1.19-bullseye
 
@@ -597,16 +602,16 @@ jobs:
           echo "{\"credsStore\":\"ecr-login\"}" > /github/home/.docker/config.json
 
       - name: Pull neon image from ECR
-        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:latest neon
+        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}} neon
 
       - name: Pull compute tools image from ECR
-        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:latest compute-tools
-
-      - name: Pull compute node image from ECR
-        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node:latest compute-node
+        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{needs.tag.outputs.build-tag}} compute-tools
 
       - name: Pull compute node v14 image from ECR
-        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:latest compute-node-v14
+        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:${{needs.tag.outputs.build-tag}} compute-node-v14
+
+      - name: Pull compute node v15 image from ECR
+        run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:${{needs.tag.outputs.build-tag}} compute-node-v15
 
       - name: Pull rust image from ECR
         run: crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned rust
@@ -616,9 +621,10 @@ jobs:
           (github.ref_name == 'main' || github.ref_name == 'release') &&
           github.event_name != 'workflow_dispatch'
         run: |
-          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:$GITHUB_RUN_ID 093970136003.dkr.ecr.us-east-2.amazonaws.com/neon:latest
-          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:$GITHUB_RUN_ID 093970136003.dkr.ecr.us-east-2.amazonaws.com/compute-tools:latest
-          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node:$GITHUB_RUN_ID 093970136003.dkr.ecr.us-east-2.amazonaws.com/compute-node:latest
+          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/neon:latest
+          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:latest
+          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:latest
+          crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:latest
 
       - name: Configure Docker Hub login
         run: |
@@ -632,12 +638,12 @@ jobs:
       - name: Push compute tools image to Docker Hub
         run: crane push compute-tools neondatabase/compute-tools:${{needs.tag.outputs.build-tag}}
 
-      - name: Push compute node image to Docker Hub
-        run: crane push compute-node neondatabase/compute-node:${{needs.tag.outputs.build-tag}}
-
       - name: Push compute node v14 image to Docker Hub
         run: crane push compute-node-v14 neondatabase/compute-node-v14:${{needs.tag.outputs.build-tag}}
 
+      - name: Push compute node v15 image to Docker Hub
+        run: crane push compute-node-v15 neondatabase/compute-node-v15:${{needs.tag.outputs.build-tag}}
+
       - name: Push rust image to Docker Hub
         run: crane push rust neondatabase/rust:pinned
 
@@ -648,8 +654,8 @@ jobs:
         run: |
           crane tag neondatabase/neon:${{needs.tag.outputs.build-tag}} latest
           crane tag neondatabase/compute-tools:${{needs.tag.outputs.build-tag}} latest
-          crane tag neondatabase/compute-node:${{needs.tag.outputs.build-tag}} latest
           crane tag neondatabase/compute-node-v14:${{needs.tag.outputs.build-tag}} latest
+          crane tag neondatabase/compute-node-v15:${{needs.tag.outputs.build-tag}} latest
 
   calculate-deploy-targets:
     runs-on: [ self-hosted, Linux, k8s-runner ]
@@ -662,12 +668,12 @@ jobs:
       - id: set-matrix
         run: |
           if [[ "$GITHUB_REF_NAME" == "main" ]]; then
-            STAGING='{"env_name": "staging", "proxy_job": "neon-proxy", "proxy_config": "staging.proxy", "kubeconfig_secret": "STAGING_KUBECONFIG_DATA"}'
-            NEON_STRESS='{"env_name": "neon-stress", "proxy_job": "neon-stress-proxy", "proxy_config": "neon-stress.proxy", "kubeconfig_secret": "NEON_STRESS_KUBECONFIG_DATA"}'
-            echo "::set-output name=include::[$STAGING, $NEON_STRESS]"
+            STAGING='{"env_name": "staging", "proxy_job": "neon-proxy", "proxy_config": "staging.proxy", "kubeconfig_secret": "STAGING_KUBECONFIG_DATA", "console_api_key_secret": "NEON_STAGING_API_KEY"}'
+            NEON_STRESS='{"env_name": "neon-stress", "proxy_job": "neon-stress-proxy", "proxy_config": "neon-stress.proxy", "kubeconfig_secret": "NEON_STRESS_KUBECONFIG_DATA", "console_api_key_secret": "NEON_CAPTEST_API_KEY"}'
+            echo "include=[$STAGING, $NEON_STRESS]" >> $GITHUB_OUTPUT
           elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
-            PRODUCTION='{"env_name": "production", "proxy_job": "neon-proxy", "proxy_config": "production.proxy", "kubeconfig_secret": "PRODUCTION_KUBECONFIG_DATA"}'
-            echo "::set-output name=include::[$PRODUCTION]"
+            PRODUCTION='{"env_name": "production", "proxy_job": "neon-proxy", "proxy_config": "production.proxy", "kubeconfig_secret": "PRODUCTION_KUBECONFIG_DATA", "console_api_key_secret": "NEON_PRODUCTION_API_KEY"}'
+            echo "include=[$PRODUCTION]" >> $GITHUB_OUTPUT
           else
             echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
             exit 1
@@ -703,7 +709,7 @@ jobs:
       - name: Setup ansible
         run: |
           export PATH="/root/.local/bin:$PATH"
-          pip install --progress-bar off --user ansible boto3
+          pip install --progress-bar off --user ansible boto3 toml
 
       - name: Redeploy
         run: |
@@ -725,12 +731,123 @@ jobs:
           chmod 0600 ssh-key
           ssh-add ssh-key
           rm -f ssh-key ssh-key-cert.pub
+          ansible-galaxy collection install sivel.toiletwater
+          ansible-playbook deploy.yaml -i ${{ matrix.env_name }}.hosts.yaml -e CONSOLE_API_TOKEN=${{ secrets[matrix.console_api_key_secret] }}
+          rm -f neon_install.tar.gz .neon_current_version
 
-          ansible-playbook deploy.yaml -i ${{ matrix.env_name }}.hosts
+  deploy-new:
+    runs-on: [ self-hosted, dev, x64 ]
+    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
+    # We need both storage **and** compute images for deploy, because control plane picks the compute version based on the storage version.
+    # If it notices a fresh storage it may bump the compute version. And if compute image failed to build it may break things badly
+    needs: [ push-docker-hub, calculate-deploy-targets, tag, regress-tests ]
+    if: |
+      (github.ref_name == 'main') &&
+      github.event_name != 'workflow_dispatch'
+    defaults:
+      run:
+        shell: bash
+    strategy:
+      matrix:
+        target_region: [ us-east-2 ]
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          submodules: true
+          fetch-depth: 0
+
+      - name: Redeploy
+        run: |
+          export DOCKER_TAG=${{needs.tag.outputs.build-tag}}
+          cd "$(pwd)/.github/ansible"
+          if [[ "$GITHUB_REF_NAME" == "main" ]]; then
+            ./get_binaries.sh
+          elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
+            RELEASE=true ./get_binaries.sh
+          else
+            echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
+            exit 1
+          fi
+          ansible-galaxy collection install sivel.toiletwater
+          ansible-playbook deploy.yaml -i staging.${{ matrix.target_region }}.hosts.yaml -e @ssm_config -e CONSOLE_API_TOKEN=${{secrets.NEON_STAGING_API_KEY}}
+          rm -f neon_install.tar.gz .neon_current_version
+
+  deploy-pr-test-new:
+    runs-on: [ self-hosted, dev, x64 ]
+    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
+    # We need both storage **and** compute images for deploy, because control plane picks the compute version based on the storage version.
+    # If it notices a fresh storage it may bump the compute version. And if compute image failed to build it may break things badly
+    needs: [ push-docker-hub, tag, regress-tests ]
+    if: |
+      contains(github.event.pull_request.labels.*.name, 'deploy-test-storage') && 
+      github.event_name != 'workflow_dispatch'
+    defaults:
+      run:
+        shell: bash
+    strategy:
+      matrix:
+        target_region: [ eu-west-1 ]
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          submodules: true
+          fetch-depth: 0
+
+      - name: Redeploy
+        run: |
+          export DOCKER_TAG=${{needs.tag.outputs.build-tag}}
+          cd "$(pwd)/.github/ansible"
+
+          ./get_binaries.sh
+
+          ansible-galaxy collection install sivel.toiletwater
+          ansible-playbook deploy.yaml -i staging.${{ matrix.target_region }}.hosts.yaml -e @ssm_config -e CONSOLE_API_TOKEN=${{secrets.NEON_STAGING_API_KEY}}
+          rm -f neon_install.tar.gz .neon_current_version
+
+  deploy-prod-new:
+    runs-on: prod
+    container: 093970136003.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
+    # We need both storage **and** compute images for deploy, because control plane picks the compute version based on the storage version.
+    # If it notices a fresh storage it may bump the compute version. And if compute image failed to build it may break things badly
+    needs: [ push-docker-hub, tag, regress-tests ]
+    if: |
+      (github.ref_name == 'release') &&
+      github.event_name != 'workflow_dispatch'
+    defaults:
+      run:
+        shell: bash
+    strategy:
+      matrix:
+        target_region: [ us-east-2, eu-central-1, ap-southeast-1 ]
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          submodules: true
+          fetch-depth: 0
+
+      - name: Redeploy
+        run: |
+          export DOCKER_TAG=${{needs.tag.outputs.build-tag}}
+          cd "$(pwd)/.github/ansible"
+
+          if [[ "$GITHUB_REF_NAME" == "main" ]]; then
+            ./get_binaries.sh
+          elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
+            RELEASE=true ./get_binaries.sh
+          else
+            echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
+            exit 1
+          fi
+
+          ansible-galaxy collection install sivel.toiletwater
+          ansible-playbook deploy.yaml -i prod.${{ matrix.target_region }}.hosts.yaml -e @ssm_config -e CONSOLE_API_TOKEN=${{secrets.NEON_PRODUCTION_API_KEY}}
           rm -f neon_install.tar.gz .neon_current_version
 
   deploy-proxy:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/base:latest
     # Compute image isn't strictly required for proxy deploy, but let's still wait for it to run all deploy jobs consistently.
     needs: [ push-docker-hub, calculate-deploy-targets, tag, regress-tests ]
@@ -770,3 +887,111 @@ jobs:
           DOCKER_TAG=${{needs.tag.outputs.build-tag}}
           helm upgrade ${{ matrix.proxy_job }}       neondatabase/neon-proxy --namespace neon-proxy --install -f .github/helm-values/${{ matrix.proxy_config }}.yaml --set image.tag=${DOCKER_TAG} --wait --timeout 15m0s
           helm upgrade ${{ matrix.proxy_job }}-scram neondatabase/neon-proxy --namespace neon-proxy --install -f .github/helm-values/${{ matrix.proxy_config }}-scram.yaml --set image.tag=${DOCKER_TAG} --wait --timeout 15m0s
+
+  deploy-proxy-new:
+    runs-on: [ self-hosted, dev, x64 ]
+    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
+    # Compute image isn't strictly required for proxy deploy, but let's still wait for it to run all deploy jobs consistently.
+    needs: [ push-docker-hub, tag, regress-tests ]
+    if: |
+      (github.ref_name == 'main') &&
+      github.event_name != 'workflow_dispatch'
+    defaults:
+      run:
+        shell: bash
+    strategy:
+      matrix:
+        include:
+          - target_region:  us-east-2
+            target_cluster: dev-us-east-2-beta
+          - target_region:  eu-west-1
+            target_cluster: dev-eu-west-1-zeta
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          submodules: true
+          fetch-depth: 0
+
+      - name: Configure environment
+        run: |
+          helm repo add neondatabase https://neondatabase.github.io/helm-charts
+          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
+
+      - name: Re-deploy proxy
+        run: |
+          DOCKER_TAG=${{needs.tag.outputs.build-tag}}
+          helm upgrade neon-proxy-scram neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-scram.yaml --set image.tag=${DOCKER_TAG} --wait --timeout 15m0s
+
+  deploy-proxy-prod-new:
+    runs-on: prod
+    container: 093970136003.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
+    # Compute image isn't strictly required for proxy deploy, but let's still wait for it to run all deploy jobs consistently.
+    needs: [ push-docker-hub, tag, regress-tests ]
+    if: |
+      (github.ref_name == 'release') &&
+      github.event_name != 'workflow_dispatch'
+    defaults:
+      run:
+        shell: bash
+    strategy:
+      matrix:
+        include:
+          - target_region:  us-east-2
+            target_cluster: prod-us-east-2-delta
+          - target_region: eu-central-1
+            target_cluster: prod-eu-central-1-gamma
+          - target_region: ap-southeast-1
+            target_cluster: prod-ap-southeast-1-epsilon
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+        with:
+          submodules: true
+          fetch-depth: 0
+
+      - name: Configure environment
+        run: |
+          helm repo add neondatabase https://neondatabase.github.io/helm-charts
+          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
+
+      - name: Re-deploy proxy
+        run: |
+          DOCKER_TAG=${{needs.tag.outputs.build-tag}}
+          helm upgrade neon-proxy-scram neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-scram.yaml --set image.tag=${DOCKER_TAG} --wait --timeout 15m0s
+
+  promote-compatibility-data:
+    runs-on: [ self-hosted, dev, x64 ]
+    container:
+      image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
+      options: --init
+    needs: [ deploy, deploy-proxy ]
+    if: github.ref_name == 'release' && github.event_name != 'workflow_dispatch'
+    steps:
+      - name: Promote compatibility snapshot for the release
+        shell: bash -euxo pipefail {0}
+        env:
+          BUCKET: neon-github-public-dev
+          PREFIX: artifacts/latest
+        run: |
+          # Update compatibility snapshot for the release
+          for build_type in debug release; do
+            OLD_FILENAME=compatibility-snapshot-${build_type}-pg14-${GITHUB_RUN_ID}.tar.zst
+            NEW_FILENAME=compatibility-snapshot-${build_type}-pg14.tar.zst
+
+            time aws s3 mv --only-show-errors s3://${BUCKET}/${PREFIX}/${OLD_FILENAME} s3://${BUCKET}/${PREFIX}/${NEW_FILENAME}
+          done
+
+          # Update Neon artifact for the release (reuse already uploaded artifact)
+          for build_type in debug release; do
+            OLD_PREFIX=artifacts/${GITHUB_RUN_ID}
+            FILENAME=neon-${{ runner.os }}-${build_type}-artifact.tar.zst
+
+            S3_KEY=$(aws s3api list-objects-v2 --bucket ${BUCKET} --prefix ${OLD_PREFIX} | jq -r '.Contents[].Key' | grep ${FILENAME} | sort --version-sort | tail -1 || true)
+            if [ -z "${S3_KEY}" ]; then
+              echo 2>&1 "Neither s3://${BUCKET}/${OLD_PREFIX}/${FILENAME} nor its version from previous attempts exist"
+              exit 1
+            fi
+
+            time aws s3 cp --only-show-errors s3://${BUCKET}/${S3_KEY} s3://${BUCKET}/${PREFIX}/${FILENAME}
+          done

.github/workflows/codestyle.yml

@@ -36,7 +36,7 @@ jobs:
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           submodules: true
           fetch-depth: 2
@@ -56,12 +56,12 @@ jobs:
 
       - name: Set pg 14 revision for caching
         id: pg_v14_rev
-        run: echo ::set-output name=pg_rev::$(git rev-parse HEAD:vendor/postgres-v14)
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v14) >> $GITHUB_OUTPUT
         shell: bash -euxo pipefail {0}
 
       - name: Set pg 15 revision for caching
         id: pg_v15_rev
-        run: echo ::set-output name=pg_rev::$(git rev-parse HEAD:vendor/postgres-v15)
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v15) >> $GITHUB_OUTPUT
         shell: bash -euxo pipefail {0}
 
       - name: Cache postgres v14 build
@@ -106,7 +106,7 @@ jobs:
             !~/.cargo/registry/src
             ~/.cargo/git
             target
-          key: v5-${{ runner.os }}-cargo-${{ hashFiles('./Cargo.lock') }}-rust
+          key: v6-${{ runner.os }}-cargo-${{ hashFiles('./Cargo.lock') }}-rust
 
       - name: Run cargo clippy
         run: ./run_clippy.sh
@@ -115,7 +115,7 @@ jobs:
         run: cargo build --locked --all --all-targets
 
   check-rust-dependencies:
-    runs-on: dev
+    runs-on: [ self-hosted, dev, x64 ]
     container:
       image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:pinned
       options: --init

.gitmodules

@@ -1,7 +1,7 @@
 [submodule "vendor/postgres-v14"]
 	path = vendor/postgres-v14
 	url = https://github.com/neondatabase/postgres.git
-	branch = main
+	branch = REL_14_STABLE_neon
 [submodule "vendor/postgres-v15"]
 	path = vendor/postgres-v15
 	url = https://github.com/neondatabase/postgres.git

CODEOWNERS

@@ -0,0 +1,11 @@
+/compute_tools/ @neondatabase/control-plane
+/control_plane/ @neondatabase/compute @neondatabase/storage
+/libs/pageserver_api/ @neondatabase/compute @neondatabase/storage
+/libs/postgres_ffi/ @neondatabase/compute 
+/libs/remote_storage/ @neondatabase/storage 
+/libs/safekeeper_api/ @neondatabase/safekeepers  
+/pageserver/ @neondatabase/compute @neondatabase/storage 
+/pgxn/ @neondatabase/compute
+/proxy/ @neondatabase/control-plane 
+/safekeeper/ @neondatabase/safekeepers
+/vendor/ @neondatabase/compute

Cargo.toml

@@ -1,3 +1,14 @@
+# 'named-profiles' feature was stabilized in cargo 1.57. This line makes the
+# build work with older cargo versions.
+#
+# We have this because as of this writing, the latest cargo Debian package
+# that's available is 1.56. (Confusingly, the Debian package version number
+# is 0.57, whereas 'cargo --version' says 1.56.)
+#
+# See https://tracker.debian.org/pkg/cargo for the current status of the
+# package. When that gets updated, we can remove this.
+cargo-features = ["named-profiles"]
+
 [workspace]
 members = [
     "compute_tools",
@@ -14,6 +25,10 @@ members = [
 # Besides, debug info should not affect the performance.
 debug = true
 
+# disable debug symbols for all packages except this one to decrease binaries size
+[profile.release.package."*"]
+debug = false
+
 [profile.release-line-debug]
 inherits = "release"
 debug = 1 # true = 2 = all symbols, 1 = line only

Dockerfile

@@ -44,7 +44,7 @@ COPY . .
 # Show build caching stats to check if it was used in the end.
 # Has to be the part of the same RUN since cachepot daemon is killed in the end of this RUN, losing the compilation stats.
 RUN set -e \
-&& mold -run cargo build --bin pageserver --bin safekeeper --bin proxy --locked --release \
+&& mold -run cargo build --bin pageserver --bin pageserver_binutils --bin draw_timeline_dir --bin safekeeper --bin proxy --locked --release \
     && cachepot -s
 
 # Build final image
@@ -63,9 +63,11 @@ RUN set -e \
     && useradd -d /data neon \
     && chown -R neon:neon /data
 
-COPY --from=build --chown=neon:neon /home/nonroot/target/release/pageserver /usr/local/bin
-COPY --from=build --chown=neon:neon /home/nonroot/target/release/safekeeper /usr/local/bin
-COPY --from=build --chown=neon:neon /home/nonroot/target/release/proxy      /usr/local/bin
+COPY --from=build --chown=neon:neon /home/nonroot/target/release/pageserver          /usr/local/bin
+COPY --from=build --chown=neon:neon /home/nonroot/target/release/pageserver_binutils /usr/local/bin
+COPY --from=build --chown=neon:neon /home/nonroot/target/release/draw_timeline_dir   /usr/local/bin
+COPY --from=build --chown=neon:neon /home/nonroot/target/release/safekeeper          /usr/local/bin
+COPY --from=build --chown=neon:neon /home/nonroot/target/release/proxy               /usr/local/bin
 
 COPY --from=pg-build /home/nonroot/pg_install/v14 /usr/local/v14/
 COPY --from=pg-build /home/nonroot/pg_install/v15 /usr/local/v15/
@@ -85,4 +87,3 @@ VOLUME ["/data"]
 USER neon
 EXPOSE 6400
 EXPOSE 9898
-CMD ["/bin/bash"]

Dockerfile.compute-node-v14

@@ -1,50 +1,50 @@
-ARG TAG=pinned
-# apparently, ARGs don't get replaced in RUN commands in kaniko
-# ARG POSTGIS_VERSION=3.3.0
-# ARG PLV8_VERSION=3.1.4
-# ARG PG_VERSION=v14
+#
+# This file is identical to the Dockerfile.compute-node-v15 file
+# except for the version of Postgres that is built.
+#
 
+ARG TAG=pinned
+
+#########################################################################################
 #
 # Layer "build-deps"
 #
+#########################################################################################
 FROM debian:bullseye-slim AS build-deps
-RUN echo "deb http://ftp.debian.org/debian testing main" >> /etc/apt/sources.list && \
-    echo "APT::Default-Release \"stable\";" > /etc/apt/apt.conf.d/default-release && \
-    apt update
 RUN apt update &&  \
-    apt install -y git autoconf automake libtool build-essential bison flex libreadline-dev zlib1g-dev libxml2-dev \
-    libcurl4-openssl-dev libossp-uuid-dev wget pkg-config libglib2.0-dev
+    apt install -y git autoconf automake libtool build-essential bison flex libreadline-dev \
+    zlib1g-dev libxml2-dev libcurl4-openssl-dev libossp-uuid-dev wget pkg-config libssl-dev
 
+#########################################################################################
 #
 # Layer "pg-build"
 # Build Postgres from the neon postgres repository.
 #
+#########################################################################################
 FROM build-deps AS pg-build
 COPY vendor/postgres-v14 postgres
 RUN cd postgres && \
-    ./configure CFLAGS='-O2 -g3' --enable-debug --with-uuid=ossp && \
+    ./configure CFLAGS='-O2 -g3' --enable-debug --with-openssl --with-uuid=ossp && \
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s install && \
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C contrib/ install && \
     # Install headers
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C src/include install && \
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C src/interfaces/libpq install
 
+#########################################################################################
 #
 # Layer "postgis-build"
 # Build PostGIS from the upstream PostGIS mirror.
 #
-# PostGIS compiles against neon postgres sources without changes. Perhaps we
-# could even use the upstream binaries, compiled against vanilla Postgres, but
-# it would require some investigation to check that it works, and also keeps
-# working in the future. So for now, we compile our own binaries.
+#########################################################################################
 FROM build-deps AS postgis-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 RUN apt update && \
     apt install -y gdal-bin libgdal-dev libprotobuf-c-dev protobuf-c-compiler xsltproc
 
-RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.0.tar.gz && \
-    tar xvzf postgis-3.3.0.tar.gz && \
-    cd postgis-3.3.0 && \
+RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.1.tar.gz && \
+    tar xvzf postgis-3.3.1.tar.gz && \
+    cd postgis-3.3.1 && \
     ./autogen.sh && \
     export PATH="/usr/local/pgsql/bin:$PATH" && \
     ./configure && \
@@ -57,39 +57,55 @@ RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.0.tar.gz && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/postgis_tiger_geocoder.control && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/postgis_topology.control
 
+#########################################################################################
 #
 # Layer "plv8-build"
 # Build plv8
 #
+#########################################################################################
 FROM build-deps AS plv8-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 RUN apt update && \
-    apt install -y ninja-build python3-dev libc++-dev libc++abi-dev libncurses5
+    apt install -y ninja-build python3-dev libc++-dev libc++abi-dev libncurses5 binutils
 
-# https://github.com/plv8/plv8/issues/475
-# Debian bullseye provides binutils 2.35 when >= 2.38 is necessary
-RUN apt update && \
-    apt install -y --no-install-recommends -t testing binutils
+# https://github.com/plv8/plv8/issues/475:
+#   v8 uses gold for linking and sets `--thread-count=4` which breaks
+#   gold version <= 1.35 (https://sourceware.org/bugzilla/show_bug.cgi?id=23607)
+# Install newer gold version manually as debian-testing binutils version updates
+# libc version, which in turn breaks other extension built against non-testing libc.
+RUN wget https://ftp.gnu.org/gnu/binutils/binutils-2.38.tar.gz && \
+    tar xvzf binutils-2.38.tar.gz && \
+    cd binutils-2.38 && \
+    cd libiberty && ./configure && make -j $(getconf _NPROCESSORS_ONLN) && \
+    cd ../bfd && ./configure && make bfdver.h && \
+    cd ../gold && ./configure && make -j $(getconf _NPROCESSORS_ONLN) && make install && \
+    cp /usr/local/bin/ld.gold /usr/bin/gold
 
+# Sed is used to patch for https://github.com/plv8/plv8/issues/503
 RUN wget https://github.com/plv8/plv8/archive/refs/tags/v3.1.4.tar.gz && \
     tar xvzf v3.1.4.tar.gz && \
     cd plv8-3.1.4 && \
     export PATH="/usr/local/pgsql/bin:$PATH" && \
-    make -j $(getconf _NPROCESSORS_ONLN) && \
-    make -j $(getconf _NPROCESSORS_ONLN) install && \
+    sed -i 's/MemoryContextAlloc(/MemoryContextAllocZero(/' plv8.cc && \
+    make DOCKER=1 -j $(getconf _NPROCESSORS_ONLN) install && \
     rm -rf /plv8-* && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/plv8.control
 
+#########################################################################################
 #
 # Layer "h3-pg-build"
 # Build h3_pg
 #
+#########################################################################################
 FROM build-deps AS h3-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
 # packaged cmake is too old
-RUN apt update && \
-    apt install -y --no-install-recommends -t testing cmake
+RUN wget https://github.com/Kitware/CMake/releases/download/v3.24.2/cmake-3.24.2-linux-x86_64.sh \
+      -q -O /tmp/cmake-install.sh \
+      && chmod u+x /tmp/cmake-install.sh \
+      && /tmp/cmake-install.sh --skip-license --prefix=/usr/local/ \
+      && rm /tmp/cmake-install.sh
 
 RUN wget https://github.com/uber/h3/archive/refs/tags/v4.0.1.tar.gz -O h3.tgz && \
     tar xvzf h3.tgz  && \
@@ -108,16 +124,18 @@ RUN wget https://github.com/zachasme/h3-pg/archive/refs/tags/v4.0.1.tar.gz -O h3
     export PATH="/usr/local/pgsql/bin:$PATH" && \
     make -j $(getconf _NPROCESSORS_ONLN) && \
     make -j $(getconf _NPROCESSORS_ONLN) install && \
-    echo 'trusted = true' >> /usr/local/pgsql/share/extension/h3.control
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/h3.control && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/h3_postgis.control
 
+#########################################################################################
 #
 # Layer "neon-pg-ext-build"
 # compile neon extensions
 #
+#########################################################################################
 FROM build-deps AS neon-pg-ext-build
 COPY --from=postgis-build /usr/local/pgsql/ /usr/local/pgsql/
-# plv8 still sometimes crashes during the creation
-# COPY --from=plv8-build /usr/local/pgsql/ /usr/local/pgsql/
+COPY --from=plv8-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=h3-pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=h3-pg-build /h3/usr /
 COPY pgxn/ pgxn/
@@ -127,16 +145,22 @@ RUN make -j $(getconf _NPROCESSORS_ONLN) \
         -C pgxn/neon \
         -s install
 
+#########################################################################################
+#
 # Compile and run the Neon-specific `compute_ctl` binary
+#
+#########################################################################################
 FROM 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:$TAG AS compute-tools
 USER nonroot
 # Copy entire project to get Cargo.* files with proper dependencies for the whole project
 COPY --chown=nonroot . .
 RUN cd compute_tools && cargo build --locked --profile release-line-debug-size-lto
 
+#########################################################################################
 #
 # Clean up postgres folder before inclusion
 #
+#########################################################################################
 FROM neon-pg-ext-build AS postgres-cleanup-layer
 COPY --from=neon-pg-ext-build /usr/local/pgsql /usr/local/pgsql
 
@@ -154,10 +178,12 @@ RUN rm -r /usr/local/pgsql/lib/pgxs/src
 # if they were to be used by other libraries.
 RUN rm /usr/local/pgsql/lib/lib*.a
 
+#########################################################################################
 #
 # Final layer
 # Put it all together into the final image
 #
+#########################################################################################
 FROM debian:bullseye-slim
 # Add user postgres
 RUN mkdir /var/db && useradd -m -d /var/db/postgres postgres && \
@@ -174,8 +200,6 @@ COPY --from=compute-tools --chown=postgres /home/nonroot/target/release-line-deb
 # libreadline8 for psql
 # libossp-uuid16 for extension ossp-uuid
 # libgeos, libgdal, libproj and libprotobuf-c1 for PostGIS
-# GLIBC 2.34 for plv8.
-#     Debian bullseye provides GLIBC 2.31, so we install the library from testing
 #
 # Lastly, link compute_ctl into zenith_ctl while we're at it,
 # so that we don't need to put this in another layer.
@@ -188,12 +212,6 @@ RUN apt update &&  \
         libproj19 \
         libprotobuf-c1 && \
     rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
-    echo "Installing GLIBC 2.34" && \
-    echo "deb http://ftp.debian.org/debian testing main" >> /etc/apt/sources.list && \
-    echo "APT::Default-Release \"stable\";" > /etc/apt/apt.conf.d/default-release && \
-    apt update && \
-    apt install -y --no-install-recommends -t testing libc6 && \
-    rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
     ln /usr/local/bin/compute_ctl /usr/local/bin/zenith_ctl
 
 USER postgres

Dockerfile.compute-node-v15

@@ -4,49 +4,47 @@
 #
 
 ARG TAG=pinned
-# apparently, ARGs don't get replaced in RUN commands in kaniko
-# ARG POSTGIS_VERSION=3.3.0
-# ARG PLV8_VERSION=3.1.4
-# ARG PG_VERSION=v15
 
+#########################################################################################
 #
 # Layer "build-deps"
 #
+#########################################################################################
 FROM debian:bullseye-slim AS build-deps
 RUN apt update &&  \
-    apt install -y git autoconf automake libtool build-essential bison flex libreadline-dev zlib1g-dev libxml2-dev \
-    libcurl4-openssl-dev libossp-uuid-dev
+    apt install -y git autoconf automake libtool build-essential bison flex libreadline-dev \
+    zlib1g-dev libxml2-dev libcurl4-openssl-dev libossp-uuid-dev wget pkg-config libssl-dev
 
+#########################################################################################
 #
 # Layer "pg-build"
 # Build Postgres from the neon postgres repository.
 #
+#########################################################################################
 FROM build-deps AS pg-build
 COPY vendor/postgres-v15 postgres
 RUN cd postgres && \
-    ./configure CFLAGS='-O2 -g3' --enable-debug --with-uuid=ossp && \
+    ./configure CFLAGS='-O2 -g3' --enable-debug --with-openssl --with-uuid=ossp && \
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s install && \
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C contrib/ install && \
     # Install headers
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C src/include install && \
     make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C src/interfaces/libpq install
 
+#########################################################################################
 #
 # Layer "postgis-build"
 # Build PostGIS from the upstream PostGIS mirror.
 #
-# PostGIS compiles against neon postgres sources without changes. Perhaps we
-# could even use the upstream binaries, compiled against vanilla Postgres, but
-# it would require some investigation to check that it works, and also keeps
-# working in the future. So for now, we compile our own binaries.
+#########################################################################################
 FROM build-deps AS postgis-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 RUN apt update && \
-    apt install -y gdal-bin libgdal-dev libprotobuf-c-dev protobuf-c-compiler xsltproc wget
+    apt install -y gdal-bin libgdal-dev libprotobuf-c-dev protobuf-c-compiler xsltproc
 
-RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.0.tar.gz && \
-    tar xvzf postgis-3.3.0.tar.gz && \
-    cd postgis-3.3.0 && \
+RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.1.tar.gz && \
+    tar xvzf postgis-3.3.1.tar.gz && \
+    cd postgis-3.3.1 && \
     ./autogen.sh && \
     export PATH="/usr/local/pgsql/bin:$PATH" && \
     ./configure && \
@@ -59,37 +57,87 @@ RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.0.tar.gz && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/postgis_tiger_geocoder.control && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/postgis_topology.control
 
+#########################################################################################
 #
 # Layer "plv8-build"
 # Build plv8
 #
+#########################################################################################
 FROM build-deps AS plv8-build
-COPY --from=postgis-build /usr/local/pgsql/ /usr/local/pgsql/
+COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 RUN apt update && \
-    apt install -y git curl wget make ninja-build build-essential libncurses5 python3-dev pkg-config libc++-dev libc++abi-dev libglib2.0-dev
+    apt install -y ninja-build python3-dev libc++-dev libc++abi-dev libncurses5 binutils
 
-# https://github.com/plv8/plv8/issues/475
-# Debian bullseye provides binutils 2.35 when >= 2.38 is necessary
-RUN echo "deb http://ftp.debian.org/debian testing main" >> /etc/apt/sources.list && \
-    echo "APT::Default-Release \"stable\";" > /etc/apt/apt.conf.d/default-release && \
-    apt update && \
-    apt install -y --no-install-recommends -t testing binutils
+# https://github.com/plv8/plv8/issues/475:
+#   v8 uses gold for linking and sets `--thread-count=4` which breaks
+#   gold version <= 1.35 (https://sourceware.org/bugzilla/show_bug.cgi?id=23607)
+# Install newer gold version manually as debian-testing binutils version updates
+# libc version, which in turn breaks other extension built against non-testing libc.
+RUN wget https://ftp.gnu.org/gnu/binutils/binutils-2.38.tar.gz && \
+    tar xvzf binutils-2.38.tar.gz && \
+    cd binutils-2.38 && \
+    cd libiberty && ./configure && make -j $(getconf _NPROCESSORS_ONLN) && \
+    cd ../bfd && ./configure && make bfdver.h && \
+    cd ../gold && ./configure && make -j $(getconf _NPROCESSORS_ONLN) && make install && \
+    cp /usr/local/bin/ld.gold /usr/bin/gold
 
+# Sed is used to patch for https://github.com/plv8/plv8/issues/503
 RUN wget https://github.com/plv8/plv8/archive/refs/tags/v3.1.4.tar.gz && \
     tar xvzf v3.1.4.tar.gz && \
     cd plv8-3.1.4 && \
     export PATH="/usr/local/pgsql/bin:$PATH" && \
-    make -j $(getconf _NPROCESSORS_ONLN) && \
-    make -j $(getconf _NPROCESSORS_ONLN) install && \
+    sed -i 's/MemoryContextAlloc(/MemoryContextAllocZero(/' plv8.cc && \
+    make DOCKER=1 -j $(getconf _NPROCESSORS_ONLN) install && \
     rm -rf /plv8-* && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/plv8.control
 
+#########################################################################################
+#
+# Layer "h3-pg-build"
+# Build h3_pg
+#
+#########################################################################################
+FROM build-deps AS h3-pg-build
+COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
+
+# packaged cmake is too old
+RUN wget https://github.com/Kitware/CMake/releases/download/v3.24.2/cmake-3.24.2-linux-x86_64.sh \
+      -q -O /tmp/cmake-install.sh \
+      && chmod u+x /tmp/cmake-install.sh \
+      && /tmp/cmake-install.sh --skip-license --prefix=/usr/local/ \
+      && rm /tmp/cmake-install.sh
+
+RUN wget https://github.com/uber/h3/archive/refs/tags/v4.0.1.tar.gz -O h3.tgz && \
+    tar xvzf h3.tgz  && \
+    cd h3-4.0.1 && \
+    mkdir build && \
+    cd build && \
+    cmake .. -DCMAKE_BUILD_TYPE=Release && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
+    DESTDIR=/h3 make install && \
+    cp -R /h3/usr / && \
+    rm -rf build
+
+RUN wget https://github.com/zachasme/h3-pg/archive/refs/tags/v4.0.1.tar.gz -O h3-pg.tgz && \
+    tar xvzf h3-pg.tgz && \
+    cd h3-pg-4.0.1 && \
+    export PATH="/usr/local/pgsql/bin:$PATH" && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
+    make -j $(getconf _NPROCESSORS_ONLN) install && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/h3.control && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/h3_postgis.control
+
+#########################################################################################
 #
 # Layer "neon-pg-ext-build"
 # compile neon extensions
 #
+#########################################################################################
 FROM build-deps AS neon-pg-ext-build
 COPY --from=postgis-build /usr/local/pgsql/ /usr/local/pgsql/
+COPY --from=plv8-build /usr/local/pgsql/ /usr/local/pgsql/
+COPY --from=h3-pg-build /usr/local/pgsql/ /usr/local/pgsql/
+COPY --from=h3-pg-build /h3/usr /
 COPY pgxn/ pgxn/
 
 RUN make -j $(getconf _NPROCESSORS_ONLN) \
@@ -97,16 +145,22 @@ RUN make -j $(getconf _NPROCESSORS_ONLN) \
         -C pgxn/neon \
         -s install
 
+#########################################################################################
+#
 # Compile and run the Neon-specific `compute_ctl` binary
+#
+#########################################################################################
 FROM 369495373322.dkr.ecr.eu-central-1.amazonaws.com/rust:$TAG AS compute-tools
 USER nonroot
 # Copy entire project to get Cargo.* files with proper dependencies for the whole project
 COPY --chown=nonroot . .
 RUN cd compute_tools && cargo build --locked --profile release-line-debug-size-lto
 
+#########################################################################################
 #
 # Clean up postgres folder before inclusion
 #
+#########################################################################################
 FROM neon-pg-ext-build AS postgres-cleanup-layer
 COPY --from=neon-pg-ext-build /usr/local/pgsql /usr/local/pgsql
 
@@ -124,10 +178,12 @@ RUN rm -r /usr/local/pgsql/lib/pgxs/src
 # if they were to be used by other libraries.
 RUN rm /usr/local/pgsql/lib/lib*.a
 
+#########################################################################################
 #
 # Final layer
 # Put it all together into the final image
 #
+#########################################################################################
 FROM debian:bullseye-slim
 # Add user postgres
 RUN mkdir /var/db && useradd -m -d /var/db/postgres postgres && \
@@ -137,8 +193,6 @@ RUN mkdir /var/db && useradd -m -d /var/db/postgres postgres && \
     chmod 0750 /var/db/postgres/compute && \
     echo '/usr/local/lib' >> /etc/ld.so.conf && /sbin/ldconfig
 
-# TODO: Check if we can make the extension setup more modular versus a linear build
-# currently plv8-build copies the output /usr/local/pgsql from postgis-build, etc#
 COPY --from=postgres-cleanup-layer --chown=postgres /usr/local/pgsql /usr/local
 COPY --from=compute-tools --chown=postgres /home/nonroot/target/release-line-debug-size-lto/compute_ctl /usr/local/bin/compute_ctl
 
@@ -146,8 +200,6 @@ COPY --from=compute-tools --chown=postgres /home/nonroot/target/release-line-deb
 # libreadline8 for psql
 # libossp-uuid16 for extension ossp-uuid
 # libgeos, libgdal, libproj and libprotobuf-c1 for PostGIS
-# GLIBC 2.34 for plv8.
-#     Debian bullseye provides GLIBC 2.31, so we install the library from testing
 #
 # Lastly, link compute_ctl into zenith_ctl while we're at it,
 # so that we don't need to put this in another layer.
@@ -160,12 +212,6 @@ RUN apt update &&  \
         libproj19 \
         libprotobuf-c1 && \
     rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
-    echo "Installing GLIBC 2.34" && \
-    echo "deb http://ftp.debian.org/debian testing main" >> /etc/apt/sources.list && \
-    echo "APT::Default-Release \"stable\";" > /etc/apt/apt.conf.d/default-release && \
-    apt update && \
-    apt install -y --no-install-recommends -t testing libc6 && \
-    rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
     ln /usr/local/bin/compute_ctl /usr/local/bin/zenith_ctl
 
 USER postgres

Dockerfile.compute-node.legacy

@@ -1,88 +0,0 @@
-#
-# Legacy version of the Dockerfile for the compute node.
-# Used by e2e CI. Building Dockerfile.compute-node will take
-# unreasonable ammount of time without v2 runners.
-#
-# TODO: remove once cloud repo CI is moved to v2 runners.
-#
-
-
-# Allow specifiyng different compute-tools tag and image repo, so we are
-# able to use different images
-ARG REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
-ARG IMAGE=compute-tools
-ARG TAG=latest
-
-#
-# Image with pre-built tools
-#
-FROM $REPOSITORY/$IMAGE:$TAG AS compute-deps
-# Only to get ready compute_ctl binary as deppendency
-
-#
-# Image with Postgres build deps
-#
-FROM debian:bullseye-slim AS build-deps
-
-RUN apt-get update && apt-get -yq install automake libtool build-essential bison flex libreadline-dev zlib1g-dev libxml2-dev \
-                                          libcurl4-openssl-dev libossp-uuid-dev
-
-#
-# Image with built Postgres
-#
-FROM build-deps AS pg-build
-
-# Add user postgres
-RUN adduser postgres
-RUN mkdir /pg && chown postgres:postgres /pg
-
-# Copy source files
-# version 14 is default for now
-COPY ./vendor/postgres-v14 /pg/
-COPY ./pgxn /pg/
-
-# Build and install Postgres locally
-RUN mkdir /pg/compute_build && cd /pg/compute_build && \
-    ../configure CFLAGS='-O2 -g3' --prefix=$(pwd)/postgres_bin --enable-debug --with-uuid=ossp && \
-    # Install main binaries and contribs
-    make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s install && \
-    make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C contrib/ install && \
-    # Install headers
-    make MAKELEVEL=0 -j $(getconf _NPROCESSORS_ONLN) -s -C src/include install
-
-# Install neon contrib
-RUN make MAKELEVEL=0 PG_CONFIG=/pg/compute_build/postgres_bin/bin/pg_config -j $(getconf _NPROCESSORS_ONLN) -C /pg/neon install
-
-USER postgres
-WORKDIR /pg
-
-#
-# Final compute node image to be exported
-#
-FROM debian:bullseye-slim
-
-# libreadline-dev is required to run psql
-RUN apt-get update && apt-get -yq install libreadline-dev libossp-uuid-dev
-
-# Add user postgres
-RUN mkdir /var/db && useradd -m -d /var/db/postgres postgres && \
-    echo "postgres:test_console_pass" | chpasswd && \
-    mkdir /var/db/postgres/compute && mkdir /var/db/postgres/specs && \
-    chown -R postgres:postgres /var/db/postgres && \
-    chmod 0750 /var/db/postgres/compute
-
-# Copy ready Postgres binaries
-COPY --from=pg-build /pg/compute_build/postgres_bin /usr/local
-
-# Copy binaries from compute-tools
-COPY --from=compute-deps /usr/local/bin/compute_ctl /usr/local/bin/compute_ctl
-
-# XXX: temporary symlink for compatibility with old control-plane
-RUN ln -s /usr/local/bin/compute_ctl /usr/local/bin/zenith_ctl
-
-# Add postgres shared objects to the search path
-RUN echo '/usr/local/lib' >> /etc/ld.so.conf && /sbin/ldconfig
-
-USER postgres
-
-ENTRYPOINT ["/usr/local/bin/compute_ctl"]

Makefile

@@ -20,18 +20,18 @@ else
 	$(error Bad build type '$(BUILD_TYPE)', see Makefile for options)
 endif
 
-# Seccomp BPF is only available for Linux
 UNAME_S := $(shell uname -s)
 ifeq ($(UNAME_S),Linux)
+	# Seccomp BPF is only available for Linux
 	PG_CONFIGURE_OPTS += --with-libseccomp
-endif
-
-# macOS with brew-installed openssl requires explicit paths
-# It can be configured with OPENSSL_PREFIX variable
-UNAME_S := $(shell uname -s)
-ifeq ($(UNAME_S),Darwin)
-    OPENSSL_PREFIX ?= $(shell brew --prefix openssl@3)
-    PG_CONFIGURE_OPTS += --with-includes=$(OPENSSL_PREFIX)/include --with-libraries=$(OPENSSL_PREFIX)/lib
+else ifeq ($(UNAME_S),Darwin)
+	# macOS with brew-installed openssl requires explicit paths
+	# It can be configured with OPENSSL_PREFIX variable
+	OPENSSL_PREFIX ?= $(shell brew --prefix openssl@3)
+	PG_CONFIGURE_OPTS += --with-includes=$(OPENSSL_PREFIX)/include --with-libraries=$(OPENSSL_PREFIX)/lib
+	# macOS already has bison and flex in the system, but they are old and result in postgres-v14 target failure
+	# brew formulae are keg-only and not symlinked into HOMEBREW_PREFIX, force their usage
+	EXTRA_PATH_OVERRIDES += $(shell brew --prefix bison)/bin/:$(shell brew --prefix flex)/bin/:
 endif
 
 # Use -C option so that when PostgreSQL "make install" installs the
@@ -73,7 +73,8 @@ $(POSTGRES_INSTALL_DIR)/build/v14/config.status:
 	+@echo "Configuring Postgres v14 build"
 	mkdir -p $(POSTGRES_INSTALL_DIR)/build/v14
 	(cd $(POSTGRES_INSTALL_DIR)/build/v14 && \
-	$(ROOT_PROJECT_DIR)/vendor/postgres-v14/configure CFLAGS='$(PG_CFLAGS)' \
+	env PATH="$(EXTRA_PATH_OVERRIDES):$$PATH" $(ROOT_PROJECT_DIR)/vendor/postgres-v14/configure \
+		CFLAGS='$(PG_CFLAGS)' \
 		$(PG_CONFIGURE_OPTS) \
 		--prefix=$(abspath $(POSTGRES_INSTALL_DIR))/v14 > configure.log)
 
@@ -81,7 +82,8 @@ $(POSTGRES_INSTALL_DIR)/build/v15/config.status:
 	+@echo "Configuring Postgres v15 build"
 	mkdir -p $(POSTGRES_INSTALL_DIR)/build/v15
 	(cd $(POSTGRES_INSTALL_DIR)/build/v15 && \
-	$(ROOT_PROJECT_DIR)/vendor/postgres-v15/configure CFLAGS='$(PG_CFLAGS)' \
+	env PATH="$(EXTRA_PATH_OVERRIDES):$$PATH" $(ROOT_PROJECT_DIR)/vendor/postgres-v15/configure \
+		CFLAGS='$(PG_CFLAGS)' \
 		$(PG_CONFIGURE_OPTS) \
 		--prefix=$(abspath $(POSTGRES_INSTALL_DIR))/v15 > configure.log)
 
@@ -111,6 +113,8 @@ postgres-v14: postgres-v14-configure \
 	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v14 MAKELEVEL=0 install
 	+@echo "Compiling libpq v14"
 	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v14/src/interfaces/libpq install
+	+@echo "Compiling pg_prewarm v14"
+	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v14/contrib/pg_prewarm install
 	+@echo "Compiling pg_buffercache v14"
 	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v14/contrib/pg_buffercache install
 	+@echo "Compiling pageinspect v14"
@@ -123,6 +127,8 @@ postgres-v15: postgres-v15-configure \
 	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v15 MAKELEVEL=0 install
 	+@echo "Compiling libpq v15"
 	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v15/src/interfaces/libpq install
+	+@echo "Compiling pg_prewarm v15"
+	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v15/contrib/pg_prewarm install
 	+@echo "Compiling pg_buffercache v15"
 	$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/v15/contrib/pg_buffercache install
 	+@echo "Compiling pageinspect v15"
@@ -151,6 +157,11 @@ neon-pg-ext-v14: postgres-v14
 	(cd $(POSTGRES_INSTALL_DIR)/build/neon-v14 && \
 	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/v14/bin/pg_config CFLAGS='$(PG_CFLAGS) $(COPT)' \
 		-f $(ROOT_PROJECT_DIR)/pgxn/neon/Makefile install)
+	+@echo "Compiling neon_walredo v14"
+	mkdir -p $(POSTGRES_INSTALL_DIR)/build/neon-walredo-v14
+	(cd $(POSTGRES_INSTALL_DIR)/build/neon-walredo-v14 && \
+	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/v14/bin/pg_config CFLAGS='$(PG_CFLAGS) $(COPT)' \
+		-f $(ROOT_PROJECT_DIR)/pgxn/neon_walredo/Makefile install)
 	+@echo "Compiling neon_test_utils" v14
 	mkdir -p $(POSTGRES_INSTALL_DIR)/build/neon-test-utils-v14
 	(cd $(POSTGRES_INSTALL_DIR)/build/neon-test-utils-v14 && \
@@ -163,6 +174,11 @@ neon-pg-ext-v15: postgres-v15
 	(cd $(POSTGRES_INSTALL_DIR)/build/neon-v15 && \
 	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/v15/bin/pg_config CFLAGS='$(PG_CFLAGS) $(COPT)' \
 		-f $(ROOT_PROJECT_DIR)/pgxn/neon/Makefile install)
+	+@echo "Compiling neon_walredo v15"
+	mkdir -p $(POSTGRES_INSTALL_DIR)/build/neon-walredo-v15
+	(cd $(POSTGRES_INSTALL_DIR)/build/neon-walredo-v15 && \
+	$(MAKE) PG_CONFIG=$(POSTGRES_INSTALL_DIR)/v15/bin/pg_config CFLAGS='$(PG_CFLAGS) $(COPT)' \
+		-f $(ROOT_PROJECT_DIR)/pgxn/neon_walredo/Makefile install)
 	+@echo "Compiling neon_test_utils" v15
 	mkdir -p $(POSTGRES_INSTALL_DIR)/build/neon-test-utils-v15
 	(cd $(POSTGRES_INSTALL_DIR)/build/neon-test-utils-v15 && \

README.md

@@ -53,7 +53,7 @@ curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
 1. Install XCode and dependencies
 ```
 xcode-select --install
-brew install protobuf etcd openssl
+brew install protobuf etcd openssl flex bison
 ```
 
 2. [Install Rust](https://www.rust-lang.org/tools/install)
@@ -125,24 +125,23 @@ Python (3.9 or higher), and install python3 packages using `./scripts/pysync` (r
 # Create repository in .neon with proper paths to binaries and data
 # Later that would be responsibility of a package install script
 > ./target/debug/neon_local init
-Starting pageserver at '127.0.0.1:64000' in '.neon'
-
-Pageserver started
-Successfully initialized timeline 7dd0907914ac399ff3be45fb252bfdb7
-Stopping pageserver gracefully...done!
+Starting pageserver at '127.0.0.1:64000' in '.neon'.
+pageserver started, pid: 2545906
+Successfully initialized timeline de200bd42b49cc1814412c7e592dd6e9
+Stopped pageserver 1 process with pid 2545906
 
 # start pageserver and safekeeper
 > ./target/debug/neon_local start
-Starting etcd broker using /usr/bin/etcd
-Starting pageserver at '127.0.0.1:64000' in '.neon'
-
-Pageserver started
-Starting safekeeper at '127.0.0.1:5454' in '.neon/safekeepers/sk1'
-Safekeeper started
+Starting etcd broker using "/usr/bin/etcd"
+etcd started, pid: 2545996
+Starting pageserver at '127.0.0.1:64000' in '.neon'.
+pageserver started, pid: 2546005
+Starting safekeeper at '127.0.0.1:5454' in '.neon/safekeepers/sk1'.
+safekeeper 1 started, pid: 2546041
 
 # start postgres compute node
 > ./target/debug/neon_local pg start main
-Starting new postgres main on timeline de200bd42b49cc1814412c7e592dd6e9 ...
+Starting new postgres (v14) main on timeline de200bd42b49cc1814412c7e592dd6e9 ...
 Extracting base backup to create postgres instance: path=.neon/pgdatadirs/tenants/9ef87a5bf0d92544f6fafeeb3239695c/main port=55432
 Starting postgres node at 'host=127.0.0.1 port=55432 user=cloud_admin dbname=postgres'
 
@@ -223,10 +222,7 @@ Ensure your dependencies are installed as described [here](https://github.com/ne
 ```sh
 git clone --recursive https://github.com/neondatabase/neon.git
 
-# either:
 CARGO_BUILD_FLAGS="--features=testing" make
-# or:
-make debug
 
 ./scripts/pytest
 ```

cli-v2-story.md

@@ -1,188 +0,0 @@
-Create a new Zenith repository in the current directory:
-
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli init
-    The files belonging to this database system will be owned by user "heikki".
-    This user must also own the server process.
-    
-    The database cluster will be initialized with locale "en_GB.UTF-8".
-    The default database encoding has accordingly been set to "UTF8".
-    The default text search configuration will be set to "english".
-    
-    Data page checksums are disabled.
-    
-    creating directory tmp ... ok
-    creating subdirectories ... ok
-    selecting dynamic shared memory implementation ... posix
-    selecting default max_connections ... 100
-    selecting default shared_buffers ... 128MB
-    selecting default time zone ... Europe/Helsinki
-    creating configuration files ... ok
-    running bootstrap script ... ok
-    performing post-bootstrap initialization ... ok
-    syncing data to disk ... ok
-    
-    initdb: warning: enabling "trust" authentication for local connections
-    You can change this by editing pg_hba.conf or using the option -A, or
-    --auth-local and --auth-host, the next time you run initdb.
-    new zenith repository was created in .zenith
-
-Initially, there is only one branch:
-
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli branch
-      main
-
-Start a local Postgres instance on the branch:
-
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli start main
-    Creating data directory from snapshot at 0/15FFB08...
-    waiting for server to start....2021-04-13 09:27:43.919 EEST [984664] LOG:  starting PostgreSQL 14devel on x86_64-pc-linux-gnu, compiled by gcc (Debian 10.2.1-6) 10.2.1 20210110, 64-bit
-    2021-04-13 09:27:43.920 EEST [984664] LOG:  listening on IPv6 address "::1", port 5432
-    2021-04-13 09:27:43.920 EEST [984664] LOG:  listening on IPv4 address "127.0.0.1", port 5432
-    2021-04-13 09:27:43.927 EEST [984664] LOG:  listening on Unix socket "/tmp/.s.PGSQL.5432"
-    2021-04-13 09:27:43.939 EEST [984665] LOG:  database system was interrupted; last known up at 2021-04-13 09:27:33 EEST
-    2021-04-13 09:27:43.939 EEST [984665] LOG:  creating missing WAL directory "pg_wal/archive_status"
-    2021-04-13 09:27:44.189 EEST [984665] LOG:  database system was not properly shut down; automatic recovery in progress
-    2021-04-13 09:27:44.195 EEST [984665] LOG:  invalid record length at 0/15FFB80: wanted 24, got 0
-    2021-04-13 09:27:44.195 EEST [984665] LOG:  redo is not required
-    2021-04-13 09:27:44.225 EEST [984664] LOG:  database system is ready to accept connections
-     done
-    server started
-
-Run some commands against it:
-
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -c "create table foo (t text);" 
-    CREATE TABLE
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -c "insert into foo values ('inserted on the main branch');" 
-    INSERT 0 1
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -c "select * from foo" 
-                  t              
-    -----------------------------
-     inserted on the main branch
-    (1 row)
-
-Create a new branch called 'experimental'. We create it from the
-current end of the 'main' branch, but you could specify a different
-LSN as the start point instead.
-
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli branch experimental main
-    branching at end of WAL: 0/161F478
-    
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli branch 
-      experimental
-      main
-
-Start another Postgres instance off the 'experimental' branch:
-
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli start experimental -- -o -p5433
-    Creating data directory from snapshot at 0/15FFB08...
-    waiting for server to start....2021-04-13 09:28:41.874 EEST [984766] LOG:  starting PostgreSQL 14devel on x86_64-pc-linux-gnu, compiled by gcc (Debian 10.2.1-6) 10.2.1 20210110, 64-bit
-    2021-04-13 09:28:41.875 EEST [984766] LOG:  listening on IPv6 address "::1", port 5433
-    2021-04-13 09:28:41.875 EEST [984766] LOG:  listening on IPv4 address "127.0.0.1", port 5433
-    2021-04-13 09:28:41.883 EEST [984766] LOG:  listening on Unix socket "/tmp/.s.PGSQL.5433"
-    2021-04-13 09:28:41.896 EEST [984767] LOG:  database system was interrupted; last known up at 2021-04-13 09:27:33 EEST
-    2021-04-13 09:28:42.265 EEST [984767] LOG:  database system was not properly shut down; automatic recovery in progress
-    2021-04-13 09:28:42.269 EEST [984767] LOG:  redo starts at 0/15FFB80
-    2021-04-13 09:28:42.272 EEST [984767] LOG:  invalid record length at 0/161F4B0: wanted 24, got 0
-    2021-04-13 09:28:42.272 EEST [984767] LOG:  redo done at 0/161F478 system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.00 s
-    2021-04-13 09:28:42.321 EEST [984766] LOG:  database system is ready to accept connections
-     done
-    server started
-
-Insert some a row on the 'experimental' branch:
-
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -p5433 -c "select * from foo" 
-                  t              
-    -----------------------------
-     inserted on the main branch
-    (1 row)
-    
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -p5433 -c "insert into foo values ('inserted on experimental')" 
-    INSERT 0 1
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -p5433 -c "select * from foo" 
-                  t              
-    -----------------------------
-     inserted on the main branch
-     inserted on experimental
-    (2 rows)
-    
-See that the other Postgres instance is still running on 'main' branch on port 5432:
-
-
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -p5432 -c "select * from foo" 
-                  t              
-    -----------------------------
-     inserted on the main branch
-    (1 row)
-
-
-
-
-Everything is stored in the .zenith directory:
-
-    ~/git-sandbox/zenith (cli-v2)$ ls -l .zenith/
-    total 12
-    drwxr-xr-x 4 heikki heikki 4096 Apr 13 09:28 datadirs
-    drwxr-xr-x 4 heikki heikki 4096 Apr 13 09:27 refs
-    drwxr-xr-x 4 heikki heikki 4096 Apr 13 09:28 timelines
-
-The 'datadirs' directory contains the datadirs of the running instances:
-
-    ~/git-sandbox/zenith (cli-v2)$ ls -l .zenith/datadirs/
-    total 8
-    drwx------ 18 heikki heikki 4096 Apr 13 09:27 3c0c634c1674079b2c6d4edf7c91523e
-    drwx------ 18 heikki heikki 4096 Apr 13 09:28 697e3c103d4b1763cd6e82e4ff361d76
-    ~/git-sandbox/zenith (cli-v2)$ ls -l .zenith/datadirs/3c0c634c1674079b2c6d4edf7c91523e/
-    total 124
-    drwxr-xr-x 5 heikki heikki  4096 Apr 13 09:27 base
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 global
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_commit_ts
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_dynshmem
-    -rw------- 1 heikki heikki  4760 Apr 13 09:27 pg_hba.conf
-    -rw------- 1 heikki heikki  1636 Apr 13 09:27 pg_ident.conf
-    drwxr-xr-x 4 heikki heikki  4096 Apr 13 09:32 pg_logical
-    drwxr-xr-x 4 heikki heikki  4096 Apr 13 09:27 pg_multixact
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_notify
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_replslot
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_serial
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_snapshots
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_stat
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:34 pg_stat_tmp
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_subtrans
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_tblspc
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_twophase
-    -rw------- 1 heikki heikki     3 Apr 13 09:27 PG_VERSION
-    lrwxrwxrwx 1 heikki heikki    52 Apr 13 09:27 pg_wal -> ../../timelines/3c0c634c1674079b2c6d4edf7c91523e/wal
-    drwxr-xr-x 2 heikki heikki  4096 Apr 13 09:27 pg_xact
-    -rw------- 1 heikki heikki    88 Apr 13 09:27 postgresql.auto.conf
-    -rw------- 1 heikki heikki 28688 Apr 13 09:27 postgresql.conf
-    -rw------- 1 heikki heikki    96 Apr 13 09:27 postmaster.opts
-    -rw------- 1 heikki heikki   149 Apr 13 09:27 postmaster.pid
-
-Note how 'pg_wal' is just a symlink to the 'timelines' directory. The
-datadir is ephemeral, you can delete it at any time, and it can be reconstructed
-from the snapshots and WAL stored in the 'timelines' directory. So if you push/pull
-the repository, the 'datadirs' are not included. (They are like git working trees)
-
-    ~/git-sandbox/zenith (cli-v2)$ killall -9 postgres
-    ~/git-sandbox/zenith (cli-v2)$ rm -rf .zenith/datadirs/*
-    ~/git-sandbox/zenith (cli-v2)$ ./target/debug/cli start experimental -- -o -p5433
-    Creating data directory from snapshot at 0/15FFB08...
-    waiting for server to start....2021-04-13 09:37:05.476 EEST [985340] LOG:  starting PostgreSQL 14devel on x86_64-pc-linux-gnu, compiled by gcc (Debian 10.2.1-6) 10.2.1 20210110, 64-bit
-    2021-04-13 09:37:05.477 EEST [985340] LOG:  listening on IPv6 address "::1", port 5433
-    2021-04-13 09:37:05.477 EEST [985340] LOG:  listening on IPv4 address "127.0.0.1", port 5433
-    2021-04-13 09:37:05.487 EEST [985340] LOG:  listening on Unix socket "/tmp/.s.PGSQL.5433"
-    2021-04-13 09:37:05.498 EEST [985341] LOG:  database system was interrupted; last known up at 2021-04-13 09:27:33 EEST
-    2021-04-13 09:37:05.808 EEST [985341] LOG:  database system was not properly shut down; automatic recovery in progress
-    2021-04-13 09:37:05.813 EEST [985341] LOG:  redo starts at 0/15FFB80
-    2021-04-13 09:37:05.815 EEST [985341] LOG:  invalid record length at 0/161F770: wanted 24, got 0
-    2021-04-13 09:37:05.815 EEST [985341] LOG:  redo done at 0/161F738 system usage: CPU: user: 0.00 s, system: 0.00 s, elapsed: 0.00 s
-    2021-04-13 09:37:05.866 EEST [985340] LOG:  database system is ready to accept connections
-     done
-    server started
-    ~/git-sandbox/zenith (cli-v2)$ psql postgres -p5433 -c "select * from foo" 
-                  t              
-    -----------------------------
-     inserted on the main branch
-     inserted on experimental
-    (2 rows)
-

compute_tools/Cargo.toml

@@ -6,7 +6,7 @@ edition = "2021"
 [dependencies]
 anyhow = "1.0"
 chrono = "0.4"
-clap = "3.0"
+clap = "4.0"
 env_logger = "0.9"
 futures = "0.3.13"
 hyper = { version = "0.14", features = ["full"] }

compute_tools/src/bin/compute_ctl.rs

@@ -51,53 +51,19 @@ fn main() -> Result<()> {
     // TODO: re-use `utils::logging` later
     init_logger(DEFAULT_LOG_LEVEL)?;
 
-    // Env variable is set by `cargo`
-    let version: Option<&str> = option_env!("CARGO_PKG_VERSION");
-    let matches = clap::App::new("compute_ctl")
-        .version(version.unwrap_or("unknown"))
-        .arg(
-            Arg::new("connstr")
-                .short('C')
-                .long("connstr")
-                .value_name("DATABASE_URL")
-                .required(true),
-        )
-        .arg(
-            Arg::new("pgdata")
-                .short('D')
-                .long("pgdata")
-                .value_name("DATADIR")
-                .required(true),
-        )
-        .arg(
-            Arg::new("pgbin")
-                .short('b')
-                .long("pgbin")
-                .value_name("POSTGRES_PATH"),
-        )
-        .arg(
-            Arg::new("spec")
-                .short('s')
-                .long("spec")
-                .value_name("SPEC_JSON"),
-        )
-        .arg(
-            Arg::new("spec-path")
-                .short('S')
-                .long("spec-path")
-                .value_name("SPEC_PATH"),
-        )
-        .get_matches();
+    let matches = cli().get_matches();
 
-    let pgdata = matches.value_of("pgdata").expect("PGDATA path is required");
+    let pgdata = matches
+        .get_one::<String>("pgdata")
+        .expect("PGDATA path is required");
     let connstr = matches
-        .value_of("connstr")
+        .get_one::<String>("connstr")
         .expect("Postgres connection string is required");
-    let spec = matches.value_of("spec");
-    let spec_path = matches.value_of("spec-path");
+    let spec = matches.get_one::<String>("spec");
+    let spec_path = matches.get_one::<String>("spec-path");
 
     // Try to use just 'postgres' if no path is provided
-    let pgbin = matches.value_of("pgbin").unwrap_or("postgres");
+    let pgbin = matches.get_one::<String>("pgbin").unwrap();
 
     let spec: ComputeSpec = match spec {
         // First, try to get cluster spec from the cli argument
@@ -173,3 +139,48 @@ fn main() -> Result<()> {
         }
     }
 }
+
+fn cli() -> clap::Command {
+    // Env variable is set by `cargo`
+    let version = option_env!("CARGO_PKG_VERSION").unwrap_or("unknown");
+    clap::Command::new("compute_ctl")
+        .version(version)
+        .arg(
+            Arg::new("connstr")
+                .short('C')
+                .long("connstr")
+                .value_name("DATABASE_URL")
+                .required(true),
+        )
+        .arg(
+            Arg::new("pgdata")
+                .short('D')
+                .long("pgdata")
+                .value_name("DATADIR")
+                .required(true),
+        )
+        .arg(
+            Arg::new("pgbin")
+                .short('b')
+                .long("pgbin")
+                .default_value("postgres")
+                .value_name("POSTGRES_PATH"),
+        )
+        .arg(
+            Arg::new("spec")
+                .short('s')
+                .long("spec")
+                .value_name("SPEC_JSON"),
+        )
+        .arg(
+            Arg::new("spec-path")
+                .short('S')
+                .long("spec-path")
+                .value_name("SPEC_PATH"),
+        )
+}
+
+#[test]
+fn verify_cli() {
+    cli().debug_assert()
+}

compute_tools/src/compute.rs

@@ -178,7 +178,6 @@ impl ComputeNode {
             .args(&["--sync-safekeepers"])
             .env("PGDATA", &self.pgdata) // we cannot use -D in this mode
             .stdout(Stdio::piped())
-            .stderr(Stdio::piped())
             .spawn()
             .expect("postgres --sync-safekeepers failed to start");
 
@@ -191,10 +190,10 @@ impl ComputeNode {
 
         if !sync_output.status.success() {
             anyhow::bail!(
-                "postgres --sync-safekeepers exited with non-zero status: {}. stdout: {}, stderr: {}",
+                "postgres --sync-safekeepers exited with non-zero status: {}. stdout: {}",
                 sync_output.status,
-                String::from_utf8(sync_output.stdout).expect("postgres --sync-safekeepers exited, and stdout is not utf-8"),
-                String::from_utf8(sync_output.stderr).expect("postgres --sync-safekeepers exited, and stderr is not utf-8"),
+                String::from_utf8(sync_output.stdout)
+                    .expect("postgres --sync-safekeepers exited, and stdout is not utf-8"),
             );
         }
 

compute_tools/src/pg_helpers.rs

@@ -8,11 +8,10 @@ use std::process::Child;
 use std::time::{Duration, Instant};
 
 use anyhow::{bail, Result};
+use notify::{RecursiveMode, Watcher};
 use postgres::{Client, Transaction};
 use serde::Deserialize;
 
-use notify::{RecursiveMode, Watcher};
-
 const POSTGRES_WAIT_TIMEOUT: Duration = Duration::from_millis(60 * 1000); // milliseconds
 
 /// Rust representation of Postgres role info with only those fields
@@ -66,7 +65,7 @@ impl GenericOption {
             let name = match self.name.as_str() {
                 "safekeepers" => "neon.safekeepers",
                 "wal_acceptor_reconnect" => "neon.safekeeper_reconnect_timeout",
-                "wal_acceptor_connect_timeout" => "neon.safekeeper_connect_timeout",
+                "wal_acceptor_connection_timeout" => "neon.safekeeper_connection_timeout",
                 it => it,
             };
 
@@ -169,7 +168,7 @@ impl Database {
     /// it may require a proper quoting too.
     pub fn to_pg_options(&self) -> String {
         let mut params: String = self.options.as_pg_options();
-        write!(params, " OWNER {}", &self.owner.quote())
+        write!(params, " OWNER {}", &self.owner.pg_quote())
             .expect("String is documented to not to error during write operations");
 
         params
@@ -180,18 +179,17 @@ impl Database {
 /// intended to be used for DB / role names.
 pub type PgIdent = String;
 
-/// Generic trait used to provide quoting for strings used in the
-/// Postgres SQL queries. Currently used only to implement quoting
-/// of identifiers, but could be used for literals in the future.
-pub trait PgQuote {
-    fn quote(&self) -> String;
+/// Generic trait used to provide quoting / encoding for strings used in the
+/// Postgres SQL queries and DATABASE_URL.
+pub trait Escaping {
+    fn pg_quote(&self) -> String;
 }
 
-impl PgQuote for PgIdent {
+impl Escaping for PgIdent {
     /// This is intended to mimic Postgres quote_ident(), but for simplicity it
-    /// always quotes provided string with `""` and escapes every `"`. Not idempotent,
-    /// i.e. if string is already escaped it will be escaped again.
-    fn quote(&self) -> String {
+    /// always quotes provided string with `""` and escapes every `"`.
+    /// **Not idempotent**, i.e. if string is already escaped it will be escaped again.
+    fn pg_quote(&self) -> String {
         let result = format!("\"{}\"", self.replace('"', "\"\""));
         result
     }
@@ -250,9 +248,36 @@ pub fn wait_for_postgres(pg: &mut Child, pgdata: &Path) -> Result<()> {
     // case we miss some events for some reason. Not strictly necessary, but
     // better safe than sorry.
     let (tx, rx) = std::sync::mpsc::channel();
-    let mut watcher = notify::recommended_watcher(move |res| {
+    let (mut watcher, rx): (Box<dyn Watcher>, _) = match notify::recommended_watcher(move |res| {
         let _ = tx.send(res);
-    })?;
+    }) {
+        Ok(watcher) => (Box::new(watcher), rx),
+        Err(e) => {
+            match e.kind {
+                notify::ErrorKind::Io(os) if os.raw_os_error() == Some(38) => {
+                    // docker on m1 macs does not support recommended_watcher
+                    // but return "Function not implemented (os error 38)"
+                    // see https://github.com/notify-rs/notify/issues/423
+                    let (tx, rx) = std::sync::mpsc::channel();
+
+                    // let's poll it faster than what we check the results for (100ms)
+                    let config =
+                        notify::Config::default().with_poll_interval(Duration::from_millis(50));
+
+                    let watcher = notify::PollWatcher::new(
+                        move |res| {
+                            let _ = tx.send(res);
+                        },
+                        config,
+                    )?;
+
+                    (Box::new(watcher), rx)
+                }
+                _ => return Err(e.into()),
+            }
+        }
+    };
+
     watcher.watch(pgdata, RecursiveMode::NonRecursive)?;
 
     let started_at = Instant::now();

compute_tools/src/spec.rs

@@ -1,7 +1,9 @@
 use std::path::Path;
+use std::str::FromStr;
 
 use anyhow::Result;
 use log::{info, log_enabled, warn, Level};
+use postgres::config::Config;
 use postgres::{Client, NoTls};
 use serde::Deserialize;
 
@@ -115,8 +117,8 @@ pub fn handle_roles(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
                     if existing_roles.iter().any(|r| r.name == op.name) {
                         let query: String = format!(
                             "ALTER ROLE {} RENAME TO {}",
-                            op.name.quote(),
-                            new_name.quote()
+                            op.name.pg_quote(),
+                            new_name.pg_quote()
                         );
 
                         warn!("renaming role '{}' to '{}'", op.name, new_name);
@@ -162,7 +164,7 @@ pub fn handle_roles(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
             }
 
             if update_role {
-                let mut query: String = format!("ALTER ROLE {} ", name.quote());
+                let mut query: String = format!("ALTER ROLE {} ", name.pg_quote());
                 info_print!(" -> update");
 
                 query.push_str(&role.to_pg_options());
@@ -170,7 +172,7 @@ pub fn handle_roles(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
             }
         } else {
             info!("role name: '{}'", &name);
-            let mut query: String = format!("CREATE ROLE {} ", name.quote());
+            let mut query: String = format!("CREATE ROLE {} ", name.pg_quote());
             info!("role create query: '{}'", &query);
             info_print!(" -> create");
 
@@ -179,7 +181,7 @@ pub fn handle_roles(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
 
             let grant_query = format!(
                 "GRANT pg_read_all_data, pg_write_all_data TO {}",
-                name.quote()
+                name.pg_quote()
             );
             xact.execute(grant_query.as_str(), &[])?;
             info!("role grant query: '{}'", &grant_query);
@@ -215,7 +217,7 @@ pub fn handle_role_deletions(node: &ComputeNode, client: &mut Client) -> Result<
             // We do not check either role exists or not,
             // Postgres will take care of it for us
             if op.action == "delete_role" {
-                let query: String = format!("DROP ROLE IF EXISTS {}", &op.name.quote());
+                let query: String = format!("DROP ROLE IF EXISTS {}", &op.name.pg_quote());
 
                 warn!("deleting role '{}'", &op.name);
                 xact.execute(query.as_str(), &[])?;
@@ -230,17 +232,16 @@ pub fn handle_role_deletions(node: &ComputeNode, client: &mut Client) -> Result<
 fn reassign_owned_objects(node: &ComputeNode, role_name: &PgIdent) -> Result<()> {
     for db in &node.spec.cluster.databases {
         if db.owner != *role_name {
-            let mut connstr = node.connstr.clone();
-            // database name is always the last and the only component of the path
-            connstr.set_path(&db.name);
+            let mut conf = Config::from_str(node.connstr.as_str())?;
+            conf.dbname(&db.name);
 
-            let mut client = Client::connect(connstr.as_str(), NoTls)?;
+            let mut client = conf.connect(NoTls)?;
 
             // This will reassign all dependent objects to the db owner
             let reassign_query = format!(
                 "REASSIGN OWNED BY {} TO {}",
-                role_name.quote(),
-                db.owner.quote()
+                role_name.pg_quote(),
+                db.owner.pg_quote()
             );
             info!(
                 "reassigning objects owned by '{}' in db '{}' to '{}'",
@@ -249,7 +250,7 @@ fn reassign_owned_objects(node: &ComputeNode, role_name: &PgIdent) -> Result<()>
             client.simple_query(&reassign_query)?;
 
             // This now will only drop privileges of the role
-            let drop_query = format!("DROP OWNED BY {}", role_name.quote());
+            let drop_query = format!("DROP OWNED BY {}", role_name.pg_quote());
             client.simple_query(&drop_query)?;
         }
     }
@@ -279,7 +280,7 @@ pub fn handle_databases(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
                 // We do not check either DB exists or not,
                 // Postgres will take care of it for us
                 "delete_db" => {
-                    let query: String = format!("DROP DATABASE IF EXISTS {}", &op.name.quote());
+                    let query: String = format!("DROP DATABASE IF EXISTS {}", &op.name.pg_quote());
 
                     warn!("deleting database '{}'", &op.name);
                     client.execute(query.as_str(), &[])?;
@@ -291,8 +292,8 @@ pub fn handle_databases(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
                     if existing_dbs.iter().any(|r| r.name == op.name) {
                         let query: String = format!(
                             "ALTER DATABASE {} RENAME TO {}",
-                            op.name.quote(),
-                            new_name.quote()
+                            op.name.pg_quote(),
+                            new_name.pg_quote()
                         );
 
                         warn!("renaming database '{}' to '{}'", op.name, new_name);
@@ -320,7 +321,7 @@ pub fn handle_databases(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
             // XXX: db owner name is returned as quoted string from Postgres,
             // when quoting is needed.
             let new_owner = if r.owner.starts_with('"') {
-                db.owner.quote()
+                db.owner.pg_quote()
             } else {
                 db.owner.clone()
             };
@@ -328,15 +329,15 @@ pub fn handle_databases(spec: &ComputeSpec, client: &mut Client) -> Result<()> {
             if new_owner != r.owner {
                 let query: String = format!(
                     "ALTER DATABASE {} OWNER TO {}",
-                    name.quote(),
-                    db.owner.quote()
+                    name.pg_quote(),
+                    db.owner.pg_quote()
                 );
                 info_print!(" -> update");
 
                 client.execute(query.as_str(), &[])?;
             }
         } else {
-            let mut query: String = format!("CREATE DATABASE {} ", name.quote());
+            let mut query: String = format!("CREATE DATABASE {} ", name.pg_quote());
             info_print!(" -> create");
 
             query.push_str(&db.to_pg_options());
@@ -366,7 +367,7 @@ pub fn handle_grants(node: &ComputeNode, client: &mut Client) -> Result<()> {
         .cluster
         .roles
         .iter()
-        .map(|r| r.name.quote())
+        .map(|r| r.name.pg_quote())
         .collect::<Vec<_>>();
 
     for db in &spec.cluster.databases {
@@ -374,7 +375,7 @@ pub fn handle_grants(node: &ComputeNode, client: &mut Client) -> Result<()> {
 
         let query: String = format!(
             "GRANT CREATE ON DATABASE {} TO {}",
-            dbname.quote(),
+            dbname.pg_quote(),
             roles.join(", ")
         );
         info!("grant query {}", &query);
@@ -385,12 +386,11 @@ pub fn handle_grants(node: &ComputeNode, client: &mut Client) -> Result<()> {
     // Do some per-database access adjustments. We'd better do this at db creation time,
     // but CREATE DATABASE isn't transactional. So we cannot create db + do some grants
     // atomically.
-    let mut db_connstr = node.connstr.clone();
     for db in &node.spec.cluster.databases {
-        // database name is always the last and the only component of the path
-        db_connstr.set_path(&db.name);
+        let mut conf = Config::from_str(node.connstr.as_str())?;
+        conf.dbname(&db.name);
 
-        let mut db_client = Client::connect(db_connstr.as_str(), NoTls)?;
+        let mut db_client = conf.connect(NoTls)?;
 
         // This will only change ownership on the schema itself, not the objects
         // inside it. Without it owner of the `public` schema will be `cloud_admin`
@@ -419,9 +419,36 @@ pub fn handle_grants(node: &ComputeNode, client: &mut Client) -> Result<()> {
                     END IF;\n\
                 END\n\
             $$;",
-            db.owner.quote()
+            db.owner.pg_quote()
         );
         db_client.simple_query(&alter_query)?;
+
+        // Explicitly grant CREATE ON SCHEMA PUBLIC to the web_access user.
+        // This is needed because since postgres 15 this privilege is removed by default.
+        let grant_query = "DO $$\n\
+                BEGIN\n\
+                    IF EXISTS(\n\
+                        SELECT nspname\n\
+                        FROM pg_catalog.pg_namespace\n\
+                        WHERE nspname = 'public'\n\
+                    ) AND\n\
+                    current_setting('server_version_num')::int/10000 >= 15\n\
+                    THEN\n\
+                        IF EXISTS(\n\
+                            SELECT rolname\n\
+                            FROM pg_catalog.pg_roles\n\
+                            WHERE rolname = 'web_access'\n\
+                        )\n\
+                        THEN\n\
+                            GRANT CREATE ON SCHEMA public TO web_access;\n\
+                        END IF;\n\
+                    END IF;\n\
+                END\n\
+            $$;"
+        .to_string();
+
+        info!("grant query for db {} : {}", &db.name, &grant_query);
+        db_client.simple_query(&grant_query)?;
     }
 
     Ok(())

compute_tools/tests/pg_helpers_tests.rs

@@ -33,9 +33,9 @@ mod pg_helpers_tests {
     }
 
     #[test]
-    fn quote_ident() {
+    fn ident_pg_quote() {
         let ident: PgIdent = PgIdent::from("\"name\";\\n select 1;");
 
-        assert_eq!(ident.quote(), "\"\"\"name\"\";\\n select 1;\"");
+        assert_eq!(ident.pg_quote(), "\"\"\"name\"\";\\n select 1;\"");
     }
 }

control_plane/Cargo.toml

@@ -4,20 +4,21 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
-clap = "3.0"
-comfy-table = "5.0.1"
-git-version = "0.3.5"
-tar = "0.4.38"
-postgres = { git = "https://github.com/neondatabase/rust-postgres.git", rev="d052ee8b86fff9897c77b0fe89ea9daba0e1fa38" }
-serde = { version = "1.0", features = ["derive"] }
-serde_with = "1.12.0"
-toml = "0.5"
-once_cell = "1.13.0"
-regex = "1"
 anyhow = "1.0"
-thiserror = "1"
-nix = "0.23"
+clap = "4.0"
+comfy-table = "6.1"
+git-version = "0.3.5"
+nix = "0.25"
+once_cell = "1.13.0"
+postgres = { git = "https://github.com/neondatabase/rust-postgres.git", rev = "d052ee8b86fff9897c77b0fe89ea9daba0e1fa38" }
+regex = "1"
 reqwest = { version = "0.11", default-features = false, features = ["blocking", "json", "rustls-tls"] }
+serde = { version = "1.0", features = ["derive"] }
+serde_with = "2.0"
+tar = "0.4.38"
+thiserror = "1"
+toml = "0.5"
+url = "2.2.2"
 
 # Note: Do not directly depend on pageserver or safekeeper; use pageserver_api or safekeeper_api
 # instead, so that recompile times are better.

control_plane/src/background_process.rs

@@ -0,0 +1,279 @@
+//! Spawns and kills background processes that are needed by Neon CLI.
+//! Applies common set-up such as log and pid files (if needed) to every process.
+//!
+//! Neon CLI does not run in background, so it needs to store the information about
+//! spawned processes, which it does in this module.
+//! We do that by storing the pid of the process in the "${process_name}.pid" file.
+//! The pid file can be created by the process itself
+//! (Neon storage binaries do that and also ensure that a lock is taken onto that file)
+//! or we create such file after starting the process
+//! (non-Neon binaries don't necessarily follow our pidfile conventions).
+//! The pid stored in the file is later used to stop the service.
+//!
+//! See [`lock_file`] module for more info.
+
+use std::ffi::OsStr;
+use std::io::Write;
+use std::path::Path;
+use std::process::{Child, Command};
+use std::time::Duration;
+use std::{fs, io, thread};
+
+use anyhow::{anyhow, bail, Context, Result};
+use nix::errno::Errno;
+use nix::sys::signal::{kill, Signal};
+use nix::unistd::Pid;
+
+use utils::lock_file;
+
+// These constants control the loop used to poll for process start / stop.
+//
+// The loop waits for at most 10 seconds, polling every 100 ms.
+// Once a second, it prints a dot ("."), to give the user an indication that
+// it's waiting. If the process hasn't started/stopped after 5 seconds,
+// it prints a notice that it's taking long, but keeps waiting.
+//
+const RETRY_UNTIL_SECS: u64 = 10;
+const RETRIES: u64 = (RETRY_UNTIL_SECS * 1000) / RETRY_INTERVAL_MILLIS;
+const RETRY_INTERVAL_MILLIS: u64 = 100;
+const DOT_EVERY_RETRIES: u64 = 10;
+const NOTICE_AFTER_RETRIES: u64 = 50;
+
+/// Argument to `start_process`, to indicate whether it should create pidfile or if the process creates
+/// it itself.
+pub enum InitialPidFile<'t> {
+    /// Create a pidfile, to allow future CLI invocations to manipulate the process.
+    Create(&'t Path),
+    /// The process will create the pidfile itself, need to wait for that event.
+    Expect(&'t Path),
+}
+
+/// Start a background child process using the parameters given.
+pub fn start_process<F, S: AsRef<OsStr>>(
+    process_name: &str,
+    datadir: &Path,
+    command: &Path,
+    args: &[S],
+    initial_pid_file: InitialPidFile,
+    process_status_check: F,
+) -> anyhow::Result<Child>
+where
+    F: Fn() -> anyhow::Result<bool>,
+{
+    let log_path = datadir.join(format!("{process_name}.log"));
+    let process_log_file = fs::OpenOptions::new()
+        .create(true)
+        .write(true)
+        .append(true)
+        .open(&log_path)
+        .with_context(|| {
+            format!("Could not open {process_name} log file {log_path:?} for writing")
+        })?;
+    let same_file_for_stderr = process_log_file.try_clone().with_context(|| {
+        format!("Could not reuse {process_name} log file {log_path:?} for writing stderr")
+    })?;
+
+    let mut command = Command::new(command);
+    let background_command = command
+        .stdout(process_log_file)
+        .stderr(same_file_for_stderr)
+        .args(args);
+    let filled_cmd = fill_aws_secrets_vars(fill_rust_env_vars(background_command));
+
+    let mut spawned_process = filled_cmd.spawn().with_context(|| {
+        format!("Could not spawn {process_name}, see console output and log files for details.")
+    })?;
+    let pid = spawned_process.id();
+    let pid = Pid::from_raw(
+        i32::try_from(pid)
+            .with_context(|| format!("Subprocess {process_name} has invalid pid {pid}"))?,
+    );
+
+    let pid_file_to_check = match initial_pid_file {
+        InitialPidFile::Create(target_pid_file_path) => {
+            match lock_file::create_lock_file(target_pid_file_path, pid.to_string()) {
+                lock_file::LockCreationResult::Created { .. } => {
+                    // We use "lock" file here only to create the pid file. The lock on the pidfile will be dropped as soon
+                    // as this CLI invocation exits, so it's a bit useless, but doesn't any harm either.
+                }
+                lock_file::LockCreationResult::AlreadyLocked { .. } => {
+                    anyhow::bail!("Cannot write pid file for {process_name} at path {target_pid_file_path:?}: file is already locked by another process")
+                }
+                lock_file::LockCreationResult::CreationFailed(e) => {
+                    return Err(e.context(format!(
+                    "Failed to create pid file for {process_name} at path {target_pid_file_path:?}"
+                )))
+                }
+            }
+            None
+        }
+        InitialPidFile::Expect(pid_file_path) => Some(pid_file_path),
+    };
+
+    for retries in 0..RETRIES {
+        match process_started(pid, pid_file_to_check, &process_status_check) {
+            Ok(true) => {
+                println!("\n{process_name} started, pid: {pid}");
+                return Ok(spawned_process);
+            }
+            Ok(false) => {
+                if retries == NOTICE_AFTER_RETRIES {
+                    // The process is taking a long time to start up. Keep waiting, but
+                    // print a message
+                    print!("\n{process_name} has not started yet, continuing to wait");
+                }
+                if retries % DOT_EVERY_RETRIES == 0 {
+                    print!(".");
+                    io::stdout().flush().unwrap();
+                }
+                thread::sleep(Duration::from_millis(RETRY_INTERVAL_MILLIS));
+            }
+            Err(e) => {
+                println!("{process_name} failed to start: {e:#}");
+                if let Err(e) = spawned_process.kill() {
+                    println!("Could not stop {process_name} subprocess: {e:#}")
+                };
+                return Err(e);
+            }
+        }
+    }
+    println!();
+    anyhow::bail!("{process_name} did not start in {RETRY_UNTIL_SECS} seconds");
+}
+
+/// Stops the process, using the pid file given. Returns Ok also if the process is already not running.
+pub fn stop_process(immediate: bool, process_name: &str, pid_file: &Path) -> anyhow::Result<()> {
+    if !pid_file.exists() {
+        println!("{process_name} is already stopped: no pid file {pid_file:?} is present");
+        return Ok(());
+    }
+    let pid = read_pidfile(pid_file)?;
+
+    let sig = if immediate {
+        print!("Stopping {process_name} with pid {pid} immediately..");
+        Signal::SIGQUIT
+    } else {
+        print!("Stopping {process_name} with pid {pid} gracefully..");
+        Signal::SIGTERM
+    };
+    io::stdout().flush().unwrap();
+    match kill(pid, sig) {
+        Ok(()) => (),
+        Err(Errno::ESRCH) => {
+            println!(
+                "{process_name} with pid {pid} does not exist, but a pid file {pid_file:?} was found"
+            );
+            return Ok(());
+        }
+        Err(e) => anyhow::bail!("Failed to send signal to {process_name} with pid {pid}: {e}"),
+    }
+
+    // Wait until process is gone
+    for retries in 0..RETRIES {
+        match process_has_stopped(pid) {
+            Ok(true) => {
+                println!("\n{process_name} stopped");
+                if let Err(e) = fs::remove_file(pid_file) {
+                    if e.kind() != io::ErrorKind::NotFound {
+                        eprintln!("Failed to remove pid file {pid_file:?} after stopping the process: {e:#}");
+                    }
+                }
+                return Ok(());
+            }
+            Ok(false) => {
+                if retries == NOTICE_AFTER_RETRIES {
+                    // The process is taking a long time to start up. Keep waiting, but
+                    // print a message
+                    print!("\n{process_name} has not stopped yet, continuing to wait");
+                }
+                if retries % DOT_EVERY_RETRIES == 0 {
+                    print!(".");
+                    io::stdout().flush().unwrap();
+                }
+                thread::sleep(Duration::from_millis(RETRY_INTERVAL_MILLIS));
+            }
+            Err(e) => {
+                println!("{process_name} with pid {pid} failed to stop: {e:#}");
+                return Err(e);
+            }
+        }
+    }
+    println!();
+    anyhow::bail!("{process_name} with pid {pid} did not stop in {RETRY_UNTIL_SECS} seconds");
+}
+
+fn fill_rust_env_vars(cmd: &mut Command) -> &mut Command {
+    let mut filled_cmd = cmd.env_clear().env("RUST_BACKTRACE", "1");
+
+    // Pass through these environment variables to the command
+    for var in ["LLVM_PROFILE_FILE", "FAILPOINTS", "RUST_LOG"] {
+        if let Some(val) = std::env::var_os(var) {
+            filled_cmd = filled_cmd.env(var, val);
+        }
+    }
+
+    filled_cmd
+}
+
+fn fill_aws_secrets_vars(mut cmd: &mut Command) -> &mut Command {
+    for env_key in [
+        "AWS_ACCESS_KEY_ID",
+        "AWS_SECRET_ACCESS_KEY",
+        "AWS_SESSION_TOKEN",
+    ] {
+        if let Ok(value) = std::env::var(env_key) {
+            cmd = cmd.env(env_key, value);
+        }
+    }
+    cmd
+}
+
+fn process_started<F>(
+    pid: Pid,
+    pid_file_to_check: Option<&Path>,
+    status_check: &F,
+) -> anyhow::Result<bool>
+where
+    F: Fn() -> anyhow::Result<bool>,
+{
+    match status_check() {
+        Ok(true) => match pid_file_to_check {
+            Some(pid_file_path) => {
+                if pid_file_path.exists() {
+                    let pid_in_file = read_pidfile(pid_file_path)?;
+                    Ok(pid_in_file == pid)
+                } else {
+                    Ok(false)
+                }
+            }
+            None => Ok(true),
+        },
+        Ok(false) => Ok(false),
+        Err(e) => anyhow::bail!("process failed to start: {e}"),
+    }
+}
+
+/// Read a PID file
+///
+/// We expect a file that contains a single integer.
+fn read_pidfile(pidfile: &Path) -> Result<Pid> {
+    let pid_str = fs::read_to_string(pidfile)
+        .with_context(|| format!("failed to read pidfile {pidfile:?}"))?;
+    let pid: i32 = pid_str
+        .parse()
+        .map_err(|_| anyhow!("failed to parse pidfile {pidfile:?}"))?;
+    if pid < 1 {
+        bail!("pidfile {pidfile:?} contained bad value '{pid}'");
+    }
+    Ok(Pid::from_raw(pid))
+}
+
+fn process_has_stopped(pid: Pid) -> anyhow::Result<bool> {
+    match kill(pid, None) {
+        // Process exists, keep waiting
+        Ok(_) => Ok(false),
+        // Process not found, we're done
+        Err(Errno::ESRCH) => Ok(true),
+        Err(err) => anyhow::bail!("Failed to send signal to process with pid {pid}: {err}"),
+    }
+}

control_plane/src/bin/neon_local.rs

@@ -6,11 +6,11 @@
 //! rely on `neon_local` to set up the environment for each test.
 //!
 use anyhow::{anyhow, bail, Context, Result};
-use clap::{App, AppSettings, Arg, ArgMatches};
+use clap::{value_parser, Arg, ArgAction, ArgMatches, Command};
 use control_plane::compute::ComputeControlPlane;
 use control_plane::local_env::{EtcdBroker, LocalEnv};
+use control_plane::pageserver::PageServerNode;
 use control_plane::safekeeper::SafekeeperNode;
-use control_plane::storage::PageServerNode;
 use control_plane::{etcd, local_env};
 use pageserver_api::models::TimelineInfo;
 use pageserver_api::{
@@ -85,212 +85,7 @@ struct TimelineTreeEl {
 //   * Providing CLI api to the pageserver
 //   * TODO: export/import to/from usual postgres
 fn main() -> Result<()> {
-    let branch_name_arg = Arg::new("branch-name")
-        .long("branch-name")
-        .takes_value(true)
-        .help("Name of the branch to be created or used as an alias for other services")
-        .required(false);
-
-    let pg_node_arg = Arg::new("node").help("Postgres node name").required(false);
-
-    let safekeeper_id_arg = Arg::new("id").help("safekeeper id").required(false);
-
-    let tenant_id_arg = Arg::new("tenant-id")
-        .long("tenant-id")
-        .help("Tenant id. Represented as a hexadecimal string 32 symbols length")
-        .takes_value(true)
-        .required(false);
-
-    let timeline_id_arg = Arg::new("timeline-id")
-        .long("timeline-id")
-        .help("Timeline id. Represented as a hexadecimal string 32 symbols length")
-        .takes_value(true)
-        .required(false);
-
-    let pg_version_arg = Arg::new("pg-version")
-        .long("pg-version")
-        .help("Postgres version to use for the initial tenant")
-        .required(false)
-        .takes_value(true)
-        .default_value(DEFAULT_PG_VERSION);
-
-    let port_arg = Arg::new("port")
-        .long("port")
-        .required(false)
-        .value_name("port");
-
-    let stop_mode_arg = Arg::new("stop-mode")
-        .short('m')
-        .takes_value(true)
-        .possible_values(&["fast", "immediate"])
-        .help("If 'immediate', don't flush repository data at shutdown")
-        .required(false)
-        .value_name("stop-mode");
-
-    let pageserver_config_args = Arg::new("pageserver-config-override")
-        .long("pageserver-config-override")
-        .takes_value(true)
-        .number_of_values(1)
-        .multiple_occurrences(true)
-        .help("Additional pageserver's configuration options or overrides, refer to pageserver's 'config-override' CLI parameter docs for more")
-        .required(false);
-
-    let lsn_arg = Arg::new("lsn")
-        .long("lsn")
-        .help("Specify Lsn on the timeline to start from. By default, end of the timeline would be used.")
-        .takes_value(true)
-        .required(false);
-
-    let matches = App::new("Neon CLI")
-        .setting(AppSettings::ArgRequiredElseHelp)
-        .version(GIT_VERSION)
-        .subcommand(
-            App::new("init")
-                .about("Initialize a new Neon repository")
-                .arg(pageserver_config_args.clone())
-                .arg(timeline_id_arg.clone().help("Use a specific timeline id when creating a tenant and its initial timeline"))
-                .arg(
-                    Arg::new("config")
-                        .long("config")
-                        .required(false)
-                        .value_name("config"),
-                )
-                .arg(pg_version_arg.clone())
-        )
-        .subcommand(
-            App::new("timeline")
-            .about("Manage timelines")
-            .subcommand(App::new("list")
-                .about("List all timelines, available to this pageserver")
-                .arg(tenant_id_arg.clone()))
-            .subcommand(App::new("branch")
-                .about("Create a new timeline, using another timeline as a base, copying its data")
-                .arg(tenant_id_arg.clone())
-                .arg(branch_name_arg.clone())
-                .arg(Arg::new("ancestor-branch-name").long("ancestor-branch-name").takes_value(true)
-                    .help("Use last Lsn of another timeline (and its data) as base when creating the new timeline. The timeline gets resolved by its branch name.").required(false))
-                .arg(Arg::new("ancestor-start-lsn").long("ancestor-start-lsn").takes_value(true)
-                    .help("When using another timeline as base, use a specific Lsn in it instead of the latest one").required(false)))
-            .subcommand(App::new("create")
-                .about("Create a new blank timeline")
-                .arg(tenant_id_arg.clone())
-                .arg(branch_name_arg.clone())
-                .arg(pg_version_arg.clone())
-            )
-            .subcommand(App::new("import")
-                .about("Import timeline from basebackup directory")
-                .arg(tenant_id_arg.clone())
-                .arg(timeline_id_arg.clone())
-                .arg(Arg::new("node-name").long("node-name").takes_value(true)
-                    .help("Name to assign to the imported timeline"))
-                .arg(Arg::new("base-tarfile").long("base-tarfile").takes_value(true)
-                    .help("Basebackup tarfile to import"))
-                .arg(Arg::new("base-lsn").long("base-lsn").takes_value(true)
-                    .help("Lsn the basebackup starts at"))
-                .arg(Arg::new("wal-tarfile").long("wal-tarfile").takes_value(true)
-                    .help("Wal to add after base"))
-                .arg(Arg::new("end-lsn").long("end-lsn").takes_value(true)
-                    .help("Lsn the basebackup ends at"))
-                .arg(pg_version_arg.clone())
-            )
-        ).subcommand(
-            App::new("tenant")
-            .setting(AppSettings::ArgRequiredElseHelp)
-            .about("Manage tenants")
-            .subcommand(App::new("list"))
-            .subcommand(App::new("create")
-                .arg(tenant_id_arg.clone())
-                .arg(timeline_id_arg.clone().help("Use a specific timeline id when creating a tenant and its initial timeline"))
-                .arg(Arg::new("config").short('c').takes_value(true).multiple_occurrences(true).required(false))
-                .arg(pg_version_arg.clone())
-                )
-            .subcommand(App::new("config")
-                .arg(tenant_id_arg.clone())
-                .arg(Arg::new("config").short('c').takes_value(true).multiple_occurrences(true).required(false))
-                )
-        )
-        .subcommand(
-            App::new("pageserver")
-                .setting(AppSettings::ArgRequiredElseHelp)
-                .about("Manage pageserver")
-                .subcommand(App::new("status"))
-                .subcommand(App::new("start").about("Start local pageserver").arg(pageserver_config_args.clone()))
-                .subcommand(App::new("stop").about("Stop local pageserver")
-                            .arg(stop_mode_arg.clone()))
-                .subcommand(App::new("restart").about("Restart local pageserver").arg(pageserver_config_args.clone()))
-        )
-        .subcommand(
-            App::new("safekeeper")
-                .setting(AppSettings::ArgRequiredElseHelp)
-                .about("Manage safekeepers")
-                .subcommand(App::new("start")
-                            .about("Start local safekeeper")
-                            .arg(safekeeper_id_arg.clone())
-                )
-                .subcommand(App::new("stop")
-                            .about("Stop local safekeeper")
-                            .arg(safekeeper_id_arg.clone())
-                            .arg(stop_mode_arg.clone())
-                )
-                .subcommand(App::new("restart")
-                            .about("Restart local safekeeper")
-                            .arg(safekeeper_id_arg.clone())
-                            .arg(stop_mode_arg.clone())
-                )
-        )
-        .subcommand(
-            App::new("pg")
-                .setting(AppSettings::ArgRequiredElseHelp)
-                .about("Manage postgres instances")
-                .subcommand(App::new("list").arg(tenant_id_arg.clone()))
-                .subcommand(App::new("create")
-                    .about("Create a postgres compute node")
-                    .arg(pg_node_arg.clone())
-                    .arg(branch_name_arg.clone())
-                    .arg(tenant_id_arg.clone())
-                    .arg(lsn_arg.clone())
-                    .arg(port_arg.clone())
-                    .arg(
-                        Arg::new("config-only")
-                            .help("Don't do basebackup, create compute node with only config files")
-                            .long("config-only")
-                            .required(false))
-                    .arg(pg_version_arg.clone())
-                )
-                .subcommand(App::new("start")
-                    .about("Start a postgres compute node.\n This command actually creates new node from scratch, but preserves existing config files")
-                    .arg(pg_node_arg.clone())
-                    .arg(tenant_id_arg.clone())
-                    .arg(branch_name_arg.clone())
-                    .arg(timeline_id_arg.clone())
-                    .arg(lsn_arg.clone())
-                    .arg(port_arg.clone())
-                    .arg(pg_version_arg.clone())
-                )
-                .subcommand(
-                    App::new("stop")
-                    .arg(pg_node_arg.clone())
-                    .arg(tenant_id_arg.clone())
-                    .arg(
-                        Arg::new("destroy")
-                            .help("Also delete data directory (now optional, should be default in future)")
-                            .long("destroy")
-                            .required(false)
-                    )
-                    )
-
-        )
-        .subcommand(
-            App::new("start")
-                .about("Start page server and safekeepers")
-                .arg(pageserver_config_args)
-        )
-        .subcommand(
-            App::new("stop")
-                .about("Stop page server and safekeepers")
-                .arg(stop_mode_arg.clone())
-        )
-        .get_matches();
+    let matches = cli().get_matches();
 
     let (sub_name, sub_args) = match matches.subcommand() {
         Some(subcommand_data) => subcommand_data,
@@ -358,9 +153,7 @@ fn print_timelines_tree(
 
     // Memorize all direct children of each timeline.
     for timeline in timelines.iter() {
-        if let Some(ancestor_timeline_id) =
-            timeline.local.as_ref().and_then(|l| l.ancestor_timeline_id)
-        {
+        if let Some(ancestor_timeline_id) = timeline.ancestor_timeline_id {
             timelines_hash
                 .get_mut(&ancestor_timeline_id)
                 .context("missing timeline info in the HashMap")?
@@ -371,13 +164,7 @@ fn print_timelines_tree(
 
     for timeline in timelines_hash.values() {
         // Start with root local timelines (no ancestors) first.
-        if timeline
-            .info
-            .local
-            .as_ref()
-            .and_then(|l| l.ancestor_timeline_id)
-            .is_none()
-        {
+        if timeline.info.ancestor_timeline_id.is_none() {
             print_timeline(0, &Vec::from([true]), timeline, &timelines_hash)?;
         }
     }
@@ -394,17 +181,8 @@ fn print_timeline(
     timeline: &TimelineTreeEl,
     timelines: &HashMap<TimelineId, TimelineTreeEl>,
 ) -> Result<()> {
-    let local_remote = match (timeline.info.local.as_ref(), timeline.info.remote.as_ref()) {
-        (None, None) => unreachable!("in this case no info for a timeline is found"),
-        (None, Some(_)) => "(R)",
-        (Some(_), None) => "(L)",
-        (Some(_), Some(_)) => "(L+R)",
-    };
-    // Draw main padding
-    print!("{} ", local_remote);
-
     if nesting_level > 0 {
-        let ancestor_lsn = match timeline.info.local.as_ref().and_then(|i| i.ancestor_lsn) {
+        let ancestor_lsn = match timeline.info.ancestor_lsn {
             Some(lsn) => lsn.to_string(),
             None => "Unknown Lsn".to_string(),
         };
@@ -492,16 +270,16 @@ fn get_tenant_id(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> anyhow::R
 
 fn parse_tenant_id(sub_match: &ArgMatches) -> anyhow::Result<Option<TenantId>> {
     sub_match
-        .value_of("tenant-id")
-        .map(TenantId::from_str)
+        .get_one::<String>("tenant-id")
+        .map(|tenant_id| TenantId::from_str(tenant_id))
         .transpose()
         .context("Failed to parse tenant id from the argument string")
 }
 
 fn parse_timeline_id(sub_match: &ArgMatches) -> anyhow::Result<Option<TimelineId>> {
     sub_match
-        .value_of("timeline-id")
-        .map(TimelineId::from_str)
+        .get_one::<String>("timeline-id")
+        .map(|timeline_id| TimelineId::from_str(timeline_id))
         .transpose()
         .context("Failed to parse timeline id from the argument string")
 }
@@ -510,19 +288,22 @@ fn handle_init(init_match: &ArgMatches) -> anyhow::Result<LocalEnv> {
     let initial_timeline_id_arg = parse_timeline_id(init_match)?;
 
     // Create config file
-    let toml_file: String = if let Some(config_path) = init_match.value_of("config") {
+    let toml_file: String = if let Some(config_path) = init_match.get_one::<PathBuf>("config") {
         // load and parse the file
-        std::fs::read_to_string(std::path::Path::new(config_path))
-            .with_context(|| format!("Could not read configuration file '{config_path}'"))?
+        std::fs::read_to_string(config_path).with_context(|| {
+            format!(
+                "Could not read configuration file '{}'",
+                config_path.display()
+            )
+        })?
     } else {
         // Built-in default config
         default_conf(&EtcdBroker::locate_etcd()?)
     };
 
     let pg_version = init_match
-        .value_of("pg-version")
-        .unwrap()
-        .parse::<u32>()
+        .get_one::<u32>("pg-version")
+        .copied()
         .context("Failed to parse postgres version from the argument string")?;
 
     let mut env =
@@ -558,9 +339,10 @@ fn handle_init(init_match: &ArgMatches) -> anyhow::Result<LocalEnv> {
 
 fn pageserver_config_overrides(init_match: &ArgMatches) -> Vec<&str> {
     init_match
-        .values_of("pageserver-config-override")
+        .get_many::<String>("pageserver-config-override")
         .into_iter()
         .flatten()
+        .map(|s| s.as_str())
         .collect()
 }
 
@@ -575,7 +357,7 @@ fn handle_tenant(tenant_match: &ArgMatches, env: &mut local_env::LocalEnv) -> an
         Some(("create", create_match)) => {
             let initial_tenant_id = parse_tenant_id(create_match)?;
             let tenant_conf: HashMap<_, _> = create_match
-                .values_of("config")
+                .get_many::<String>("config")
                 .map(|vals| vals.flat_map(|c| c.split_once(':')).collect())
                 .unwrap_or_default();
             let new_tenant_id = pageserver.tenant_create(initial_tenant_id, tenant_conf)?;
@@ -584,9 +366,8 @@ fn handle_tenant(tenant_match: &ArgMatches, env: &mut local_env::LocalEnv) -> an
             // Create an initial timeline for the new tenant
             let new_timeline_id = parse_timeline_id(create_match)?;
             let pg_version = create_match
-                .value_of("pg-version")
-                .unwrap()
-                .parse::<u32>()
+                .get_one::<u32>("pg-version")
+                .copied()
                 .context("Failed to parse postgres version from the argument string")?;
 
             let timeline_info = pageserver.timeline_create(
@@ -597,10 +378,7 @@ fn handle_tenant(tenant_match: &ArgMatches, env: &mut local_env::LocalEnv) -> an
                 Some(pg_version),
             )?;
             let new_timeline_id = timeline_info.timeline_id;
-            let last_record_lsn = timeline_info
-                .local
-                .context(format!("Failed to get last record LSN: no local timeline info for timeline {new_timeline_id}"))?
-                .last_record_lsn;
+            let last_record_lsn = timeline_info.last_record_lsn;
 
             env.register_branch_mapping(
                 DEFAULT_BRANCH_NAME.to_string(),
@@ -615,7 +393,7 @@ fn handle_tenant(tenant_match: &ArgMatches, env: &mut local_env::LocalEnv) -> an
         Some(("config", create_match)) => {
             let tenant_id = get_tenant_id(create_match, env)?;
             let tenant_conf: HashMap<_, _> = create_match
-                .values_of("config")
+                .get_many::<String>("config")
                 .map(|vals| vals.flat_map(|c| c.split_once(':')).collect())
                 .unwrap_or_default();
 
@@ -642,23 +420,19 @@ fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::LocalEnv) -
         Some(("create", create_match)) => {
             let tenant_id = get_tenant_id(create_match, env)?;
             let new_branch_name = create_match
-                .value_of("branch-name")
+                .get_one::<String>("branch-name")
                 .ok_or_else(|| anyhow!("No branch name provided"))?;
 
             let pg_version = create_match
-                .value_of("pg-version")
-                .unwrap()
-                .parse::<u32>()
+                .get_one::<u32>("pg-version")
+                .copied()
                 .context("Failed to parse postgres version from the argument string")?;
 
             let timeline_info =
                 pageserver.timeline_create(tenant_id, None, None, None, Some(pg_version))?;
             let new_timeline_id = timeline_info.timeline_id;
 
-            let last_record_lsn = timeline_info
-                .local
-                .expect("no local timeline info")
-                .last_record_lsn;
+            let last_record_lsn = timeline_info.last_record_lsn;
             env.register_branch_mapping(new_branch_name.to_string(), tenant_id, new_timeline_id)?;
 
             println!(
@@ -670,35 +444,32 @@ fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::LocalEnv) -
             let tenant_id = get_tenant_id(import_match, env)?;
             let timeline_id = parse_timeline_id(import_match)?.expect("No timeline id provided");
             let name = import_match
-                .value_of("node-name")
+                .get_one::<String>("node-name")
                 .ok_or_else(|| anyhow!("No node name provided"))?;
 
             // Parse base inputs
             let base_tarfile = import_match
-                .value_of("base-tarfile")
-                .map(|s| PathBuf::from_str(s).unwrap())
-                .ok_or_else(|| anyhow!("No base-tarfile provided"))?;
+                .get_one::<PathBuf>("base-tarfile")
+                .ok_or_else(|| anyhow!("No base-tarfile provided"))?
+                .to_owned();
             let base_lsn = Lsn::from_str(
                 import_match
-                    .value_of("base-lsn")
+                    .get_one::<String>("base-lsn")
                     .ok_or_else(|| anyhow!("No base-lsn provided"))?,
             )?;
             let base = (base_lsn, base_tarfile);
 
             // Parse pg_wal inputs
-            let wal_tarfile = import_match
-                .value_of("wal-tarfile")
-                .map(|s| PathBuf::from_str(s).unwrap());
+            let wal_tarfile = import_match.get_one::<PathBuf>("wal-tarfile").cloned();
             let end_lsn = import_match
-                .value_of("end-lsn")
+                .get_one::<String>("end-lsn")
                 .map(|s| Lsn::from_str(s).unwrap());
             // TODO validate both or none are provided
             let pg_wal = end_lsn.zip(wal_tarfile);
 
             let pg_version = import_match
-                .value_of("pg-version")
-                .unwrap()
-                .parse::<u32>()
+                .get_one::<u32>("pg-version")
+                .copied()
                 .context("Failed to parse postgres version from the argument string")?;
 
             let mut cplane = ComputeControlPlane::load(env.clone())?;
@@ -713,10 +484,11 @@ fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::LocalEnv) -
         Some(("branch", branch_match)) => {
             let tenant_id = get_tenant_id(branch_match, env)?;
             let new_branch_name = branch_match
-                .value_of("branch-name")
+                .get_one::<String>("branch-name")
                 .ok_or_else(|| anyhow!("No branch name provided"))?;
             let ancestor_branch_name = branch_match
-                .value_of("ancestor-branch-name")
+                .get_one::<String>("ancestor-branch-name")
+                .map(|s| s.as_str())
                 .unwrap_or(DEFAULT_BRANCH_NAME);
             let ancestor_timeline_id = env
                 .get_branch_timeline_id(ancestor_branch_name, tenant_id)
@@ -725,8 +497,8 @@ fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::LocalEnv) -
                 })?;
 
             let start_lsn = branch_match
-                .value_of("ancestor-start-lsn")
-                .map(Lsn::from_str)
+                .get_one::<String>("ancestor-start-lsn")
+                .map(|lsn_str| Lsn::from_str(lsn_str))
                 .transpose()
                 .context("Failed to parse ancestor start Lsn from the request")?;
             let timeline_info = pageserver.timeline_create(
@@ -738,10 +510,7 @@ fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::LocalEnv) -
             )?;
             let new_timeline_id = timeline_info.timeline_id;
 
-            let last_record_lsn = timeline_info
-                .local
-                .expect("no local timeline info")
-                .last_record_lsn;
+            let last_record_lsn = timeline_info.last_record_lsn;
 
             env.register_branch_mapping(new_branch_name.to_string(), tenant_id, new_timeline_id)?;
 
@@ -801,7 +570,7 @@ fn handle_pg(pg_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
                         // Use the LSN at the end of the timeline.
                         timeline_infos
                             .get(&node.timeline_id)
-                            .and_then(|bi| bi.local.as_ref().map(|l| l.last_record_lsn.to_string()))
+                            .map(|bi| bi.last_record_lsn.to_string())
                             .unwrap_or_else(|| "?".to_string())
                     }
                     Some(lsn) => {
@@ -830,45 +599,39 @@ fn handle_pg(pg_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
         }
         "create" => {
             let branch_name = sub_args
-                .value_of("branch-name")
+                .get_one::<String>("branch-name")
+                .map(|s| s.as_str())
                 .unwrap_or(DEFAULT_BRANCH_NAME);
             let node_name = sub_args
-                .value_of("node")
-                .map(ToString::to_string)
-                .unwrap_or_else(|| format!("{}_node", branch_name));
+                .get_one::<String>("node")
+                .map(|node_name| node_name.to_string())
+                .unwrap_or_else(|| format!("{branch_name}_node"));
 
             let lsn = sub_args
-                .value_of("lsn")
-                .map(Lsn::from_str)
+                .get_one::<String>("lsn")
+                .map(|lsn_str| Lsn::from_str(lsn_str))
                 .transpose()
                 .context("Failed to parse Lsn from the request")?;
             let timeline_id = env
                 .get_branch_timeline_id(branch_name, tenant_id)
-                .ok_or_else(|| anyhow!("Found no timeline id for branch name '{}'", branch_name))?;
+                .ok_or_else(|| anyhow!("Found no timeline id for branch name '{branch_name}'"))?;
 
-            let port: Option<u16> = match sub_args.value_of("port") {
-                Some(p) => Some(p.parse()?),
-                None => None,
-            };
+            let port: Option<u16> = sub_args.get_one::<u16>("port").copied();
 
             let pg_version = sub_args
-                .value_of("pg-version")
-                .unwrap()
-                .parse::<u32>()
+                .get_one::<u32>("pg-version")
+                .copied()
                 .context("Failed to parse postgres version from the argument string")?;
 
             cplane.new_node(tenant_id, &node_name, timeline_id, lsn, port, pg_version)?;
         }
         "start" => {
-            let port: Option<u16> = match sub_args.value_of("port") {
-                Some(p) => Some(p.parse()?),
-                None => None,
-            };
+            let port: Option<u16> = sub_args.get_one::<u16>("port").copied();
             let node_name = sub_args
-                .value_of("node")
+                .get_one::<String>("node")
                 .ok_or_else(|| anyhow!("No node name was provided to start"))?;
 
-            let node = cplane.nodes.get(&(tenant_id, node_name.to_owned()));
+            let node = cplane.nodes.get(&(tenant_id, node_name.to_string()));
 
             let auth_token = if matches!(env.pageserver.auth_type, AuthType::NeonJWT) {
                 let claims = Claims::new(Some(tenant_id), Scope::Tenant);
@@ -879,36 +642,33 @@ fn handle_pg(pg_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
             };
 
             if let Some(node) = node {
-                println!("Starting existing postgres {}...", node_name);
+                println!("Starting existing postgres {node_name}...");
                 node.start(&auth_token)?;
             } else {
                 let branch_name = sub_args
-                    .value_of("branch-name")
+                    .get_one::<String>("branch-name")
+                    .map(|s| s.as_str())
                     .unwrap_or(DEFAULT_BRANCH_NAME);
                 let timeline_id = env
                     .get_branch_timeline_id(branch_name, tenant_id)
                     .ok_or_else(|| {
-                        anyhow!("Found no timeline id for branch name '{}'", branch_name)
+                        anyhow!("Found no timeline id for branch name '{branch_name}'")
                     })?;
                 let lsn = sub_args
-                    .value_of("lsn")
-                    .map(Lsn::from_str)
+                    .get_one::<String>("lsn")
+                    .map(|lsn_str| Lsn::from_str(lsn_str))
                     .transpose()
                     .context("Failed to parse Lsn from the request")?;
                 let pg_version = sub_args
-                    .value_of("pg-version")
-                    .unwrap()
-                    .parse::<u32>()
-                    .context("Failed to parse postgres version from the argument string")?;
+                    .get_one::<u32>("pg-version")
+                    .copied()
+                    .context("Failed to `pg-version` from the argument string")?;
                 // when used with custom port this results in non obvious behaviour
                 // port is remembered from first start command, i e
                 // start --port X
                 // stop
                 // start <-- will also use port X even without explicit port argument
-                println!(
-                    "Starting new postgres (v{}) {} on timeline {} ...",
-                    pg_version, node_name, timeline_id
-                );
+                println!("Starting new postgres (v{pg_version}) {node_name} on timeline {timeline_id} ...");
 
                 let node =
                     cplane.new_node(tenant_id, node_name, timeline_id, lsn, port, pg_version)?;
@@ -917,18 +677,18 @@ fn handle_pg(pg_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
         }
         "stop" => {
             let node_name = sub_args
-                .value_of("node")
+                .get_one::<String>("node")
                 .ok_or_else(|| anyhow!("No node name was provided to stop"))?;
-            let destroy = sub_args.is_present("destroy");
+            let destroy = sub_args.get_flag("destroy");
 
             let node = cplane
                 .nodes
-                .get(&(tenant_id, node_name.to_owned()))
-                .with_context(|| format!("postgres {} is not found", node_name))?;
+                .get(&(tenant_id, node_name.to_string()))
+                .with_context(|| format!("postgres {node_name} is not found"))?;
             node.stop(destroy)?;
         }
 
-        _ => bail!("Unexpected pg subcommand '{}'", sub_name),
+        _ => bail!("Unexpected pg subcommand '{sub_name}'"),
     }
 
     Ok(())
@@ -946,7 +706,10 @@ fn handle_pageserver(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Resul
         }
 
         Some(("stop", stop_match)) => {
-            let immediate = stop_match.value_of("stop-mode") == Some("immediate");
+            let immediate = stop_match
+                .get_one::<String>("stop-mode")
+                .map(|s| s.as_str())
+                == Some("immediate");
 
             if let Err(e) = pageserver.stop(immediate) {
                 eprintln!("pageserver stop failed: {}", e);
@@ -996,7 +759,7 @@ fn handle_safekeeper(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Resul
     };
 
     // All the commands take an optional safekeeper name argument
-    let sk_id = if let Some(id_str) = sub_args.value_of("id") {
+    let sk_id = if let Some(id_str) = sub_args.get_one::<String>("id") {
         NodeId(id_str.parse().context("while parsing safekeeper id")?)
     } else {
         DEFAULT_SAFEKEEPER_ID
@@ -1012,7 +775,8 @@ fn handle_safekeeper(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Resul
         }
 
         "stop" => {
-            let immediate = sub_args.value_of("stop-mode") == Some("immediate");
+            let immediate =
+                sub_args.get_one::<String>("stop-mode").map(|s| s.as_str()) == Some("immediate");
 
             if let Err(e) = safekeeper.stop(immediate) {
                 eprintln!("safekeeper stop failed: {}", e);
@@ -1021,7 +785,8 @@ fn handle_safekeeper(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Resul
         }
 
         "restart" => {
-            let immediate = sub_args.value_of("stop-mode") == Some("immediate");
+            let immediate =
+                sub_args.get_one::<String>("stop-mode").map(|s| s.as_str()) == Some("immediate");
 
             if let Err(e) = safekeeper.stop(immediate) {
                 eprintln!("safekeeper stop failed: {}", e);
@@ -1065,7 +830,8 @@ fn handle_start_all(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> anyhow
 }
 
 fn handle_stop_all(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
-    let immediate = sub_match.value_of("stop-mode") == Some("immediate");
+    let immediate =
+        sub_match.get_one::<String>("stop-mode").map(|s| s.as_str()) == Some("immediate");
 
     let pageserver = PageServerNode::from_env(env);
 
@@ -1098,3 +864,219 @@ fn try_stop_etcd_process(env: &local_env::LocalEnv) {
         eprintln!("etcd stop failed: {e}");
     }
 }
+
+fn cli() -> Command {
+    let branch_name_arg = Arg::new("branch-name")
+        .long("branch-name")
+        .help("Name of the branch to be created or used as an alias for other services")
+        .required(false);
+
+    let pg_node_arg = Arg::new("node").help("Postgres node name").required(false);
+
+    let safekeeper_id_arg = Arg::new("id").help("safekeeper id").required(false);
+
+    let tenant_id_arg = Arg::new("tenant-id")
+        .long("tenant-id")
+        .help("Tenant id. Represented as a hexadecimal string 32 symbols length")
+        .required(false);
+
+    let timeline_id_arg = Arg::new("timeline-id")
+        .long("timeline-id")
+        .help("Timeline id. Represented as a hexadecimal string 32 symbols length")
+        .required(false);
+
+    let pg_version_arg = Arg::new("pg-version")
+        .long("pg-version")
+        .help("Postgres version to use for the initial tenant")
+        .required(false)
+        .value_parser(value_parser!(u32))
+        .default_value(DEFAULT_PG_VERSION);
+
+    let port_arg = Arg::new("port")
+        .long("port")
+        .required(false)
+        .value_parser(value_parser!(u16))
+        .value_name("port");
+
+    let stop_mode_arg = Arg::new("stop-mode")
+        .short('m')
+        .value_parser(["fast", "immediate"])
+        .help("If 'immediate', don't flush repository data at shutdown")
+        .required(false)
+        .value_name("stop-mode");
+
+    let pageserver_config_args = Arg::new("pageserver-config-override")
+        .long("pageserver-config-override")
+        .num_args(1)
+        .action(ArgAction::Append)
+        .help("Additional pageserver's configuration options or overrides, refer to pageserver's 'config-override' CLI parameter docs for more")
+        .required(false);
+
+    let lsn_arg = Arg::new("lsn")
+        .long("lsn")
+        .help("Specify Lsn on the timeline to start from. By default, end of the timeline would be used.")
+        .required(false);
+
+    Command::new("Neon CLI")
+        .arg_required_else_help(true)
+        .version(GIT_VERSION)
+        .subcommand(
+            Command::new("init")
+                .about("Initialize a new Neon repository")
+                .arg(pageserver_config_args.clone())
+                .arg(timeline_id_arg.clone().help("Use a specific timeline id when creating a tenant and its initial timeline"))
+                .arg(
+                    Arg::new("config")
+                        .long("config")
+                        .required(false)
+                        .value_parser(value_parser!(PathBuf))
+                        .value_name("config"),
+                )
+                .arg(pg_version_arg.clone())
+        )
+        .subcommand(
+            Command::new("timeline")
+            .about("Manage timelines")
+            .subcommand(Command::new("list")
+                .about("List all timelines, available to this pageserver")
+                .arg(tenant_id_arg.clone()))
+            .subcommand(Command::new("branch")
+                .about("Create a new timeline, using another timeline as a base, copying its data")
+                .arg(tenant_id_arg.clone())
+                .arg(branch_name_arg.clone())
+                .arg(Arg::new("ancestor-branch-name").long("ancestor-branch-name")
+                    .help("Use last Lsn of another timeline (and its data) as base when creating the new timeline. The timeline gets resolved by its branch name.").required(false))
+                .arg(Arg::new("ancestor-start-lsn").long("ancestor-start-lsn")
+                    .help("When using another timeline as base, use a specific Lsn in it instead of the latest one").required(false)))
+            .subcommand(Command::new("create")
+                .about("Create a new blank timeline")
+                .arg(tenant_id_arg.clone())
+                .arg(branch_name_arg.clone())
+                .arg(pg_version_arg.clone())
+            )
+            .subcommand(Command::new("import")
+                .about("Import timeline from basebackup directory")
+                .arg(tenant_id_arg.clone())
+                .arg(timeline_id_arg.clone())
+                .arg(Arg::new("node-name").long("node-name")
+                    .help("Name to assign to the imported timeline"))
+                .arg(Arg::new("base-tarfile")
+                    .long("base-tarfile")
+                    .value_parser(value_parser!(PathBuf))
+                    .help("Basebackup tarfile to import")
+                )
+                .arg(Arg::new("base-lsn").long("base-lsn")
+                    .help("Lsn the basebackup starts at"))
+                .arg(Arg::new("wal-tarfile")
+                    .long("wal-tarfile")
+                    .value_parser(value_parser!(PathBuf))
+                    .help("Wal to add after base")
+                )
+                .arg(Arg::new("end-lsn").long("end-lsn")
+                    .help("Lsn the basebackup ends at"))
+                .arg(pg_version_arg.clone())
+            )
+        ).subcommand(
+            Command::new("tenant")
+            .arg_required_else_help(true)
+            .about("Manage tenants")
+            .subcommand(Command::new("list"))
+            .subcommand(Command::new("create")
+                .arg(tenant_id_arg.clone())
+                .arg(timeline_id_arg.clone().help("Use a specific timeline id when creating a tenant and its initial timeline"))
+                .arg(Arg::new("config").short('c').num_args(1).action(ArgAction::Append).required(false))
+                .arg(pg_version_arg.clone())
+                )
+            .subcommand(Command::new("config")
+                .arg(tenant_id_arg.clone())
+                .arg(Arg::new("config").short('c').num_args(1).action(ArgAction::Append).required(false))
+                )
+        )
+        .subcommand(
+            Command::new("pageserver")
+                .arg_required_else_help(true)
+                .about("Manage pageserver")
+                .subcommand(Command::new("status"))
+                .subcommand(Command::new("start").about("Start local pageserver").arg(pageserver_config_args.clone()))
+                .subcommand(Command::new("stop").about("Stop local pageserver")
+                            .arg(stop_mode_arg.clone()))
+                .subcommand(Command::new("restart").about("Restart local pageserver").arg(pageserver_config_args.clone()))
+        )
+        .subcommand(
+            Command::new("safekeeper")
+                .arg_required_else_help(true)
+                .about("Manage safekeepers")
+                .subcommand(Command::new("start")
+                            .about("Start local safekeeper")
+                            .arg(safekeeper_id_arg.clone())
+                )
+                .subcommand(Command::new("stop")
+                            .about("Stop local safekeeper")
+                            .arg(safekeeper_id_arg.clone())
+                            .arg(stop_mode_arg.clone())
+                )
+                .subcommand(Command::new("restart")
+                            .about("Restart local safekeeper")
+                            .arg(safekeeper_id_arg)
+                            .arg(stop_mode_arg.clone())
+                )
+        )
+        .subcommand(
+            Command::new("pg")
+                .arg_required_else_help(true)
+                .about("Manage postgres instances")
+                .subcommand(Command::new("list").arg(tenant_id_arg.clone()))
+                .subcommand(Command::new("create")
+                    .about("Create a postgres compute node")
+                    .arg(pg_node_arg.clone())
+                    .arg(branch_name_arg.clone())
+                    .arg(tenant_id_arg.clone())
+                    .arg(lsn_arg.clone())
+                    .arg(port_arg.clone())
+                    .arg(
+                        Arg::new("config-only")
+                            .help("Don't do basebackup, create compute node with only config files")
+                            .long("config-only")
+                            .required(false))
+                    .arg(pg_version_arg.clone())
+                )
+                .subcommand(Command::new("start")
+                    .about("Start a postgres compute node.\n This command actually creates new node from scratch, but preserves existing config files")
+                    .arg(pg_node_arg.clone())
+                    .arg(tenant_id_arg.clone())
+                    .arg(branch_name_arg)
+                    .arg(timeline_id_arg)
+                    .arg(lsn_arg)
+                    .arg(port_arg)
+                    .arg(pg_version_arg)
+                )
+                .subcommand(
+                    Command::new("stop")
+                    .arg(pg_node_arg)
+                    .arg(tenant_id_arg)
+                    .arg(
+                        Arg::new("destroy")
+                            .help("Also delete data directory (now optional, should be default in future)")
+                            .long("destroy")
+                            .action(ArgAction::SetTrue)
+                            .required(false)
+                        )
+                )
+
+        )
+        .subcommand(
+            Command::new("start")
+                .about("Start page server and safekeepers")
+                .arg(pageserver_config_args)
+        )
+        .subcommand(
+            Command::new("stop")
+                .about("Stop page server and safekeepers")
+                .arg(stop_mode_arg)
+        )
+}
+
+#[test]
+fn verify_cli() {
+    cli().debug_assert();
+}

control_plane/src/compute.rs

@@ -12,15 +12,14 @@ use std::time::Duration;
 
 use anyhow::{Context, Result};
 use utils::{
-    connstring::connection_host_port,
     id::{TenantId, TimelineId},
     lsn::Lsn,
     postgres_backend::AuthType,
 };
 
 use crate::local_env::{LocalEnv, DEFAULT_PG_VERSION};
+use crate::pageserver::PageServerNode;
 use crate::postgresql_conf::PostgresConf;
-use crate::storage::PageServerNode;
 
 //
 // ComputeControlPlane
@@ -183,18 +182,18 @@ impl PostgresNode {
     }
 
     fn sync_safekeepers(&self, auth_token: &Option<String>, pg_version: u32) -> Result<Lsn> {
-        let pg_path = self.env.pg_bin_dir(pg_version).join("postgres");
+        let pg_path = self.env.pg_bin_dir(pg_version)?.join("postgres");
         let mut cmd = Command::new(&pg_path);
 
         cmd.arg("--sync-safekeepers")
             .env_clear()
             .env(
                 "LD_LIBRARY_PATH",
-                self.env.pg_lib_dir(pg_version).to_str().unwrap(),
+                self.env.pg_lib_dir(pg_version)?.to_str().unwrap(),
             )
             .env(
                 "DYLD_LIBRARY_PATH",
-                self.env.pg_lib_dir(pg_version).to_str().unwrap(),
+                self.env.pg_lib_dir(pg_version)?.to_str().unwrap(),
             )
             .env("PGDATA", self.pgdata().to_str().unwrap())
             .stdout(Stdio::piped())
@@ -282,9 +281,7 @@ impl PostgresNode {
     fn setup_pg_conf(&self, auth_type: AuthType) -> Result<()> {
         let mut conf = PostgresConf::new();
         conf.append("max_wal_senders", "10");
-        // wal_log_hints is mandatory when running against pageserver (see gh issue#192)
-        // TODO: is it possible to check wal_log_hints at pageserver side via XLOG_PARAMETER_CHANGE?
-        conf.append("wal_log_hints", "on");
+        conf.append("wal_log_hints", "off");
         conf.append("max_replication_slots", "10");
         conf.append("hot_standby", "on");
         conf.append("shared_buffers", "1MB");
@@ -302,7 +299,8 @@ impl PostgresNode {
 
         // Configure the node to fetch pages from pageserver
         let pageserver_connstr = {
-            let (host, port) = connection_host_port(&self.pageserver.pg_connection_config);
+            let config = &self.pageserver.pg_connection_config;
+            let (host, port) = (config.host(), config.port());
 
             // Set up authentication
             //
@@ -345,7 +343,7 @@ impl PostgresNode {
         //   To be able to restore database in case of pageserver node crash, safekeeper should not
         //   remove WAL beyond this point. Too large lag can cause space exhaustion in safekeepers
         //   (if they are not able to upload WAL to S3).
-        conf.append("max_replication_write_lag", "500MB");
+        conf.append("max_replication_write_lag", "15MB");
         conf.append("max_replication_flush_lag", "10GB");
 
         if !self.env.safekeepers.is_empty() {
@@ -422,7 +420,7 @@ impl PostgresNode {
     }
 
     fn pg_ctl(&self, args: &[&str], auth_token: &Option<String>) -> Result<()> {
-        let pg_ctl_path = self.env.pg_bin_dir(self.pg_version).join("pg_ctl");
+        let pg_ctl_path = self.env.pg_bin_dir(self.pg_version)?.join("pg_ctl");
         let mut cmd = Command::new(pg_ctl_path);
         cmd.args(
             [
@@ -440,11 +438,11 @@ impl PostgresNode {
         .env_clear()
         .env(
             "LD_LIBRARY_PATH",
-            self.env.pg_lib_dir(self.pg_version).to_str().unwrap(),
+            self.env.pg_lib_dir(self.pg_version)?.to_str().unwrap(),
         )
         .env(
             "DYLD_LIBRARY_PATH",
-            self.env.pg_lib_dir(self.pg_version).to_str().unwrap(),
+            self.env.pg_lib_dir(self.pg_version)?.to_str().unwrap(),
         );
         if let Some(token) = auth_token {
             cmd.env("ZENITH_AUTH_TOKEN", token);

control_plane/src/connection.rs

@@ -0,0 +1,57 @@
+use url::Url;
+
+#[derive(Debug)]
+pub struct PgConnectionConfig {
+    url: Url,
+}
+
+impl PgConnectionConfig {
+    pub fn host(&self) -> &str {
+        self.url.host_str().expect("BUG: no host")
+    }
+
+    pub fn port(&self) -> u16 {
+        self.url.port().expect("BUG: no port")
+    }
+
+    /// Return a `<host>:<port>` string.
+    pub fn raw_address(&self) -> String {
+        format!("{}:{}", self.host(), self.port())
+    }
+
+    /// Connect using postgres protocol with TLS disabled.
+    pub fn connect_no_tls(&self) -> Result<postgres::Client, postgres::Error> {
+        postgres::Client::connect(self.url.as_str(), postgres::NoTls)
+    }
+}
+
+impl std::str::FromStr for PgConnectionConfig {
+    type Err = anyhow::Error;
+
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        let mut url: Url = s.parse()?;
+
+        match url.scheme() {
+            "postgres" | "postgresql" => {}
+            other => anyhow::bail!("invalid scheme: {other}"),
+        }
+
+        // It's not a valid connection url if host is unavailable.
+        if url.host().is_none() {
+            anyhow::bail!(url::ParseError::EmptyHost);
+        }
+
+        // E.g. `postgres:bar`.
+        if url.cannot_be_a_base() {
+            anyhow::bail!("URL cannot be a base");
+        }
+
+        // Set the default PG port if it's missing.
+        if url.port().is_none() {
+            url.set_port(Some(5432))
+                .expect("BUG: couldn't set the default port");
+        }
+
+        Ok(Self { url })
+    }
+}

control_plane/src/etcd.rs

@@ -1,95 +1,75 @@
-use std::{
-    fs,
-    path::PathBuf,
-    process::{Command, Stdio},
-};
+use std::{fs, path::PathBuf};
 
 use anyhow::Context;
-use nix::{
-    sys::signal::{kill, Signal},
-    unistd::Pid,
-};
 
-use crate::{local_env, read_pidfile};
+use crate::{background_process, local_env};
 
 pub fn start_etcd_process(env: &local_env::LocalEnv) -> anyhow::Result<()> {
     let etcd_broker = &env.etcd_broker;
-    println!(
-        "Starting etcd broker using {}",
-        etcd_broker.etcd_binary_path.display()
+    print!(
+        "Starting etcd broker using {:?}",
+        etcd_broker.etcd_binary_path
     );
 
     let etcd_data_dir = env.base_data_dir.join("etcd");
-    fs::create_dir_all(&etcd_data_dir).with_context(|| {
-        format!(
-            "Failed to create etcd data dir: {}",
-            etcd_data_dir.display()
-        )
-    })?;
+    fs::create_dir_all(&etcd_data_dir)
+        .with_context(|| format!("Failed to create etcd data dir {etcd_data_dir:?}"))?;
 
-    let etcd_stdout_file =
-        fs::File::create(etcd_data_dir.join("etcd.stdout.log")).with_context(|| {
-            format!(
-                "Failed to create etcd stout file in directory {}",
-                etcd_data_dir.display()
-            )
-        })?;
-    let etcd_stderr_file =
-        fs::File::create(etcd_data_dir.join("etcd.stderr.log")).with_context(|| {
-            format!(
-                "Failed to create etcd stderr file in directory {}",
-                etcd_data_dir.display()
-            )
-        })?;
     let client_urls = etcd_broker.comma_separated_endpoints();
+    let args = [
+        format!("--data-dir={}", etcd_data_dir.display()),
+        format!("--listen-client-urls={client_urls}"),
+        format!("--advertise-client-urls={client_urls}"),
+        // Set --quota-backend-bytes to keep the etcd virtual memory
+        // size smaller. Our test etcd clusters are very small.
+        // See https://github.com/etcd-io/etcd/issues/7910
+        "--quota-backend-bytes=100000000".to_string(),
+        // etcd doesn't compact (vacuum) with default settings,
+        // enable it to prevent space exhaustion.
+        "--auto-compaction-mode=revision".to_string(),
+        "--auto-compaction-retention=1".to_string(),
+    ];
 
-    let etcd_process = Command::new(&etcd_broker.etcd_binary_path)
-        .args(&[
-            format!("--data-dir={}", etcd_data_dir.display()),
-            format!("--listen-client-urls={client_urls}"),
-            format!("--advertise-client-urls={client_urls}"),
-            // Set --quota-backend-bytes to keep the etcd virtual memory
-            // size smaller. Our test etcd clusters are very small.
-            // See https://github.com/etcd-io/etcd/issues/7910
-            "--quota-backend-bytes=100000000".to_string(),
-        ])
-        .stdout(Stdio::from(etcd_stdout_file))
-        .stderr(Stdio::from(etcd_stderr_file))
-        .spawn()
-        .context("Failed to spawn etcd subprocess")?;
-    let pid = etcd_process.id();
+    let pid_file_path = etcd_pid_file_path(env);
 
-    let etcd_pid_file_path = etcd_pid_file_path(env);
-    fs::write(&etcd_pid_file_path, pid.to_string()).with_context(|| {
-        format!(
-            "Failed to create etcd pid file at {}",
-            etcd_pid_file_path.display()
-        )
-    })?;
+    let client = reqwest::blocking::Client::new();
+
+    background_process::start_process(
+        "etcd",
+        &etcd_data_dir,
+        &etcd_broker.etcd_binary_path,
+        &args,
+        background_process::InitialPidFile::Create(&pid_file_path),
+        || {
+            for broker_endpoint in &etcd_broker.broker_endpoints {
+                let request = broker_endpoint
+                    .join("health")
+                    .with_context(|| {
+                        format!(
+                            "Failed to append /health path to broker endopint {}",
+                            broker_endpoint
+                        )
+                    })
+                    .and_then(|url| {
+                        client.get(&url.to_string()).build().with_context(|| {
+                            format!("Failed to construct request to etcd endpoint {url}")
+                        })
+                    })?;
+                if client.execute(request).is_ok() {
+                    return Ok(true);
+                }
+            }
+
+            Ok(false)
+        },
+    )
+    .context("Failed to spawn etcd subprocess")?;
 
     Ok(())
 }
 
 pub fn stop_etcd_process(env: &local_env::LocalEnv) -> anyhow::Result<()> {
-    let etcd_path = &env.etcd_broker.etcd_binary_path;
-    println!("Stopping etcd broker at {}", etcd_path.display());
-
-    let etcd_pid_file_path = etcd_pid_file_path(env);
-    let pid = Pid::from_raw(read_pidfile(&etcd_pid_file_path).with_context(|| {
-        format!(
-            "Failed to read etcd pid file at {}",
-            etcd_pid_file_path.display()
-        )
-    })?);
-
-    kill(pid, Signal::SIGTERM).with_context(|| {
-        format!(
-            "Failed to stop etcd with pid {pid} at {}",
-            etcd_pid_file_path.display()
-        )
-    })?;
-
-    Ok(())
+    background_process::stop_process(true, "etcd", &etcd_pid_file_path(env))
 }
 
 fn etcd_pid_file_path(env: &local_env::LocalEnv) -> PathBuf {

control_plane/src/lib.rs

@@ -6,59 +6,12 @@
 // Intended to be used in integration tests and in CLI tools for
 // local installations.
 //
-use anyhow::{anyhow, bail, Context, Result};
-use std::fs;
-use std::path::Path;
-use std::process::Command;
 
+mod background_process;
 pub mod compute;
+pub mod connection;
 pub mod etcd;
 pub mod local_env;
+pub mod pageserver;
 pub mod postgresql_conf;
 pub mod safekeeper;
-pub mod storage;
-
-/// Read a PID file
-///
-/// We expect a file that contains a single integer.
-/// We return an i32 for compatibility with libc and nix.
-pub fn read_pidfile(pidfile: &Path) -> Result<i32> {
-    let pid_str = fs::read_to_string(pidfile)
-        .with_context(|| format!("failed to read pidfile {:?}", pidfile))?;
-    let pid: i32 = pid_str
-        .parse()
-        .map_err(|_| anyhow!("failed to parse pidfile {:?}", pidfile))?;
-    if pid < 1 {
-        bail!("pidfile {:?} contained bad value '{}'", pidfile, pid);
-    }
-    Ok(pid)
-}
-
-fn fill_rust_env_vars(cmd: &mut Command) -> &mut Command {
-    let cmd = cmd.env_clear().env("RUST_BACKTRACE", "1");
-
-    let var = "LLVM_PROFILE_FILE";
-    if let Some(val) = std::env::var_os(var) {
-        cmd.env(var, val);
-    }
-
-    const RUST_LOG_KEY: &str = "RUST_LOG";
-    if let Ok(rust_log_value) = std::env::var(RUST_LOG_KEY) {
-        cmd.env(RUST_LOG_KEY, rust_log_value)
-    } else {
-        cmd
-    }
-}
-
-fn fill_aws_secrets_vars(mut cmd: &mut Command) -> &mut Command {
-    for env_key in [
-        "AWS_ACCESS_KEY_ID",
-        "AWS_SECRET_ACCESS_KEY",
-        "AWS_SESSION_TOKEN",
-    ] {
-        if let Ok(value) = std::env::var(env_key) {
-            cmd = cmd.env(env_key, value);
-        }
-    }
-    cmd
-}

control_plane/src/local_env.rs

@@ -201,37 +201,37 @@ impl LocalEnv {
         self.pg_distrib_dir.clone()
     }
 
-    pub fn pg_distrib_dir(&self, pg_version: u32) -> PathBuf {
+    pub fn pg_distrib_dir(&self, pg_version: u32) -> anyhow::Result<PathBuf> {
         let path = self.pg_distrib_dir.clone();
 
         match pg_version {
-            14 => path.join(format!("v{pg_version}")),
-            15 => path.join(format!("v{pg_version}")),
-            _ => panic!("Unsupported postgres version: {}", pg_version),
+            14 => Ok(path.join(format!("v{pg_version}"))),
+            15 => Ok(path.join(format!("v{pg_version}"))),
+            _ => bail!("Unsupported postgres version: {}", pg_version),
         }
     }
 
-    pub fn pg_bin_dir(&self, pg_version: u32) -> PathBuf {
+    pub fn pg_bin_dir(&self, pg_version: u32) -> anyhow::Result<PathBuf> {
         match pg_version {
-            14 => self.pg_distrib_dir(pg_version).join("bin"),
-            15 => self.pg_distrib_dir(pg_version).join("bin"),
-            _ => panic!("Unsupported postgres version: {}", pg_version),
+            14 => Ok(self.pg_distrib_dir(pg_version)?.join("bin")),
+            15 => Ok(self.pg_distrib_dir(pg_version)?.join("bin")),
+            _ => bail!("Unsupported postgres version: {}", pg_version),
         }
     }
-    pub fn pg_lib_dir(&self, pg_version: u32) -> PathBuf {
+    pub fn pg_lib_dir(&self, pg_version: u32) -> anyhow::Result<PathBuf> {
         match pg_version {
-            14 => self.pg_distrib_dir(pg_version).join("lib"),
-            15 => self.pg_distrib_dir(pg_version).join("lib"),
-            _ => panic!("Unsupported postgres version: {}", pg_version),
+            14 => Ok(self.pg_distrib_dir(pg_version)?.join("lib")),
+            15 => Ok(self.pg_distrib_dir(pg_version)?.join("lib")),
+            _ => bail!("Unsupported postgres version: {}", pg_version),
         }
     }
 
-    pub fn pageserver_bin(&self) -> anyhow::Result<PathBuf> {
-        Ok(self.neon_distrib_dir.join("pageserver"))
+    pub fn pageserver_bin(&self) -> PathBuf {
+        self.neon_distrib_dir.join("pageserver")
     }
 
-    pub fn safekeeper_bin(&self) -> anyhow::Result<PathBuf> {
-        Ok(self.neon_distrib_dir.join("safekeeper"))
+    pub fn safekeeper_bin(&self) -> PathBuf {
+        self.neon_distrib_dir.join("safekeeper")
     }
 
     pub fn pg_data_dirs_path(&self) -> PathBuf {
@@ -422,10 +422,10 @@ impl LocalEnv {
             "directory '{}' already exists. Perhaps already initialized?",
             base_path.display()
         );
-        if !self.pg_bin_dir(pg_version).join("postgres").exists() {
+        if !self.pg_bin_dir(pg_version)?.join("postgres").exists() {
             bail!(
                 "Can't find postgres binary at {}",
-                self.pg_bin_dir(pg_version).display()
+                self.pg_bin_dir(pg_version)?.display()
             );
         }
         for binary in ["pageserver", "safekeeper"] {

control_plane/src/pageserver.rs

@@ -1,33 +1,27 @@
 use std::collections::HashMap;
-use std::fs::File;
+use std::fs::{self, File};
 use std::io::{BufReader, Write};
 use std::num::NonZeroU64;
 use std::path::{Path, PathBuf};
-use std::process::Command;
-use std::time::Duration;
-use std::{io, result, thread};
+use std::process::Child;
+use std::{io, result};
 
+use crate::connection::PgConnectionConfig;
 use anyhow::{bail, Context};
-use nix::errno::Errno;
-use nix::sys::signal::{kill, Signal};
-use nix::unistd::Pid;
 use pageserver_api::models::{
     TenantConfigRequest, TenantCreateRequest, TenantInfo, TimelineCreateRequest, TimelineInfo,
 };
-use postgres::{Config, NoTls};
 use reqwest::blocking::{Client, RequestBuilder, Response};
 use reqwest::{IntoUrl, Method};
 use thiserror::Error;
 use utils::{
-    connstring::connection_address,
     http::error::HttpErrorBody,
     id::{TenantId, TimelineId},
     lsn::Lsn,
     postgres_backend::AuthType,
 };
 
-use crate::local_env::LocalEnv;
-use crate::{fill_aws_secrets_vars, fill_rust_env_vars, read_pidfile};
+use crate::{background_process, local_env::LocalEnv};
 
 #[derive(Error, Debug)]
 pub enum PageserverHttpError {
@@ -75,7 +69,7 @@ impl ResponseErrorMessageExt for Response {
 //
 #[derive(Debug)]
 pub struct PageServerNode {
-    pub pg_connection_config: Config,
+    pub pg_connection_config: PgConnectionConfig,
     pub env: LocalEnv,
     pub http_client: Client,
     pub http_base_url: String,
@@ -101,7 +95,7 @@ impl PageServerNode {
     }
 
     /// Construct libpq connection string for connecting to the pageserver.
-    fn pageserver_connection_config(password: &str, listen_addr: &str) -> Config {
+    fn pageserver_connection_config(password: &str, listen_addr: &str) -> PgConnectionConfig {
         format!("postgresql://no_user:{password}@{listen_addr}/no_db")
             .parse()
             .unwrap()
@@ -161,7 +155,15 @@ impl PageServerNode {
             init_config_overrides.push("auth_validation_public_key_path='auth_public_key.pem'");
         }
 
-        self.start_node(&init_config_overrides, &self.env.base_data_dir, true)?;
+        let mut pageserver_process = self
+            .start_node(&init_config_overrides, &self.env.base_data_dir, true)
+            .with_context(|| {
+                format!(
+                    "Failed to start a process for pageserver {}",
+                    self.env.pageserver.id,
+                )
+            })?;
+
         let init_result = self
             .try_init_timeline(create_tenant, initial_timeline_id, pg_version)
             .context("Failed to create initial tenant and timeline for pageserver");
@@ -171,7 +173,29 @@ impl PageServerNode {
             }
             Err(e) => eprintln!("{e:#}"),
         }
-        self.stop(false)?;
+        match pageserver_process.kill() {
+            Err(e) => {
+                eprintln!(
+                    "Failed to stop pageserver {} process with pid {}: {e:#}",
+                    self.env.pageserver.id,
+                    pageserver_process.id(),
+                )
+            }
+            Ok(()) => {
+                println!(
+                    "Stopped pageserver {} process with pid {}",
+                    self.env.pageserver.id,
+                    pageserver_process.id(),
+                );
+                // cleanup after pageserver startup, since we do not call regular `stop_process` during init
+                let pid_file = self.pid_file();
+                if let Err(e) = fs::remove_file(&pid_file) {
+                    if e.kind() != io::ErrorKind::NotFound {
+                        eprintln!("Failed to remove pid file {pid_file:?} after stopping the process: {e:#}");
+                    }
+                }
+            }
+        }
         init_result
     }
 
@@ -196,11 +220,14 @@ impl PageServerNode {
         self.env.pageserver_data_dir()
     }
 
-    pub fn pid_file(&self) -> PathBuf {
+    /// The pid file is created by the pageserver process, with its pid stored inside.
+    /// Other pageservers cannot lock the same file and overwrite it for as long as the current
+    /// pageserver runs. (Unless someone removes the file manually; never do that!)
+    fn pid_file(&self) -> PathBuf {
         self.repo_path().join("pageserver.pid")
     }
 
-    pub fn start(&self, config_overrides: &[&str]) -> anyhow::Result<()> {
+    pub fn start(&self, config_overrides: &[&str]) -> anyhow::Result<Child> {
         self.start_node(config_overrides, &self.repo_path(), false)
     }
 
@@ -209,10 +236,10 @@ impl PageServerNode {
         config_overrides: &[&str],
         datadir: &Path,
         update_config: bool,
-    ) -> anyhow::Result<()> {
-        println!(
+    ) -> anyhow::Result<Child> {
+        print!(
             "Starting pageserver at '{}' in '{}'",
-            connection_address(&self.pg_connection_config),
+            self.pg_connection_config.raw_address(),
             datadir.display()
         );
         io::stdout().flush()?;
@@ -220,10 +247,7 @@ impl PageServerNode {
         let mut args = vec![
             "-D",
             datadir.to_str().with_context(|| {
-                format!(
-                    "Datadir path '{}' cannot be represented as a unicode string",
-                    datadir.display()
-                )
+                format!("Datadir path {datadir:?} cannot be represented as a unicode string")
             })?,
         ];
 
@@ -235,48 +259,18 @@ impl PageServerNode {
             args.extend(["-c", config_override]);
         }
 
-        let mut cmd = Command::new(self.env.pageserver_bin()?);
-        let mut filled_cmd = fill_rust_env_vars(cmd.args(&args).arg("--daemonize"));
-        filled_cmd = fill_aws_secrets_vars(filled_cmd);
-
-        if !filled_cmd.status()?.success() {
-            bail!(
-                "Pageserver failed to start. See console output and '{}' for details.",
-                datadir.join("pageserver.log").display()
-            );
-        }
-
-        // It takes a while for the page server to start up. Wait until it is
-        // open for business.
-        const RETRIES: i8 = 15;
-        for retries in 1..RETRIES {
-            match self.check_status() {
-                Ok(()) => {
-                    println!("\nPageserver started");
-                    return Ok(());
-                }
-                Err(err) => {
-                    match err {
-                        PageserverHttpError::Transport(err) => {
-                            if err.is_connect() && retries < 5 {
-                                print!(".");
-                                io::stdout().flush().unwrap();
-                            } else {
-                                if retries == 5 {
-                                    println!() // put a line break after dots for second message
-                                }
-                                println!("Pageserver not responding yet, err {err} retrying ({retries})...");
-                            }
-                        }
-                        PageserverHttpError::Response(msg) => {
-                            bail!("pageserver failed to start: {msg} ")
-                        }
-                    }
-                    thread::sleep(Duration::from_secs(1));
-                }
-            }
-        }
-        bail!("pageserver failed to start in {RETRIES} seconds");
+        background_process::start_process(
+            "pageserver",
+            datadir,
+            &self.env.pageserver_bin(),
+            &args,
+            background_process::InitialPidFile::Expect(&self.pid_file()),
+            || match self.check_status() {
+                Ok(()) => Ok(true),
+                Err(PageserverHttpError::Transport(_)) => Ok(false),
+                Err(e) => Err(anyhow::anyhow!("Failed to check node status: {e}")),
+            },
+        )
     }
 
     ///
@@ -288,69 +282,18 @@ impl PageServerNode {
     /// If the server is not running, returns success
     ///
     pub fn stop(&self, immediate: bool) -> anyhow::Result<()> {
-        let pid_file = self.pid_file();
-        if !pid_file.exists() {
-            println!("Pageserver is already stopped");
-            return Ok(());
-        }
-        let pid = Pid::from_raw(read_pidfile(&pid_file)?);
-
-        let sig = if immediate {
-            print!("Stopping pageserver immediately..");
-            Signal::SIGQUIT
-        } else {
-            print!("Stopping pageserver gracefully..");
-            Signal::SIGTERM
-        };
-        io::stdout().flush().unwrap();
-        match kill(pid, sig) {
-            Ok(_) => (),
-            Err(Errno::ESRCH) => {
-                println!("Pageserver with pid {pid} does not exist, but a PID file was found");
-                return Ok(());
-            }
-            Err(err) => bail!(
-                "Failed to send signal to pageserver with pid {pid}: {}",
-                err.desc()
-            ),
-        }
-
-        // Wait until process is gone
-        for i in 0..600 {
-            let signal = None; // Send no signal, just get the error code
-            match kill(pid, signal) {
-                Ok(_) => (), // Process exists, keep waiting
-                Err(Errno::ESRCH) => {
-                    // Process not found, we're done
-                    println!("done!");
-                    return Ok(());
-                }
-                Err(err) => bail!(
-                    "Failed to send signal to pageserver with pid {}: {}",
-                    pid,
-                    err.desc()
-                ),
-            };
-
-            if i % 10 == 0 {
-                print!(".");
-                io::stdout().flush().unwrap();
-            }
-            thread::sleep(Duration::from_millis(100));
-        }
-
-        bail!("Failed to stop pageserver with pid {pid}");
+        background_process::stop_process(immediate, "pageserver", &self.pid_file())
     }
 
     pub fn page_server_psql(&self, sql: &str) -> Vec<postgres::SimpleQueryMessage> {
-        let mut client = self.pg_connection_config.connect(NoTls).unwrap();
+        let mut client = self.pg_connection_config.connect_no_tls().unwrap();
 
         println!("Pageserver query: '{sql}'");
         client.simple_query(sql).unwrap()
     }
 
     pub fn page_server_psql_client(&self) -> result::Result<postgres::Client, postgres::Error> {
-        self.pg_connection_config.connect(NoTls)
+        self.pg_connection_config.connect_no_tls()
     }
 
     fn http_request<U: IntoUrl>(&self, method: Method, url: U) -> RequestBuilder {
@@ -419,6 +362,11 @@ impl PageServerNode {
                 .map(|x| x.parse::<NonZeroU64>())
                 .transpose()
                 .context("Failed to parse 'max_lsn_wal_lag' as non zero integer")?,
+            trace_read_requests: settings
+                .remove("trace_read_requests")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'trace_read_requests' as bool")?,
         };
         if !settings.is_empty() {
             bail!("Unrecognized tenant settings: {settings:?}")
@@ -481,6 +429,11 @@ impl PageServerNode {
                     .map(|x| x.parse::<NonZeroU64>())
                     .transpose()
                     .context("Failed to parse 'max_lsn_wal_lag' as non zero integer")?,
+                trace_read_requests: settings
+                    .get("trace_read_requests")
+                    .map(|x| x.parse::<bool>())
+                    .transpose()
+                    .context("Failed to parse 'trace_read_requests' as bool")?,
             })
             .send()?
             .error_from_body()?;
@@ -549,7 +502,7 @@ impl PageServerNode {
         pg_wal: Option<(Lsn, PathBuf)>,
         pg_version: u32,
     ) -> anyhow::Result<()> {
-        let mut client = self.pg_connection_config.connect(NoTls).unwrap();
+        let mut client = self.pg_connection_config.connect_no_tls().unwrap();
 
         // Init base reader
         let (start_lsn, base_tarfile_path) = base;

control_plane/src/safekeeper.rs

@@ -1,28 +1,21 @@
 use std::io::Write;
 use std::path::PathBuf;
-use std::process::Command;
+use std::process::Child;
 use std::sync::Arc;
-use std::time::Duration;
-use std::{io, result, thread};
+use std::{io, result};
 
-use anyhow::bail;
-use nix::errno::Errno;
-use nix::sys::signal::{kill, Signal};
-use nix::unistd::Pid;
-use postgres::Config;
+use anyhow::Context;
 use reqwest::blocking::{Client, RequestBuilder, Response};
 use reqwest::{IntoUrl, Method};
-use safekeeper_api::models::TimelineCreateRequest;
 use thiserror::Error;
-use utils::{
-    connstring::connection_address,
-    http::error::HttpErrorBody,
-    id::{NodeId, TenantId, TimelineId},
-};
+use utils::{http::error::HttpErrorBody, id::NodeId};
 
-use crate::local_env::{LocalEnv, SafekeeperConf};
-use crate::storage::PageServerNode;
-use crate::{fill_aws_secrets_vars, fill_rust_env_vars, read_pidfile};
+use crate::connection::PgConnectionConfig;
+use crate::pageserver::PageServerNode;
+use crate::{
+    background_process,
+    local_env::{LocalEnv, SafekeeperConf},
+};
 
 #[derive(Error, Debug)]
 pub enum SafekeeperHttpError {
@@ -68,7 +61,7 @@ pub struct SafekeeperNode {
 
     pub conf: SafekeeperConf,
 
-    pub pg_connection_config: Config,
+    pub pg_connection_config: PgConnectionConfig,
     pub env: LocalEnv,
     pub http_client: Client,
     pub http_base_url: String,
@@ -92,15 +85,15 @@ impl SafekeeperNode {
     }
 
     /// Construct libpq connection string for connecting to this safekeeper.
-    fn safekeeper_connection_config(port: u16) -> Config {
+    fn safekeeper_connection_config(port: u16) -> PgConnectionConfig {
         // TODO safekeeper authentication not implemented yet
-        format!("postgresql://no_user@127.0.0.1:{}/no_db", port)
+        format!("postgresql://no_user@127.0.0.1:{port}/no_db")
             .parse()
             .unwrap()
     }
 
     pub fn datadir_path_by_id(env: &LocalEnv, sk_id: NodeId) -> PathBuf {
-        env.safekeeper_data_dir(format!("sk{}", sk_id).as_ref())
+        env.safekeeper_data_dir(&format!("sk{sk_id}"))
     }
 
     pub fn datadir_path(&self) -> PathBuf {
@@ -111,92 +104,78 @@ impl SafekeeperNode {
         self.datadir_path().join("safekeeper.pid")
     }
 
-    pub fn start(&self) -> anyhow::Result<()> {
+    pub fn start(&self) -> anyhow::Result<Child> {
         print!(
             "Starting safekeeper at '{}' in '{}'",
-            connection_address(&self.pg_connection_config),
+            self.pg_connection_config.raw_address(),
             self.datadir_path().display()
         );
         io::stdout().flush().unwrap();
 
         let listen_pg = format!("127.0.0.1:{}", self.conf.pg_port);
         let listen_http = format!("127.0.0.1:{}", self.conf.http_port);
+        let id = self.id;
+        let datadir = self.datadir_path();
 
-        let mut cmd = Command::new(self.env.safekeeper_bin()?);
-        fill_rust_env_vars(
-            cmd.args(&["-D", self.datadir_path().to_str().unwrap()])
-                .args(&["--id", self.id.to_string().as_ref()])
-                .args(&["--listen-pg", &listen_pg])
-                .args(&["--listen-http", &listen_http])
-                .args(&["--recall", "1 second"])
-                .arg("--daemonize"),
-        );
+        let id_string = id.to_string();
+        let mut args = vec![
+            "-D",
+            datadir.to_str().with_context(|| {
+                format!("Datadir path {datadir:?} cannot be represented as a unicode string")
+            })?,
+            "--id",
+            &id_string,
+            "--listen-pg",
+            &listen_pg,
+            "--listen-http",
+            &listen_http,
+        ];
         if !self.conf.sync {
-            cmd.arg("--no-sync");
+            args.push("--no-sync");
         }
 
         let comma_separated_endpoints = self.env.etcd_broker.comma_separated_endpoints();
         if !comma_separated_endpoints.is_empty() {
-            cmd.args(&["--broker-endpoints", &comma_separated_endpoints]);
+            args.extend(["--broker-endpoints", &comma_separated_endpoints]);
         }
         if let Some(prefix) = self.env.etcd_broker.broker_etcd_prefix.as_deref() {
-            cmd.args(&["--broker-etcd-prefix", prefix]);
+            args.extend(["--broker-etcd-prefix", prefix]);
         }
+
+        let mut backup_threads = String::new();
         if let Some(threads) = self.conf.backup_threads {
-            cmd.args(&["--backup-threads", threads.to_string().as_ref()]);
+            backup_threads = threads.to_string();
+            args.extend(["--backup-threads", &backup_threads]);
+        } else {
+            drop(backup_threads);
         }
+
         if let Some(ref remote_storage) = self.conf.remote_storage {
-            cmd.args(&["--remote-storage", remote_storage]);
+            args.extend(["--remote-storage", remote_storage]);
         }
+
+        let key_path = self.env.base_data_dir.join("auth_public_key.pem");
         if self.conf.auth_enabled {
-            cmd.arg("--auth-validation-public-key-path");
-            // PathBuf is better be passed as is, not via `String`.
-            cmd.arg(self.env.base_data_dir.join("auth_public_key.pem"));
+            args.extend([
+                "--auth-validation-public-key-path",
+                key_path.to_str().with_context(|| {
+                    format!("Key path {key_path:?} cannot be represented as a unicode string")
+                })?,
+            ]);
         }
 
-        fill_aws_secrets_vars(&mut cmd);
-
-        if !cmd.status()?.success() {
-            bail!(
-                "Safekeeper failed to start. See '{}' for details.",
-                self.datadir_path().join("safekeeper.log").display()
-            );
-        }
-
-        // It takes a while for the safekeeper to start up. Wait until it is
-        // open for business.
-        const RETRIES: i8 = 15;
-        for retries in 1..RETRIES {
-            match self.check_status() {
-                Ok(_) => {
-                    println!("\nSafekeeper started");
-                    return Ok(());
-                }
-                Err(err) => {
-                    match err {
-                        SafekeeperHttpError::Transport(err) => {
-                            if err.is_connect() && retries < 5 {
-                                print!(".");
-                                io::stdout().flush().unwrap();
-                            } else {
-                                if retries == 5 {
-                                    println!() // put a line break after dots for second message
-                                }
-                                println!(
-                                    "Safekeeper not responding yet, err {} retrying ({})...",
-                                    err, retries
-                                );
-                            }
-                        }
-                        SafekeeperHttpError::Response(msg) => {
-                            bail!("safekeeper failed to start: {} ", msg)
-                        }
-                    }
-                    thread::sleep(Duration::from_secs(1));
-                }
-            }
-        }
-        bail!("safekeeper failed to start in {} seconds", RETRIES);
+        background_process::start_process(
+            &format!("safekeeper {id}"),
+            &datadir,
+            &self.env.safekeeper_bin(),
+            &args,
+            background_process::InitialPidFile::Expect(&self.pid_file()),
+            || match self.check_status() {
+                Ok(()) => Ok(true),
+                Err(SafekeeperHttpError::Transport(_)) => Ok(false),
+                Err(e) => Err(anyhow::anyhow!("Failed to check node status: {e}")),
+            },
+        )
     }
 
     ///
@@ -208,63 +187,11 @@ impl SafekeeperNode {
     /// If the server is not running, returns success
     ///
     pub fn stop(&self, immediate: bool) -> anyhow::Result<()> {
-        let pid_file = self.pid_file();
-        if !pid_file.exists() {
-            println!("Safekeeper {} is already stopped", self.id);
-            return Ok(());
-        }
-        let pid = read_pidfile(&pid_file)?;
-        let pid = Pid::from_raw(pid);
-
-        let sig = if immediate {
-            print!("Stopping safekeeper {} immediately..", self.id);
-            Signal::SIGQUIT
-        } else {
-            print!("Stopping safekeeper {} gracefully..", self.id);
-            Signal::SIGTERM
-        };
-        io::stdout().flush().unwrap();
-        match kill(pid, sig) {
-            Ok(_) => (),
-            Err(Errno::ESRCH) => {
-                println!(
-                    "Safekeeper with pid {} does not exist, but a PID file was found",
-                    pid
-                );
-                return Ok(());
-            }
-            Err(err) => bail!(
-                "Failed to send signal to safekeeper with pid {}: {}",
-                pid,
-                err.desc()
-            ),
-        }
-
-        // Wait until process is gone
-        for i in 0..600 {
-            let signal = None; // Send no signal, just get the error code
-            match kill(pid, signal) {
-                Ok(_) => (), // Process exists, keep waiting
-                Err(Errno::ESRCH) => {
-                    // Process not found, we're done
-                    println!("done!");
-                    return Ok(());
-                }
-                Err(err) => bail!(
-                    "Failed to send signal to pageserver with pid {}: {}",
-                    pid,
-                    err.desc()
-                ),
-            };
-
-            if i % 10 == 0 {
-                print!(".");
-                io::stdout().flush().unwrap();
-            }
-            thread::sleep(Duration::from_millis(100));
-        }
-
-        bail!("Failed to stop safekeeper with pid {}", pid);
+        background_process::stop_process(
+            immediate,
+            &format!("safekeeper {}", self.id),
+            &self.pid_file(),
+        )
     }
 
     fn http_request<U: IntoUrl>(&self, method: Method, url: U) -> RequestBuilder {
@@ -281,24 +208,4 @@ impl SafekeeperNode {
             .error_from_body()?;
         Ok(())
     }
-
-    pub fn timeline_create(
-        &self,
-        tenant_id: TenantId,
-        timeline_id: TimelineId,
-        peer_ids: Vec<NodeId>,
-    ) -> Result<()> {
-        Ok(self
-            .http_request(
-                Method::POST,
-                format!("{}/tenant/{}/timeline", self.http_base_url, tenant_id),
-            )
-            .json(&TimelineCreateRequest {
-                timeline_id,
-                peer_ids,
-            })
-            .send()?
-            .error_from_body()?
-            .json()?)
-    }
 }

docker-compose/compute_wrapper/Dockerfile

@@ -0,0 +1,13 @@
+ARG REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
+ARG COMPUTE_IMAGE=compute-node-v14
+ARG TAG=latest
+
+FROM $REPOSITORY/${COMPUTE_IMAGE}:$TAG
+
+USER root
+RUN apt-get update &&       \
+    apt-get install -y curl \
+                       jq   \
+                       netcat
+
+USER postgres

docker-compose/compute_wrapper/shell/compute.sh

@@ -0,0 +1,48 @@
+#!/bin/bash
+set -eux
+
+PG_VERSION=${PG_VERSION:-14}
+
+SPEC_FILE_ORG=/var/db/postgres/specs/spec.json
+SPEC_FILE=/tmp/spec.json
+
+echo "Waiting pageserver become ready."
+while ! nc -z pageserver 6400; do
+     sleep 1;
+done
+echo "Page server is ready."
+
+echo "Create a tenant and timeline"
+PARAMS=(
+     -sb 
+     -X POST
+     -H "Content-Type: application/json"
+     -d "{}"
+     http://pageserver:9898/v1/tenant/
+)
+tenant_id=$(curl "${PARAMS[@]}" | sed 's/"//g')
+
+PARAMS=(
+     -sb 
+     -X POST
+     -H "Content-Type: application/json"
+     -d "{\"tenant_id\":\"${tenant_id}\", \"pg_version\": ${PG_VERSION}}"
+     "http://pageserver:9898/v1/tenant/${tenant_id}/timeline/"
+)
+result=$(curl "${PARAMS[@]}")
+echo $result | jq .
+
+echo "Overwrite tenant id and timeline id in spec file"
+tenant_id=$(echo ${result} | jq -r .tenant_id)
+timeline_id=$(echo ${result} | jq -r .timeline_id)
+
+sed "s/TENANT_ID/${tenant_id}/" ${SPEC_FILE_ORG} > ${SPEC_FILE}
+sed -i "s/TIMELINE_ID/${timeline_id}/" ${SPEC_FILE}
+
+cat ${SPEC_FILE}
+
+echo "Start compute node"
+/usr/local/bin/compute_ctl --pgdata /var/db/postgres/compute \
+     -C "postgresql://cloud_admin@localhost:55433/postgres"  \
+     -b /usr/local/bin/postgres                              \
+     -S ${SPEC_FILE}

docker-compose/compute_wrapper/var/db/postgres/specs/spec.json

@@ -0,0 +1,141 @@
+{
+    "format_version": 1.0,
+
+    "timestamp": "2022-10-12T18:00:00.000Z",
+    "operation_uuid": "0f657b36-4b0f-4a2d-9c2e-1dcd615e7d8c",
+
+    "cluster": {
+        "cluster_id": "docker_compose",
+        "name": "docker_compose_test",
+        "state": "restarted",
+        "roles": [
+            {
+                "name": "cloud_admin",
+                "encrypted_password": "b093c0d3b281ba6da1eacc608620abd8",
+                "options": null
+            }
+        ],
+        "databases": [
+        ],
+        "settings": [
+            {
+                "name": "fsync",
+                "value": "off",
+                "vartype": "bool"
+            },
+            {
+                "name": "wal_level",
+                "value": "replica",
+                "vartype": "enum"
+            },
+            {
+                "name": "hot_standby",
+                "value": "on",
+                "vartype": "bool"
+            },
+            {
+                "name": "wal_log_hints",
+                "value": "on",
+                "vartype": "bool"
+            },
+            {
+                "name": "log_connections",
+                "value": "on",
+                "vartype": "bool"
+            },
+            {
+                "name": "port",
+                "value": "55433",
+                "vartype": "integer"
+            },
+            {
+                "name": "shared_buffers",
+                "value": "1MB",
+                "vartype": "string"
+            },
+            {
+                "name": "max_connections",
+                "value": "100",
+                "vartype": "integer"
+            },
+            {
+                "name": "listen_addresses",
+                "value": "0.0.0.0",
+                "vartype": "string"
+            },
+            {
+                "name": "max_wal_senders",
+                "value": "10",
+                "vartype": "integer"
+            },
+            {
+                "name": "max_replication_slots",
+                "value": "10",
+                "vartype": "integer"
+            },
+            {
+                "name": "wal_sender_timeout",
+                "value": "5s",
+                "vartype": "string"
+            },
+            {
+                "name": "wal_keep_size",
+                "value": "0",
+                "vartype": "integer"
+            },
+            {
+                "name": "password_encryption",
+                "value": "md5",
+                "vartype": "enum"
+            },
+            {
+                "name": "restart_after_crash",
+                "value": "off",
+                "vartype": "bool"
+            },
+            {
+                "name": "synchronous_standby_names",
+                "value": "walproposer",
+                "vartype": "string"
+            },
+            {
+                "name": "shared_preload_libraries",
+                "value": "neon",
+                "vartype": "string"
+            },
+            {
+                "name": "neon.safekeepers",
+                "value": "safekeeper1:5454,safekeeper2:5454,safekeeper3:5454",
+                "vartype": "string"
+            },
+            {
+                "name": "neon.timeline_id",
+                "value": "TIMELINE_ID",
+                "vartype": "string"
+            },
+            {
+                "name": "neon.tenant_id",
+                "value": "TENANT_ID",
+                "vartype": "string"
+            },
+            {
+                "name": "neon.pageserver_connstring",
+                "value": "host=pageserver port=6400",
+                "vartype": "string"
+            },
+            {
+                "name": "max_replication_write_lag",
+                "value": "500MB",
+                "vartype": "string"
+            },
+            {
+                "name": "max_replication_flush_lag",
+                "value": "10GB",
+                "vartype": "string"
+            }
+        ]
+    },
+
+    "delta_operations": [
+    ]
+}

docker-compose/docker-compose.yml

@@ -0,0 +1,209 @@
+version: '3'
+
+services:
+  etcd:
+    restart: always
+    image: quay.io/coreos/etcd:v3.5.4
+    ports:
+      - 2379:2379
+      - 2380:2380
+    environment:
+      # This signifficantly speeds up etcd and we anyway don't data persistency there.
+      ETCD_UNSAFE_NO_FSYNC: "1"
+    command:
+      - "etcd"
+      - "--auto-compaction-mode=revision"
+      - "--auto-compaction-retention=1"
+      - "--name=etcd-cluster"
+      - "--initial-cluster-state=new"
+      - "--initial-cluster-token=etcd-cluster-1"
+      - "--initial-cluster=etcd-cluster=http://etcd:2380"
+      - "--initial-advertise-peer-urls=http://etcd:2380"
+      - "--advertise-client-urls=http://etcd:2379"
+      - "--listen-client-urls=http://0.0.0.0:2379"
+      - "--listen-peer-urls=http://0.0.0.0:2380"
+      - "--quota-backend-bytes=134217728" # 128 MB
+
+  minio:
+    restart: always
+    image: quay.io/minio/minio:RELEASE.2022-10-20T00-55-09Z
+    ports:
+      - 9000:9000
+      - 9001:9001
+    environment:
+      - MINIO_ROOT_USER=minio
+      - MINIO_ROOT_PASSWORD=password
+    command: server /data --address :9000 --console-address ":9001"
+
+  minio_create_buckets:
+    image: minio/mc
+    environment:
+      - MINIO_ROOT_USER=minio
+      - MINIO_ROOT_PASSWORD=password
+    entrypoint:
+      - "/bin/sh"
+      - "-c"
+    command:
+      - "until (/usr/bin/mc alias set minio http://minio:9000 $$MINIO_ROOT_USER $$MINIO_ROOT_PASSWORD) do
+             echo 'Waiting to start minio...' && sleep 1;
+         done;
+         /usr/bin/mc mb minio/neon --region=eu-north-1;
+         exit 0;"
+    depends_on:
+      - minio
+
+  pageserver:
+    restart: always
+    image: ${REPOSITORY:-neondatabase}/neon:${TAG:-latest}
+    environment:
+      - BROKER_ENDPOINT='http://etcd:2379'
+      - AWS_ACCESS_KEY_ID=minio
+      - AWS_SECRET_ACCESS_KEY=password
+      #- RUST_BACKTRACE=1
+    ports:
+       #- 6400:6400  # pg protocol handler
+       - 9898:9898 # http endpoints
+    entrypoint:
+      - "/bin/sh"
+      - "-c"
+    command:
+      - "/usr/local/bin/pageserver -D /data/.neon/
+                                   -c \"broker_endpoints=[$$BROKER_ENDPOINT]\"
+                                   -c \"listen_pg_addr='0.0.0.0:6400'\"
+                                   -c \"listen_http_addr='0.0.0.0:9898'\"
+                                   -c \"remote_storage={endpoint='http://minio:9000',
+                                                        bucket_name='neon',
+                                                        bucket_region='eu-north-1',
+                                                        prefix_in_bucket='/pageserver/'}\""
+    depends_on:
+      - etcd
+      - minio_create_buckets
+
+  safekeeper1:
+    restart: always
+    image: ${REPOSITORY:-neondatabase}/neon:${TAG:-latest}
+    environment:
+      - SAFEKEEPER_ADVERTISE_URL=safekeeper1:5454
+      - SAFEKEEPER_ID=1
+      - BROKER_ENDPOINT=http://etcd:2379
+      - AWS_ACCESS_KEY_ID=minio
+      - AWS_SECRET_ACCESS_KEY=password
+      #- RUST_BACKTRACE=1
+    ports:
+      #- 5454:5454 # pg protocol handler
+      - 7676:7676 # http endpoints
+    entrypoint:
+      - "/bin/sh"
+      - "-c"
+    command:
+      - "safekeeper --listen-pg=$$SAFEKEEPER_ADVERTISE_URL
+                    --listen-http='0.0.0.0:7676'
+                    --id=$$SAFEKEEPER_ID
+                    --broker-endpoints=$$BROKER_ENDPOINT
+                    -D /data
+                    --remote-storage=\"{endpoint='http://minio:9000',
+                                        bucket_name='neon',
+                                        bucket_region='eu-north-1',
+                                        prefix_in_bucket='/safekeeper/'}\""
+    depends_on:
+      - etcd
+      - minio_create_buckets
+
+  safekeeper2:
+    restart: always
+    image: ${REPOSITORY:-neondatabase}/neon:${TAG:-latest}
+    environment:
+      - SAFEKEEPER_ADVERTISE_URL=safekeeper2:5454
+      - SAFEKEEPER_ID=2
+      - BROKER_ENDPOINT=http://etcd:2379
+      - AWS_ACCESS_KEY_ID=minio
+      - AWS_SECRET_ACCESS_KEY=password
+      #- RUST_BACKTRACE=1
+    ports:
+      #- 5454:5454 # pg protocol handler
+      - 7677:7676 # http endpoints
+    entrypoint:
+      - "/bin/sh"
+      - "-c"
+    command:
+      - "safekeeper --listen-pg=$$SAFEKEEPER_ADVERTISE_URL
+                    --listen-http='0.0.0.0:7676'
+                    --id=$$SAFEKEEPER_ID
+                    --broker-endpoints=$$BROKER_ENDPOINT
+                    -D /data
+                    --remote-storage=\"{endpoint='http://minio:9000',
+                                        bucket_name='neon',
+                                        bucket_region='eu-north-1',
+                                        prefix_in_bucket='/safekeeper/'}\""
+    depends_on:
+      - etcd
+      - minio_create_buckets
+
+  safekeeper3:
+    restart: always
+    image: ${REPOSITORY:-neondatabase}/neon:${TAG:-latest}
+    environment:
+      - SAFEKEEPER_ADVERTISE_URL=safekeeper3:5454
+      - SAFEKEEPER_ID=3
+      - BROKER_ENDPOINT=http://etcd:2379
+      - AWS_ACCESS_KEY_ID=minio
+      - AWS_SECRET_ACCESS_KEY=password
+      #- RUST_BACKTRACE=1
+    ports:
+      #- 5454:5454 # pg protocol handler
+      - 7678:7676 # http endpoints
+    entrypoint:
+      - "/bin/sh"
+      - "-c"
+    command:
+      - "safekeeper --listen-pg=$$SAFEKEEPER_ADVERTISE_URL
+                    --listen-http='0.0.0.0:7676'
+                    --id=$$SAFEKEEPER_ID
+                    --broker-endpoints=$$BROKER_ENDPOINT
+                    -D /data
+                    --remote-storage=\"{endpoint='http://minio:9000',
+                                        bucket_name='neon',
+                                        bucket_region='eu-north-1',
+                                        prefix_in_bucket='/safekeeper/'}\""
+    depends_on:
+      - etcd
+      - minio_create_buckets
+
+  compute:
+    restart: always
+    build:
+      context: ./compute_wrapper/
+      args:
+        - COMPUTE_IMAGE=compute-node-v${PG_VERSION:-14}
+        - TAG=${TAG:-latest}
+        - http_proxy=$http_proxy
+        - https_proxy=$https_proxy
+    environment:
+      - PG_VERSION=${PG_VERSION:-14}
+      #- RUST_BACKTRACE=1
+    # Mount the test files directly, for faster editing cycle.
+    volumes:
+      - ./compute_wrapper/var/db/postgres/specs/:/var/db/postgres/specs/
+      - ./compute_wrapper/shell/:/shell/
+    ports:
+      - 55433:55433 # pg protocol handler
+      - 3080:3080 # http endpoints
+    entrypoint:
+      - "/shell/compute.sh"
+    depends_on:
+      - safekeeper1
+      - safekeeper2
+      - safekeeper3
+      - pageserver
+
+  compute_is_ready:
+    image: postgres:latest
+    entrypoint:
+      - "/bin/bash"
+      - "-c"
+    command:
+      - "until pg_isready -h compute -p 55433 ; do
+            echo 'Waiting to start compute...' && sleep 1;
+         done"
+    depends_on:
+      - compute

docker-compose/docker_compose_test.sh

@@ -0,0 +1,60 @@
+#!/bin/bash
+
+# A basic test to ensure Docker images are built correctly.
+# Build a wrapper around the compute, start all services and runs a simple SQL query.
+# Repeats the process for all currenly supported Postgres versions.
+
+# Implicitly accepts `REPOSITORY` and `TAG` env vars that are passed into the compose file
+# Their defaults point at DockerHub `neondatabase/neon:latest` image.`,
+# to verify custom image builds (e.g pre-published ones).
+
+# XXX: Current does not work on M1 macs due to x86_64 Docker images compiled only, and no seccomp support in M1 Docker emulation layer.
+
+set -eux -o pipefail
+
+SCRIPT_DIR="$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )"
+COMPOSE_FILE=$SCRIPT_DIR/docker-compose.yml
+
+COMPUTE_CONTAINER_NAME=docker-compose-compute-1
+SQL="CREATE TABLE t(key int primary key, value text); insert into t values(1,1); select * from t;"
+PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -c '$SQL' postgres"
+
+cleanup() {
+    echo "show container information"
+    docker ps
+    docker compose -f $COMPOSE_FILE logs
+    echo "stop containers..."
+    docker compose -f $COMPOSE_FILE down
+}
+
+echo "clean up containers if exists"
+cleanup
+
+for pg_version in 14 15; do
+    echo "start containers (pg_version=$pg_version)."
+    PG_VERSION=$pg_version docker compose -f $COMPOSE_FILE up --build -d
+
+    echo "wait until the compute is ready. timeout after 60s. "
+    cnt=0
+    while sleep 1; do
+        # check timeout
+        cnt=`expr $cnt + 1`
+        if [ $cnt -gt 60 ]; then
+            echo "timeout before the compute is ready."
+            cleanup
+            exit 1
+        fi
+
+        # check if the compute is ready
+        set +o pipefail
+        result=`docker compose -f $COMPOSE_FILE logs "compute_is_ready" | grep "accepting connections" | wc -l`
+        set -o pipefail
+        if [ $result -eq 1 ]; then
+            echo "OK. The compute is ready to connect."
+            echo "execute simple queries."
+            docker exec $COMPUTE_CONTAINER_NAME /bin/bash -c "psql $PSQL_OPTION"
+            cleanup
+            break
+        fi
+    done
+done

docs/SUMMARY.md

@@ -37,7 +37,7 @@
 
 - [Source view](./sourcetree.md)
   - [docker.md](./docker.md) — Docker images and building pipeline.
-  - [Error handling and logging]()
+  - [Error handling and logging](./error-handling.md)
   - [Testing]()
     - [Unit testing]()
     - [Integration testing]()
@@ -80,4 +80,6 @@
 - [015-storage-messaging](rfcs/015-storage-messaging.md)
 - [016-connection-routing](rfcs/016-connection-routing.md)
 - [017-timeline-data-management](rfcs/017-timeline-data-management.md)
+- [018-storage-messaging-2](rfcs/018-storage-messaging-2.md)
+- [019-tenant-timeline-lifecycles](rfcs/019-tenant-timeline-lifecycles.md)
 - [cluster-size-limits](rfcs/cluster-size-limits.md)

docs/docker.md

@@ -18,3 +18,67 @@ We build all images after a successful `release` tests run and push automaticall
 1. `neondatabase/compute-tools` and `neondatabase/compute-node`
 
 2. `neondatabase/neon`
+
+## Docker Compose example
+
+You can see a [docker compose](https://docs.docker.com/compose/) example to create a neon cluster in [/docker-compose/docker-compose.yml](/docker-compose/docker-compose.yml). It creates the following conatainers.
+
+- etcd x 1
+- pageserver x 1
+- safekeeper x 3
+- compute x 1
+- MinIO x 1        # This is Amazon S3 compatible object storage
+
+### How to use
+
+1. create containers
+
+You can specify version of neon cluster using following environment values.
+- PG_VERSION: postgres version for compute (default is 14)
+- TAG: the tag version of [docker image](https://registry.hub.docker.com/r/neondatabase/neon/tags) (default is latest), which is tagged in [CI test](/.github/workflows/build_and_test.yml)
+```
+$ cd docker-compose/docker-compose.yml
+$ docker-compose down   # remove the conainers if exists
+$ PG_VERSION=15 TAG=2221 docker-compose up --build -d  # You can specify the postgres and image version
+Creating network "dockercompose_default" with the default driver
+Creating dockercompose_etcd3_1 ...
+(...omit...)
+```
+
+2. connect compute node
+```
+$ echo "localhost:55433:postgres:cloud_admin:cloud_admin" >> ~/.pgpass
+$ psql -h localhost -p 55433 -U cloud_admin
+postgres=# CREATE TABLE t(key int primary key, value text);
+CREATE TABLE
+postgres=# insert into t values(1,1);
+INSERT 0 1
+postgres=# select * from t;
+ key | value
+-----+-------
+   1 | 1
+(1 row)
+```
+
+3. If you want to see the log, you can use `docker-compose logs` command.
+```
+# check the container name you want to see
+$ docker ps
+CONTAINER ID   IMAGE                                              COMMAND                  CREATED         STATUS         PORTS                                                                                                                                  NAMES
+d6968a5ae912   dockercompose_compute                              "/shell/compute.sh"      5 minutes ago   Up 5 minutes   0.0.0.0:3080->3080/tcp, 0.0.0.0:55433->55433/tcp                                                                                       dockercompose_compute_1
+(...omit...)
+
+$ docker logs -f dockercompose_compute_1
+2022-10-21 06:15:48.757 GMT [56] LOG:  connection authorized: user=cloud_admin database=postgres application_name=psql
+2022-10-21 06:17:00.307 GMT [56] LOG:  [NEON_SMGR] libpagestore: connected to 'host=pageserver port=6400'
+(...omit...)
+```
+
+4. If you want to see durable data in MinIO which is s3 compatible storage
+
+Access http://localhost:9001 and sign in.
+
+- Username: `minio`
+- Password: `password`
+
+You can see durable pages and WAL data in `neon` bucket.

docs/error-handling.md

@@ -0,0 +1,198 @@
+# Error handling and logging
+
+## Logging errors
+
+The principle is that errors are logged when they are handled. If you
+just propagate an error to the caller in a function, you don't need to
+log it; the caller will. But if you consume an error in a function,
+you *must* log it (if it needs to be logged at all).
+
+For example:
+
+```rust
+fn read_motd_file() -> std::io::Result<String> {
+    let mut f = File::open("/etc/motd")?;
+    let mut result = String::new();
+    f.read_to_string(&mut result)?;
+    result
+}
+```
+
+Opening or reading the file could fail, but there is no need to log
+the error here. The function merely propagates the error to the
+caller, and it is up to the caller to log the error or propagate it
+further, if the failure is not expected. But if, for example, it is
+normal that the "/etc/motd" file doesn't exist, the caller can choose
+to silently ignore the error, or log it as an INFO or DEBUG level
+message:
+
+```rust
+fn get_message_of_the_day() -> String {
+    // Get the motd from /etc/motd, or return the default proverb
+    match read_motd_file() {
+        Ok(motd) => motd,
+        Err(err)  => {
+            // It's normal that /etc/motd doesn't exist, but if we fail to
+            // read it for some other reason, that's unexpected. The message
+            // of the day isn't very important though, so we just WARN and
+            // continue with the default in any case.
+            if err.kind() != std::io::ErrorKind::NotFound {
+                 tracing::warn!("could not read \"/etc/motd\": {err:?}");
+            }
+            "An old error is always more popular than a new truth. - German proverb"
+        }
+    }
+}
+```
+
+## Error types
+
+We use the `anyhow` crate widely. It contains many convenient macros
+like `bail!` and `ensure!` to construct and return errors, and to
+propagate many kinds of low-level errors, wrapped in `anyhow::Error`.
+
+A downside of `anyhow::Error` is that the caller cannot distinguish
+between different error cases. Most errors are propagated all the way
+to the mgmt API handler function, or the main loop that handles a
+connection with the compute node, and they are all handled the same
+way: the error is logged and returned to the client as an HTTP or
+libpq error.
+
+But in some cases, we need to distinguish between errors and handle
+them differently. For example, attaching a tenant to the pageserver
+could fail either because the tenant has already been attached, or
+because we could not load its metadata from cloud storage. The first
+case is more or less expected. The console sends the Attach request to
+the pageserver, and the pageserver completes the operation, but the
+network connection might be lost before the console receives the
+response. The console will retry the operation in that case, but the
+tenant has already been attached. It is important that the pagserver
+responds with the HTTP 403 Already Exists error in that case, rather
+than a generic HTTP 500 Internal Server Error.
+
+If you need to distinguish between different kinds of errors, create a
+new `Error` type. The `thiserror` crate is useful for that. But in
+most cases `anyhow::Error` is good enough.
+
+## Panics
+
+Depending on where a panic happens, it can cause the whole pageserver
+or safekeeper to restart, or just a single tenant. In either case,
+that is pretty bad and causes an outage. Avoid panics. Never use
+`unwrap()` or other calls that might panic, to verify inputs from the
+network or from disk.
+
+It is acceptable to use functions that might panic, like `unwrap()`, if
+it is obvious that it cannot panic. For example, if you have just
+checked that a variable is not None, it is OK to call `unwrap()` on it,
+but it is still preferable to use `expect("reason")` instead to explain
+why the function cannot fail.
+
+`assert!` and `panic!` are reserved for checking clear invariants and
+very obvious "can't happen" cases. When in doubt, use anyhow `ensure!`
+or `bail!` instead.
+
+## Error levels
+
+`tracing::Level` doesn't provide very clear guidelines on what the
+different levels mean, or when to use which level. Here is how we use
+them:
+
+### Error
+
+Examples:
+- could not open file "foobar"
+- invalid tenant id
+
+Errors are not expected to happen during normal operation. Incorrect
+inputs from client can cause ERRORs. For example, if a client tries to
+call a mgmt API that doesn't exist, or if a compute node sends passes
+an LSN that has already been garbage collected away.
+
+These should *not* happen during normal operations. "Normal
+operations" is not a very precise concept. But for example, disk
+errors are not expected to happen when the system is working, so those
+count as Errors. However, if a TCP connection to a compute node is
+lost, that is not considered an Error, because it doesn't affect the
+pageserver's or safekeeper's operation in any way, and happens fairly
+frequently when compute nodes are shut down, or are killed abruptly
+because of errors in the compute.
+
+**Errors are monitored, and always need human investigation to determine
+the cause.**
+
+Whether something should be logged at ERROR, WARNING or INFO level can
+depend on the callers and clients. For example, it might be unexpected
+and a sign of a serious issue if the console calls the
+"timeline_detail" mgmt API for a timeline that doesn't exist. ERROR
+would be appropriate in that case. But if the console routinely calls
+the API after deleting a timeline, to check if the deletion has
+completed, then it would be totally normal and an INFO or DEBUG level
+message would be more appropriate. If a message is logged as an ERROR,
+but it in fact happens frequently in production and never requires any
+action, it should probably be demoted to an INFO level message.
+
+### Warn
+
+Examples:
+- could not remove temporary file "foobar.temp"
+- unrecognized file "foobar" in timeline directory
+
+Warnings are similar to Errors, in that they should not happen
+when the system is operating normally. The difference between Error and
+Warning is that an Error means that the operation failed, whereas Warning
+means that something unexpected happened, but the operation continued anyway.
+For example, if deleting a file fails because the file already didn't exist,
+it should be logged as Warning.
+
+> **Note:** The python regression tests, under `test_regress`, check the
+> pageserver log after each test for any ERROR and WARN lines. If there are
+> any ERRORs or WARNs that have not been explicitly listed in the test as
+> allowed, the test is marked a failed. This is to catch unexpected errors
+> e.g. in background operations, that don't cause immediate misbehaviour in
+> the tested functionality.
+
+### Info
+
+Info level is used to log useful information when the system is
+operating normally. Info level is appropriate e.g. for logging state
+changes, background operations, and network connections.
+
+Examples:
+- "system is shutting down"
+- "tenant was created"
+- "retrying S3 upload"
+
+### Debug & Trace
+
+Debug and Trace level messages are not printed to the log in our normal
+production configuration, but could be enabled for a specific server or
+tenant, to aid debugging. (Although we don't actually have that
+capability as of this writing).
+
+## Context
+
+We use logging "spans" to hold context information about the current
+operation. Almost every operation happens on a particular tenant and
+timeline, so we enter a span with the "tenant_id" and "timeline_id"
+very early when processing an incoming API request, for example. All
+background operations should also run in a span containing at least
+those two fields, and any other parameters or information that might
+be useful when debugging an error that might happen when performing
+the operation.
+
+TODO: Spans are not captured in the Error when it is created, but when
+the error is logged. It would be more useful to capture them at Error
+creation. We should consider using `tracing_error::SpanTrace` to do
+that.
+
+## Error message style
+
+PostgreSQL has a style guide for writing error messages:
+
+https://www.postgresql.org/docs/current/error-style-guide.html
+
+Follow that guide when writing error messages in the PostgreSQL
+extension. We don't follow it strictly in the pageserver and
+safekeeper, but the advice in the PostgreSQL style guide is generally
+good, and you can't go wrong by following it.

docs/rfcs/018-storage-messaging-2.md

@@ -0,0 +1,163 @@
+# Storage messaging
+
+Safekeepers need to communicate to each other to
+* Trim WAL on safekeepers;
+* Decide on which SK should push WAL to the S3;
+* Decide on when to shut down SK<->pageserver connection;
+* Understand state of each other to perform peer recovery;
+
+Pageservers need to communicate to safekeepers to decide which SK should provide
+WAL to the pageserver.
+
+This is an iteration on [015-storage-messaging](https://github.com/neondatabase/neon/blob/main/docs/rfcs/015-storage-messaging.md) describing current situation,
+potential performance issue and ways to address it.
+
+## Background
+
+What we have currently is very close to etcd variant described in
+015-storage-messaging. Basically, we have single `SkTimelineInfo` message
+periodically sent by all safekeepers to etcd for each timeline.
+* Safekeepers subscribe to it to learn status of peers (currently they subscribe to
+  'everything', but they can and should fetch data only for timelines they hold).
+* Pageserver subscribes to it (separate watch per timeline) to learn safekeepers
+  positions; based on that, it decides from which safekeepers to pull WAL.
+
+Also, safekeepers use etcd elections API to make sure only single safekeeper
+offloads WAL.
+
+It works, and callmemaybe is gone. However, this has a performance
+hazard. Currently deployed etcd can do about 6k puts per second (using its own
+`benchmark` tool); on my 6 core laptop, while running on tmpfs, this gets to
+35k. Making benchmark closer to our usage [etcd watch bench](https://github.com/arssher/etcd-client/blob/watch-bench/examples/watch_bench.rs),
+I get ~10k received messages per second with various number of publisher-subscribers
+(laptop, tmpfs). Diving this by 12 (3 sks generate msg, 1 ps + 3 sk consume them) we
+get about 800 active timelines, if message is sent each second. Not extremely
+low, but quite reachable.
+
+A lot of idle watches seem to be ok though -- which is good, as pageserver
+subscribes to all its timelines regardless of their activity.
+
+Also, running etcd with fsyncs disabled is messy -- data dir must be wiped on
+each restart or there is a risk of corruption errors.
+
+The reason is etcd making much more than what we need; it is a fault tolerant
+store with strong consistency, but I claim all we need here is just simplest pub
+sub with best effort delivery, because
+* We already have centralized source of truth for long running data, like which
+  tlis are on which nodes  -- the console.
+* Momentary data (safekeeper/pageserver progress) doesn't make sense to persist.
+  Instead of putting each change to broker, expecting it to reliably deliver it
+  is better to just have constant flow of data for active timelines: 1) they
+  serve as natural heartbeats -- if node can't send, we shouldn't pull WAL from
+  it 2) it is simpler -- no need to track delivery to/from the broker.
+  Moreover, latency here is important: the faster we obtain fresh data, the
+  faster we can switch to proper safekeeper after failure.
+* As for WAL offloading leader election, it is trivial to achieve through these
+  heartbeats -- just take suitable node through deterministic rule (min node
+  id).  Once network is stable, this is a converging process (well, except
+  complicated failure topology, but even then making it converge is not
+  hard). Such elections bear some risk of several offloaders running
+  concurrently for a short period of time, but that's harmless.
+
+  Generally, if one needs strong consistency, electing leader per se is not
+  enough; it must be accompanied with number (logical clock ts), checked at
+  every action to track causality. s3 doesn't provide CAS, so it can't
+  differentiate old/new leader, this must be solved differently.
+
+  We could use etcd CAS (its most powerful/useful primitive actually) to issue
+  these leader numbers (and e.g. prefix files in s3), but currently I don't see
+  need for that.
+
+
+Obviously best effort pub sub is much more simpler and performant; the one proposed is
+
+## gRPC broker
+
+I took tonic and [prototyped](https://github.com/neondatabase/neon/blob/asher/neon-broker/broker/src/broker.rs) the replacement of functionality we currently use
+with grpc streams and tokio mpsc channels. The implementation description is at the file header.
+
+It is just 500 lines of code and core functionality is complete. 1-1 pub sub
+gives about 120k received messages per second; having multiple subscribers in
+different connecitons quickly scales to 1 million received messages per second.
+I had concerns about many concurrent streams in singe connection, but 2^20
+subscribers still work (though eat memory, with 10 publishers 20GB are consumed;
+in this implementation each publisher holds full copy of all subscribers). There
+is `bench.rs` nearby which I used for testing.
+
+`SkTimelineInfo` is wired here, but another message can be added (e.g. if
+pageservers want to communicate with each other) with templating.
+
+### Fault tolerance
+
+Since such broker is stateless, we can run it under k8s. Or add proxying to
+other members, with best-effort this is simple.
+
+### Security implications
+
+Communication happens in a private network that is not exposed to users;
+additionaly we can add auth to the broker.
+
+## Alternative: get existing pub-sub
+
+We could take some existing pub sub solution, e.g. RabbitMQ, Redis. But in this
+case IMV simplicity of our own outweights external dependency costs (RabbitMQ is
+much more complicated and needs VM; Redis Rust client maintenance is not
+ideal...). Also note that projects like CockroachDB and TiDB are based on gRPC
+as well.
+
+## Alternative: direct communication
+
+Apart from being transport, broker solves one more task: discovery, i.e. letting
+safekeepers and pageservers find each other. We can let safekeepers know, for
+each timeline, both other safekeepers for this timeline and pageservers serving
+it. In this case direct communication is possible:
+ - each safekeeper pushes to each other safekeeper status of timelines residing
+   on both of them, letting remove WAL, decide who offloads, decide on peer
+   recovery;
+ - each safekeeper pushes to each pageserver status of timelines residing on
+   both of them, letting pageserver choose from which sk to pull WAL;
+
+It was mostly described in [014-safekeeper-gossip](https://github.com/neondatabase/neon/blob/main/docs/rfcs/014-safekeepers-gossip.md), but I want to recap on that.
+
+The main pro is less one dependency: less moving parts, easier to run Neon
+locally/manually, less places to monitor. Fault tolerance for broker disappears,
+no kuber or something. To me this is a big thing.
+
+Also (though not a big thing) idle watches for inactive timelines disappear:
+naturally safekeepers learn about compute connection first and start pushing
+status to pageserver(s), notifying it should pull.
+
+Importantly, I think that eventually knowing and persisting peers and
+pageservers on safekeepers is inevitable:
+- Knowing peer safekeepers for the timeline is required for correct
+  automatic membership change -- new member set must be hardened on old
+  majority before proceeding. It is required to get rid of sync-safekeepers
+  as well (peer recovery up to flush_lsn).
+- Knowing pageservers where the timeline is attached is needed to
+  1. Understand when to shut down activity on the timeline, i.e. push data to
+     the broker. We can have a lot of timelines sleeping quietly which
+	 shouldn't occupy resources.
+  2. Preserve WAL for these (currently we offload to s3 and take it from there,
+     but serving locally is better, and we get one less condition on which WAL
+     can be removed from s3).
+
+I suppose this membership data should be passed to safekeepers directly from the
+console because
+1. Console is the original source of this data, conceptually this is the
+   simplest way (rather than passing it through compute or something).
+2. We already have similar code for deleting timeline on safekeepers
+   (and attaching/detaching timeline on pageserver), this is a typical
+   action -- queue operation against storage node and execute it until it
+   completes (or timeline is dropped).
+
+Cons of direct communication are
+- It is more complicated: each safekeeper should maintain set of peers it talks
+  to, and set of timelines for each such peer -- they ought to be multiplexed
+  into single connection.
+- Totally, we have O(n^2) connections instead of O(n) with broker schema
+  (still O(n) on each node). However, these are relatively stable, async and
+  thus not very expensive, I don't think this is a big problem. Up to 10k
+  storage nodes I doubt connection overhead would be noticeable.
+
+I'd use gRPC for direct communication, and in this sense gRPC based broker is a
+step towards it.

docs/rfcs/019-tenant-timeline-lifecycles.md

@@ -0,0 +1,91 @@
+# Managing Tenant and Timeline lifecycles
+
+## Summary
+
+The pageserver has a Tenant object in memory for each tenant it manages, and a
+Timeline for each timeline. There are a lot of tasks that operate on the tenants
+and timelines with references to those objects. We have some mechanisms to track
+which tasks are operating on each Tenant and Timeline, and to request them to
+shutdown when a tenant or timeline is deleted, but it does not cover all uses,
+and as a result we have many race conditions around tenant/timeline shutdown.
+
+## Motivation
+
+We have a bunch of race conditions that can produce weird errors and can be hard
+to track down.
+
+## Non Goals
+
+This RFC only covers the problem of ensuring that a task/thread isn't operating
+on a Tenant or Timeline. It does not cover what states, aside from Active and
+non-Active, each Tenant and Timeline should have, or when exactly the transitions
+should happen.
+
+## Impacted components (e.g. pageserver, safekeeper, console, etc)
+
+Pageserver. Although I wonder if the safekeeper should have a similar mechanism.
+
+## Current situation
+
+Most pageserver tasks of are managed by task_mgr.rs:
+
+- LibpqEndpointListener
+- HttpEndPointListener
+- WalReceiverManager and -Connection
+- GarbageCollector and Compaction
+- InitialLogicalSizeCalculation
+
+In addition to those tasks, the walreceiver performs some direct tokio::spawn
+calls to spawn tasks that are not registered with 'task_mgr'. And all of these
+tasks can spawn extra operations with tokio spawn_blocking.
+
+Whenever a tenant or timeline is removed from the system, by pageserver
+shutdown, delete_timeline or tenant-detach operation, we rely on the task
+registry in 'task_mgr.rs' to wait until there are no tasks operating on the
+tenant or timeline, before its Tenant/Timeline object is removed. That relies on
+each task to register itself with the tenant/timeline ID in
+'task_mgr.rs'. However, there are many gaps in that. For example,
+GarbageCollection and Compaction tasks are registered with the tenant, but when
+they proceed to operate on a particular timeline of the tenant, they don't
+register with timeline ID. Because of that, the timeline can be deleted while GC
+or compaction is running on it, causing failures in the GC or compaction (see
+https://github.com/neondatabase/neon/issues/2442).
+
+Another problem is that the task registry only works for tokio Tasks. There is
+no way to register a piece of code that runs inside spawn_blocking(), for
+example.
+
+## Proposed implementation
+
+This "voluntary" registration of tasks is fragile. Let's use Rust language features
+to enforce that a tenant/timeline cannot be removed from the system when there is
+still some code operating on it.
+
+Let's introduce new Guard objects for Tenant and Timeline, and do all actions through
+the Guard object. Something like:
+
+TenantActiveGuard: Guard object over Arc<Tenant>. When you acquire the guard,
+the code checks that the tenant is in Active state. If it's not, you get an
+error. You can change the state of the tenant to Stopping while there are
+ActiveTenantGuard objects still on it, to prevent new ActiveTenantGuards from
+being acquired, but the Tenant cannot be removed until all the guards are gone.
+
+TenantMaintenanceGuard: Like ActiveTenantGuard, but can be held even when the
+tenant is not in Active state. Used for operations like attach/detach. Perhaps
+allow only one such guard on a Tenant at a time.
+
+Similarly for Timelines. We don't currentl have a "state" on Timeline, but I think
+we need at least two states: Active and Stopping. The Stopping state is used at
+deletion, to prevent new TimelineActiveGuards from appearing, while you wait for
+existing TimelineActiveGuards to die out.
+
+The shutdown-signaling, using shutdown_watcher() and is_shutdown_requested(),
+probably also needs changes to deal with the new Guards. The rule is that if you
+have a TenantActiveGuard, and the tenant's state changes from Active to
+Stopping, the is_shutdown_requested() function should return true, and
+shutdown_watcher() future should return.
+
+This signaling doesn't neessarily need to cover all cases. For example, if you
+have a block of code in spawn_blocking(), it might be acceptable if
+is_shutdown_requested() doesn't return true even though the tenant is in
+Stopping state, as long as the code finishes reasonably fast.

docs/rfcs/020-pageserver-s3-coordination.md

@@ -0,0 +1,246 @@
+# Coordinating access of multiple pageservers to the same s3 data
+
+## Motivation
+
+There are some blind spots around coordinating access of multiple pageservers
+to the same s3 data. Currently this is applicable only to tenant relocation
+case, but in the future we'll need to solve similar problems for
+replica/standby pageservers.
+
+## Impacted components (e.g. pageserver, safekeeper, console, etc)
+
+Pageserver
+
+## The problem
+
+### Relocation
+
+During relocation both pageservers can write to s3. This should be ok for all
+data except the `index_part.json`. For index part it causes problems during
+compaction/gc because they remove files from index/s3.
+
+Imagine this case:
+
+```mermaid
+sequenceDiagram
+    autonumber
+    participant PS1
+    participant S3
+    participant PS2
+
+    PS1->>S3: Uploads L1, L2 <br/> Index contains L1 L2
+    PS2->>S3: Attach called, sees L1, L2
+    PS1->>S3: Compaction comes <br/> Removes L1, adds L3
+    note over S3: Index now L2, L3
+    PS2->>S3: Uploads new layer L4 <br/> (added to previous view of the index)
+    note over S3: Index now L1, L2, L4
+```
+
+At this point it is not possible to restore from index, it contains L2 which
+is no longer available in s3 and doesnt contain L3 added by compaction by the
+first pageserver. So if any of the pageservers restart initial sync will fail
+(or in on-demand world it will fail a bit later during page request from
+missing layer)
+
+### Standby pageserver
+
+Another related case is standby pageserver. In this case second pageserver can
+be used as a replica to scale reads and serve as a failover target in case
+first one fails.
+
+In this mode second pageserver needs to have the same picture of s3 files to
+be able to load layers on-demand. To accomplish that second pageserver
+cannot run gc/compaction jobs. Instead it needs to receive updates for index
+contents. (There is no need to run walreceiver on the second pageserver then).
+
+## Observations
+
+- If both pageservers ingest wal then their layer set diverges, because layer
+  file generation is not deterministic
+- If one of the pageservers does not ingest wal (and just picks up layer
+  updates) then it lags behind and cannot really answer queries in the same
+  pace as the primary one
+- Can compaction help make layers deterministic? E g we do not upload level
+  zero layers and construction of higher levels should be deterministic.
+  This way we can guarantee that layer creation by timeout wont mess things up.
+  This way one pageserver uploads data and second one can just ingest it.
+  But we still need some form of election
+
+## Solutions
+
+### Manual orchestration
+
+One possible solution for relocation case is to orchestrate background jobs
+from outside. The oracle who runs migration can turn off background jobs on
+PS1 before migration and then run migration -> enable them on PS2. The problem
+comes if migration fails. In this case in order to resume background jobs
+oracle needs to guarantee that PS2 doesnt run background jobs and if it doesnt
+respond then PS1 is stuck unable to run compaction/gc. This cannot be solved
+without human ensuring that no upload from PS2 can happen. In order to be able
+to resolve this automatically CAS is required on S3 side so pageserver can
+avoid overwriting index part if it is no longer the leading one
+
+Note that flag that disables background jobs needs to be persistent, because
+otherwise pageserver restart will clean it
+
+### Avoid index_part.json
+
+Index part consists of two parts, list of layers and metadata. List of layers
+can be easily obtained by `ListObjects` S3 API method. But what to do with
+metadata? Create metadata instance for each checkpoint and add some counter
+to the file name?
+
+Back to potentially long s3 ls.
+
+### Coordination based approach
+
+Do it like safekeepers chose leader for WAL upload. Ping each other and decide
+based on some heuristics e g smallest node id. During relocation PS1 sends
+"resign" ping message so others can start election without waiting for a timeout.
+
+This still leaves metadata question open and non deterministic layers are a
+problem as well
+
+### Avoid metadata file
+
+One way to eliminate metadata file is to store it in layer files under some
+special key. This may resonate with intention to keep all relation sizes in
+some special segment to avoid initial download during size calculation.
+Maybe with that we can even store pre calculated value.
+
+As a downside each checkpoint gets 512 bytes larger.
+
+If we entirely avoid metadata file this opens up many approaches
+
+* * *
+
+During discussion it seems that we converged on the approach consisting of:
+
+- index files stored per pageserver in the same timeline directory. With that
+  index file name starts to look like: `<pageserver_node_id>_index_part.json`.
+  In such set up there are no concurrent overwrites of index file by different
+  pageservers.
+- For replica pageservers the solution would be for primary to broadcast index
+  changes to any followers with an ability to check index files in s3 and
+  restore the full state. To properly merge changes with index files we can use
+  a counter that is persisted in an index file, is incremented on every change
+  to it and passed along with broadcasted change. This way we can determine
+  whether we need to apply change to the index state or not.
+- Responsibility for running background jobs is assigned externally. Pageserver
+  keeps locally persistent flag for each tenant that indicates whether this
+  pageserver is considered as primary one or not. TODO what happends if we
+  crash and cannot start for some extended period of time? Control plane can
+  assign ownership to some other pageserver. Pageserver needs some way to check
+  if its still the blessed one. Maybe by explicit request to control plane on
+  start.
+
+Requirement for deterministic layer generation was considered overly strict
+because of two reasons:
+
+- It can limit possible optimizations e g when pageserver wants to reshuffle
+  some data locally and doesnt want to coordinate this
+- The deterministic algorithm itself can change so during deployments for some
+  time there will be two different version running at the same time which can
+  cause non determinism
+
+### External elections
+
+The above case with lost state in this schema with externally managed
+leadership is represented like this:
+
+Note that here we keep objects list in the index file.
+
+```mermaid
+sequenceDiagram
+    autonumber
+    participant PS1
+    participant CP as Control Plane
+    participant S3
+    participant PS2
+
+    note over PS1,PS2: PS1 starts up and still a leader
+    PS1->>CP: Am I still the leader for Tenant X?
+    activate CP
+    CP->>PS1: Yes
+    deactivate CP
+    PS1->>S3: Fetch PS1 index.
+    note over PS1: Continue operations, start backround jobs
+    note over PS1,PS2: PS1 starts up and still and is not a leader anymore
+    PS1->>CP: Am I still the leader for Tenant X?
+    CP->>PS1: No
+    PS1->>PS2: Subscribe to index changes
+    PS1->>S3: Fetch PS1 and PS2 indexes
+    note over PS1: Combine index file to include layers <br> from both indexes to be able <br> to see newer files from leader (PS2)
+    note over PS1: Continue operations, do not start background jobs
+```
+
+### Internal elections
+
+To manage leadership internally we can use broker to exchange pings so nodes
+can decide on the leader roles. In case multiple pageservers are active leader
+is the one with lowest node id.
+
+Operations with internally managed elections:
+
+```mermaid
+sequenceDiagram
+    autonumber
+    participant PS1
+    participant S3
+
+    note over PS1: Starts up
+    note over PS1: Subscribes to changes, waits for two ping <br> timeouts to see if there is a leader
+    PS1->>S3: Fetch indexes from s3
+    alt there is a leader
+        note over PS1: do not start background jobs, <br> continue applying index updates
+    else there is no leader
+        note over PS1: start background jobs, <br> broadcast index changes
+    end
+
+    note over PS1,S3: Then the picture is similar to external elections <br> the difference is that follower can become a leader <br> if there are no pings after some timeout new leader gets elected
+```
+
+### Eviction
+
+When two pageservers operate on a tenant for extended period of time follower
+doesnt perform write operations in s3. When layer is evicted follower relies
+on updates from primary to get info about layers it needs to cover range for
+evicted layer.
+
+Note that it wont match evicted layer exactly, so layers will overlap and
+lookup code needs to correctly handle that.
+
+### Relocation flow
+
+Actions become:
+
+- Attach tenant to new pageserver
+- New pageserver becomes follower since previous one is still leading
+- New pageserver starts replicating from safekeepers but does not upload layers
+- Detach is called on the old one
+- New pageserver becomes leader after it realizes that old one disappeared
+
+### Index File
+
+Using `s3 ls` on startup simplifies things, but we still need metadata, so we
+need to fetch index files anyway. If they contain list of files we can combine
+them and avoid costly `s3 ls`
+
+### Remaining issues
+
+- More than one remote consistent lsn for safekeepers to know
+
+Anything else?
+
+### Proposed solution
+
+To recap. On meeting we converged on approach with external elections but I
+think it will be overall harder to manage and will introduce a dependency on
+control plane for pageserver. Using separate index files for each pageserver
+consisting of log of operations and a metadata snapshot should be enough.
+
+### What we need to get there?
+
+- Change index file structure to contain log of changes instead of just the
+  file list
+- Implement pinging/elections for pageservers

docs/sourcetree.md

@@ -52,6 +52,10 @@ PostgreSQL extension that implements storage manager API and network communicati
 
 PostgreSQL extension that contains functions needed for testing and debugging.
 
+`/pgxn/neon_walredo`:
+
+Library to run Postgres as a "WAL redo process" in the pageserver.
+
 `/safekeeper`:
 
 The neon WAL service that receives WAL from a primary compute nodes and streams it to the pageserver.
@@ -79,6 +83,16 @@ A subject for future modularization.
 `/libs/metrics`:
 Helpers for exposing Prometheus metrics from the server.
 
+### Adding dependencies
+When you add a Cargo dependency, you should update hakari manifest by running commands below and committing the updated `Cargo.lock` and `workspace_hack/`. There may be no changes, that's fine.
+
+```bash
+cargo hakari generate
+cargo hakari manage-deps
+```
+
+If you don't have hakari installed (`error: no such subcommand: hakari`), install it by running `cargo install cargo-hakari`.
+
 ## Using Python
 Note that Debian/Ubuntu Python packages are stale, as it commonly happens,
 so manual installation of dependencies is not recommended.

libs/etcd_broker/Cargo.toml

@@ -8,7 +8,7 @@
  regex = "1.4.5"
  serde = { version = "1.0", features = ["derive"] }
  serde_json = "1"
- serde_with = "1.12.0"
+ serde_with = "2.0"
  once_cell = "1.13.0"
 
  utils = { path = "../utils" }

libs/etcd_broker/src/subscription_value.rs

@@ -29,6 +29,9 @@ pub struct SkTimelineInfo {
     #[serde_as(as = "Option<DisplayFromStr>")]
     #[serde(default)]
     pub peer_horizon_lsn: Option<Lsn>,
+    #[serde_as(as = "Option<DisplayFromStr>")]
+    #[serde(default)]
+    pub local_start_lsn: Option<Lsn>,
     /// A connection string to use for WAL receiving.
     #[serde(default)]
     pub safekeeper_connstr: Option<String>,

libs/metrics/src/lib.rs

@@ -3,7 +3,7 @@
 //! Otherwise, we might not see all metrics registered via
 //! a default registry.
 use once_cell::sync::Lazy;
-use prometheus::core::{AtomicU64, GenericGauge, GenericGaugeVec};
+use prometheus::core::{AtomicU64, Collector, GenericGauge, GenericGaugeVec};
 pub use prometheus::opts;
 pub use prometheus::register;
 pub use prometheus::{core, default_registry, proto};
@@ -17,6 +17,7 @@ pub use prometheus::{register_int_counter_vec, IntCounterVec};
 pub use prometheus::{register_int_gauge, IntGauge};
 pub use prometheus::{register_int_gauge_vec, IntGaugeVec};
 pub use prometheus::{Encoder, TextEncoder};
+use prometheus::{Registry, Result};
 
 mod wrappers;
 pub use wrappers::{CountedReader, CountedWriter};
@@ -32,13 +33,27 @@ macro_rules! register_uint_gauge_vec {
     }};
 }
 
+/// Special internal registry, to collect metrics independently from the default registry.
+/// Was introduced to fix deadlock with lazy registration of metrics in the default registry.
+static INTERNAL_REGISTRY: Lazy<Registry> = Lazy::new(Registry::new);
+
+/// Register a collector in the internal registry. MUST be called before the first call to `gather()`.
+/// Otherwise, we can have a deadlock in the `gather()` call, trying to register a new collector
+/// while holding the lock.
+pub fn register_internal(c: Box<dyn Collector>) -> Result<()> {
+    INTERNAL_REGISTRY.register(c)
+}
+
 /// Gathers all Prometheus metrics and records the I/O stats just before that.
 ///
 /// Metrics gathering is a relatively simple and standalone operation, so
 /// it might be fine to do it this way to keep things simple.
 pub fn gather() -> Vec<prometheus::proto::MetricFamily> {
     update_rusage_metrics();
-    prometheus::gather()
+    let mut mfs = prometheus::gather();
+    let mut internal_mfs = INTERNAL_REGISTRY.gather();
+    mfs.append(&mut internal_mfs);
+    mfs
 }
 
 static DISK_IO_BYTES: Lazy<IntGaugeVec> = Lazy::new(|| {
@@ -62,6 +77,16 @@ pub const DISK_WRITE_SECONDS_BUCKETS: &[f64] = &[
     0.000_050, 0.000_100, 0.000_500, 0.001, 0.003, 0.005, 0.01, 0.05, 0.1, 0.3, 0.5,
 ];
 
+pub fn set_build_info_metric(revision: &str) {
+    let metric = register_int_gauge_vec!(
+        "libmetrics_build_info",
+        "Build/version information",
+        &["revision"]
+    )
+    .expect("Failed to register build info metric");
+    metric.with_label_values(&[revision]).set(1);
+}
+
 // Records I/O stats in a "cross-platform" way.
 // Compiles both on macOS and Linux, but current macOS implementation always returns 0 as values for I/O stats.
 // An alternative is to read procfs (`/proc/[pid]/io`) which does not work under macOS at all, hence abandoned.

libs/pageserver_api/Cargo.toml

@@ -5,8 +5,12 @@ edition = "2021"
 
 [dependencies]
 serde = { version = "1.0", features = ["derive"] }
-serde_with = "1.12.0"
+serde_with = "2.0"
 const_format = "0.2.21"
+anyhow = { version = "1.0", features = ["backtrace"] }
+bytes = "1.0.1"
+byteorder = "1.4.3"
 
 utils = { path = "../utils" }
+postgres_ffi = { path = "../postgres_ffi" }
 workspace_hack = { version = "0.1", path = "../../workspace_hack" }

libs/pageserver_api/src/lib.rs

@@ -2,6 +2,7 @@ use const_format::formatcp;
 
 /// Public API types
 pub mod models;
+pub mod reltag;
 
 pub const DEFAULT_PG_LISTEN_PORT: u16 = 64000;
 pub const DEFAULT_PG_LISTEN_ADDR: &str = formatcp!("127.0.0.1:{DEFAULT_PG_LISTEN_PORT}");

libs/pageserver_api/src/models.rs

@@ -1,5 +1,6 @@
 use std::num::NonZeroU64;
 
+use byteorder::{BigEndian, ReadBytesExt};
 use serde::{Deserialize, Serialize};
 use serde_with::{serde_as, DisplayFromStr};
 use utils::{
@@ -7,6 +8,10 @@ use utils::{
     lsn::Lsn,
 };
 
+use crate::reltag::RelTag;
+use anyhow::bail;
+use bytes::{BufMut, Bytes, BytesMut};
+
 /// A state of a tenant in pageserver's memory.
 #[derive(Debug, Clone, Copy, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
 pub enum TenantState {
@@ -19,6 +24,22 @@ pub enum TenantState {
     Broken,
 }
 
+/// A state of a timeline in pageserver's memory.
+#[derive(Debug, Clone, Copy, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
+pub enum TimelineState {
+    /// Timeline is fully operational, its background jobs are running.
+    Active,
+    /// A timeline is recognized by pageserver, but not yet ready to operate.
+    /// The status indicates, that the timeline could eventually go back to Active automatically:
+    /// for example, if the owning tenant goes back to Active again.
+    Suspended,
+    /// A timeline is recognized by pageserver, but not yet ready to operate and not allowed to
+    /// automatically become Active after certain events: only a management call can change this status.
+    Paused,
+    /// A timeline is recognized by the pageserver, but no longer used for any operations, as failed to get activated.
+    Broken,
+}
+
 #[serde_as]
 #[derive(Serialize, Deserialize)]
 pub struct TimelineCreateRequest {
@@ -52,6 +73,7 @@ pub struct TenantCreateRequest {
     pub walreceiver_connect_timeout: Option<String>,
     pub lagging_wal_timeout: Option<String>,
     pub max_lsn_wal_lag: Option<NonZeroU64>,
+    pub trace_read_requests: Option<bool>,
 }
 
 #[serde_as]
@@ -91,6 +113,7 @@ pub struct TenantConfigRequest {
     pub walreceiver_connect_timeout: Option<String>,
     pub lagging_wal_timeout: Option<String>,
     pub max_lsn_wal_lag: Option<NonZeroU64>,
+    pub trace_read_requests: Option<bool>,
 }
 
 impl TenantConfigRequest {
@@ -109,6 +132,7 @@ impl TenantConfigRequest {
             walreceiver_connect_timeout: None,
             lagging_wal_timeout: None,
             max_lsn_wal_lag: None,
+            trace_read_requests: None,
         }
     }
 }
@@ -123,9 +147,15 @@ pub struct TenantInfo {
     pub has_in_progress_downloads: Option<bool>,
 }
 
+/// This represents the output of the "timeline_detail" and "timeline_list" API calls.
 #[serde_as]
 #[derive(Debug, Serialize, Deserialize, Clone)]
-pub struct LocalTimelineInfo {
+pub struct TimelineInfo {
+    #[serde_as(as = "DisplayFromStr")]
+    pub tenant_id: TenantId,
+    #[serde_as(as = "DisplayFromStr")]
+    pub timeline_id: TimelineId,
+
     #[serde_as(as = "Option<DisplayFromStr>")]
     pub ancestor_timeline_id: Option<TimelineId>,
     #[serde_as(as = "Option<DisplayFromStr>")]
@@ -149,28 +179,35 @@ pub struct LocalTimelineInfo {
     /// the timestamp (in microseconds) of the last received message
     pub last_received_msg_ts: Option<u128>,
     pub pg_version: u32,
+
+    #[serde_as(as = "Option<DisplayFromStr>")]
+    pub remote_consistent_lsn: Option<Lsn>,
+    pub awaits_download: bool,
+
+    pub state: TimelineState,
+
+    // Some of the above fields are duplicated in 'local' and 'remote', for backwards-
+    // compatility with older clients.
+    pub local: LocalTimelineInfo,
+    pub remote: RemoteTimelineInfo,
+}
+
+#[serde_as]
+#[derive(Debug, Serialize, Deserialize, Clone)]
+pub struct LocalTimelineInfo {
+    #[serde_as(as = "Option<DisplayFromStr>")]
+    pub ancestor_timeline_id: Option<TimelineId>,
+    #[serde_as(as = "Option<DisplayFromStr>")]
+    pub ancestor_lsn: Option<Lsn>,
+    pub current_logical_size: Option<u64>, // is None when timeline is Unloaded
+    pub current_physical_size: Option<u64>, // is None when timeline is Unloaded
 }
 
 #[serde_as]
 #[derive(Debug, Serialize, Deserialize, Clone)]
 pub struct RemoteTimelineInfo {
-    #[serde_as(as = "DisplayFromStr")]
-    pub remote_consistent_lsn: Lsn,
-    pub awaits_download: bool,
-}
-
-///
-/// This represents the output of the "timeline_detail" API call.
-///
-#[serde_as]
-#[derive(Debug, Serialize, Deserialize, Clone)]
-pub struct TimelineInfo {
-    #[serde_as(as = "DisplayFromStr")]
-    pub tenant_id: TenantId,
-    #[serde_as(as = "DisplayFromStr")]
-    pub timeline_id: TimelineId,
-    pub local: Option<LocalTimelineInfo>,
-    pub remote: Option<RemoteTimelineInfo>,
+    #[serde_as(as = "Option<DisplayFromStr>")]
+    pub remote_consistent_lsn: Option<Lsn>,
 }
 
 pub type ConfigureFailpointsRequest = Vec<FailpointConfig>;
@@ -190,3 +227,262 @@ pub struct FailpointConfig {
 pub struct TimelineGcRequest {
     pub gc_horizon: Option<u64>,
 }
+
+// Wrapped in libpq CopyData
+#[derive(PartialEq, Eq)]
+pub enum PagestreamFeMessage {
+    Exists(PagestreamExistsRequest),
+    Nblocks(PagestreamNblocksRequest),
+    GetPage(PagestreamGetPageRequest),
+    DbSize(PagestreamDbSizeRequest),
+}
+
+// Wrapped in libpq CopyData
+pub enum PagestreamBeMessage {
+    Exists(PagestreamExistsResponse),
+    Nblocks(PagestreamNblocksResponse),
+    GetPage(PagestreamGetPageResponse),
+    Error(PagestreamErrorResponse),
+    DbSize(PagestreamDbSizeResponse),
+}
+
+#[derive(Debug, PartialEq, Eq)]
+pub struct PagestreamExistsRequest {
+    pub latest: bool,
+    pub lsn: Lsn,
+    pub rel: RelTag,
+}
+
+#[derive(Debug, PartialEq, Eq)]
+pub struct PagestreamNblocksRequest {
+    pub latest: bool,
+    pub lsn: Lsn,
+    pub rel: RelTag,
+}
+
+#[derive(Debug, PartialEq, Eq)]
+pub struct PagestreamGetPageRequest {
+    pub latest: bool,
+    pub lsn: Lsn,
+    pub rel: RelTag,
+    pub blkno: u32,
+}
+
+#[derive(Debug, PartialEq, Eq)]
+pub struct PagestreamDbSizeRequest {
+    pub latest: bool,
+    pub lsn: Lsn,
+    pub dbnode: u32,
+}
+
+#[derive(Debug)]
+pub struct PagestreamExistsResponse {
+    pub exists: bool,
+}
+
+#[derive(Debug)]
+pub struct PagestreamNblocksResponse {
+    pub n_blocks: u32,
+}
+
+#[derive(Debug)]
+pub struct PagestreamGetPageResponse {
+    pub page: Bytes,
+}
+
+#[derive(Debug)]
+pub struct PagestreamErrorResponse {
+    pub message: String,
+}
+
+#[derive(Debug)]
+pub struct PagestreamDbSizeResponse {
+    pub db_size: i64,
+}
+
+impl PagestreamFeMessage {
+    pub fn serialize(&self) -> Bytes {
+        let mut bytes = BytesMut::new();
+
+        match self {
+            Self::Exists(req) => {
+                bytes.put_u8(0);
+                bytes.put_u8(if req.latest { 1 } else { 0 });
+                bytes.put_u64(req.lsn.0);
+                bytes.put_u32(req.rel.spcnode);
+                bytes.put_u32(req.rel.dbnode);
+                bytes.put_u32(req.rel.relnode);
+                bytes.put_u8(req.rel.forknum);
+            }
+
+            Self::Nblocks(req) => {
+                bytes.put_u8(1);
+                bytes.put_u8(if req.latest { 1 } else { 0 });
+                bytes.put_u64(req.lsn.0);
+                bytes.put_u32(req.rel.spcnode);
+                bytes.put_u32(req.rel.dbnode);
+                bytes.put_u32(req.rel.relnode);
+                bytes.put_u8(req.rel.forknum);
+            }
+
+            Self::GetPage(req) => {
+                bytes.put_u8(2);
+                bytes.put_u8(if req.latest { 1 } else { 0 });
+                bytes.put_u64(req.lsn.0);
+                bytes.put_u32(req.rel.spcnode);
+                bytes.put_u32(req.rel.dbnode);
+                bytes.put_u32(req.rel.relnode);
+                bytes.put_u8(req.rel.forknum);
+                bytes.put_u32(req.blkno);
+            }
+
+            Self::DbSize(req) => {
+                bytes.put_u8(3);
+                bytes.put_u8(if req.latest { 1 } else { 0 });
+                bytes.put_u64(req.lsn.0);
+                bytes.put_u32(req.dbnode);
+            }
+        }
+
+        bytes.into()
+    }
+
+    pub fn parse<R: std::io::Read>(body: &mut R) -> anyhow::Result<PagestreamFeMessage> {
+        // TODO these gets can fail
+
+        // these correspond to the NeonMessageTag enum in pagestore_client.h
+        //
+        // TODO: consider using protobuf or serde bincode for less error prone
+        // serialization.
+        let msg_tag = body.read_u8()?;
+        match msg_tag {
+            0 => Ok(PagestreamFeMessage::Exists(PagestreamExistsRequest {
+                latest: body.read_u8()? != 0,
+                lsn: Lsn::from(body.read_u64::<BigEndian>()?),
+                rel: RelTag {
+                    spcnode: body.read_u32::<BigEndian>()?,
+                    dbnode: body.read_u32::<BigEndian>()?,
+                    relnode: body.read_u32::<BigEndian>()?,
+                    forknum: body.read_u8()?,
+                },
+            })),
+            1 => Ok(PagestreamFeMessage::Nblocks(PagestreamNblocksRequest {
+                latest: body.read_u8()? != 0,
+                lsn: Lsn::from(body.read_u64::<BigEndian>()?),
+                rel: RelTag {
+                    spcnode: body.read_u32::<BigEndian>()?,
+                    dbnode: body.read_u32::<BigEndian>()?,
+                    relnode: body.read_u32::<BigEndian>()?,
+                    forknum: body.read_u8()?,
+                },
+            })),
+            2 => Ok(PagestreamFeMessage::GetPage(PagestreamGetPageRequest {
+                latest: body.read_u8()? != 0,
+                lsn: Lsn::from(body.read_u64::<BigEndian>()?),
+                rel: RelTag {
+                    spcnode: body.read_u32::<BigEndian>()?,
+                    dbnode: body.read_u32::<BigEndian>()?,
+                    relnode: body.read_u32::<BigEndian>()?,
+                    forknum: body.read_u8()?,
+                },
+                blkno: body.read_u32::<BigEndian>()?,
+            })),
+            3 => Ok(PagestreamFeMessage::DbSize(PagestreamDbSizeRequest {
+                latest: body.read_u8()? != 0,
+                lsn: Lsn::from(body.read_u64::<BigEndian>()?),
+                dbnode: body.read_u32::<BigEndian>()?,
+            })),
+            _ => bail!("unknown smgr message tag: {:?}", msg_tag),
+        }
+    }
+}
+
+impl PagestreamBeMessage {
+    pub fn serialize(&self) -> Bytes {
+        let mut bytes = BytesMut::new();
+
+        match self {
+            Self::Exists(resp) => {
+                bytes.put_u8(100); /* tag from pagestore_client.h */
+                bytes.put_u8(resp.exists as u8);
+            }
+
+            Self::Nblocks(resp) => {
+                bytes.put_u8(101); /* tag from pagestore_client.h */
+                bytes.put_u32(resp.n_blocks);
+            }
+
+            Self::GetPage(resp) => {
+                bytes.put_u8(102); /* tag from pagestore_client.h */
+                bytes.put(&resp.page[..]);
+            }
+
+            Self::Error(resp) => {
+                bytes.put_u8(103); /* tag from pagestore_client.h */
+                bytes.put(resp.message.as_bytes());
+                bytes.put_u8(0); // null terminator
+            }
+            Self::DbSize(resp) => {
+                bytes.put_u8(104); /* tag from pagestore_client.h */
+                bytes.put_i64(resp.db_size);
+            }
+        }
+
+        bytes.into()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use bytes::Buf;
+
+    use super::*;
+
+    #[test]
+    fn test_pagestream() {
+        // Test serialization/deserialization of PagestreamFeMessage
+        let messages = vec![
+            PagestreamFeMessage::Exists(PagestreamExistsRequest {
+                latest: true,
+                lsn: Lsn(4),
+                rel: RelTag {
+                    forknum: 1,
+                    spcnode: 2,
+                    dbnode: 3,
+                    relnode: 4,
+                },
+            }),
+            PagestreamFeMessage::Nblocks(PagestreamNblocksRequest {
+                latest: false,
+                lsn: Lsn(4),
+                rel: RelTag {
+                    forknum: 1,
+                    spcnode: 2,
+                    dbnode: 3,
+                    relnode: 4,
+                },
+            }),
+            PagestreamFeMessage::GetPage(PagestreamGetPageRequest {
+                latest: true,
+                lsn: Lsn(4),
+                rel: RelTag {
+                    forknum: 1,
+                    spcnode: 2,
+                    dbnode: 3,
+                    relnode: 4,
+                },
+                blkno: 7,
+            }),
+            PagestreamFeMessage::DbSize(PagestreamDbSizeRequest {
+                latest: true,
+                lsn: Lsn(4),
+                dbnode: 7,
+            }),
+        ];
+        for msg in messages {
+            let bytes = msg.serialize();
+            let reconstructed = PagestreamFeMessage::parse(&mut bytes.reader()).unwrap();
+            assert!(msg == reconstructed);
+        }
+    }
+}

libs/persistent_range_query/Cargo.toml

@@ -0,0 +1,12 @@
+[package]
+name = "persistent_range_query"
+version = "0.1.0"
+edition = "2021"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+workspace_hack = { version = "0.1", path = "../../workspace_hack" }
+
+[dev-dependencies]
+rand = "0.8.3"

libs/persistent_range_query/src/lib.rs

@@ -0,0 +1,78 @@
+use std::ops::Range;
+
+pub mod naive;
+pub mod ops;
+pub mod segment_tree;
+
+/// Should be a monoid:
+/// * Identity element: for all a: combine(new_for_empty_range(), a) = combine(a, new_for_empty_range()) = a
+/// * Associativity: for all a, b, c: combine(combine(a, b), c) == combine(a, combine(b, c))
+pub trait RangeQueryResult<Key>: Sized + Clone {
+    // Clone is equivalent to combine with an empty range.
+
+    fn new_for_empty_range() -> Self;
+
+    // Contract: left_range.end == right_range.start
+    // left_range.start == left_range.end == right_range.start == right_range.end is still possible
+    fn combine(
+        left: &Self,
+        left_range: &Range<Key>,
+        right: &Self,
+        right_range: &Range<Key>,
+    ) -> Self;
+
+    fn add(left: &mut Self, left_range: &Range<Key>, right: &Self, right_range: &Range<Key>);
+}
+
+pub trait LazyRangeInitializer<Result: RangeQueryResult<Key>, Key> {
+    fn get(&self, range: &Range<Key>) -> Result;
+}
+
+/// Should be a monoid:
+/// * Identity element: for all op: compose(no_op(), op) == compose(op, no_op()) == op
+/// * Associativity: for all op_1, op_2, op_3: compose(compose(op_1, op_2), op_3) == compose(op_1, compose(op_2, op_3))
+///
+/// Should left act on Result:
+/// * Identity operation: for all r: no_op().apply(r) == r
+/// * Compatibility: for all op_1, op_2, r: op_1.apply(op_2.apply(r)) == compose(op_1, op_2).apply(r)
+pub trait RangeModification<Key> {
+    type Result: RangeQueryResult<Key>;
+
+    fn no_op() -> Self;
+    fn is_no_op(&self) -> bool;
+    fn is_reinitialization(&self) -> bool;
+    fn apply(&self, result: &mut Self::Result, range: &Range<Key>);
+    fn compose(later: &Self, earlier: &mut Self);
+}
+
+pub trait VecReadableVersion<Modification: RangeModification<Key>, Key> {
+    fn get(&self, keys: &Range<Key>) -> Modification::Result;
+}
+
+// TODO: use trait alias when stabilized
+pub trait VecFrozenVersion<Modification: RangeModification<Key>, Key>:
+    Clone + VecReadableVersion<Modification, Key>
+{
+}
+
+impl<
+        T: Clone + VecReadableVersion<Modification, Key>,
+        Modification: RangeModification<Key>,
+        Key,
+    > VecFrozenVersion<Modification, Key> for T
+{
+}
+
+pub trait PersistentVecStorage<
+    Modification: RangeModification<Key>,
+    Initializer: LazyRangeInitializer<Modification::Result, Key>,
+    Key,
+>: VecReadableVersion<Modification, Key>
+{
+    fn new(all_keys: Range<Key>, initializer: Initializer) -> Self;
+
+    type FrozenVersion: VecFrozenVersion<Modification, Key>;
+
+    fn modify(&mut self, keys: &Range<Key>, modification: &Modification);
+    fn freeze(&mut self) -> Self::FrozenVersion;
+}

libs/persistent_range_query/src/naive.rs

@@ -0,0 +1,115 @@
+use crate::{
+    LazyRangeInitializer, PersistentVecStorage, RangeModification, RangeQueryResult,
+    VecReadableVersion,
+};
+use std::marker::PhantomData;
+use std::ops::Range;
+use std::rc::Rc;
+
+pub struct NaiveFrozenVersion<Modification: RangeModification<Key>, Key> {
+    all_keys: Range<Key>,
+    values: Rc<Box<Vec<Modification::Result>>>,
+}
+
+pub trait IndexableKey: Clone {
+    fn index(all_keys: &Range<Self>, key: &Self) -> usize;
+    fn element_range(all_keys: &Range<Self>, index: usize) -> Range<Self>;
+}
+
+fn get<Modification: RangeModification<Key>, Key: IndexableKey>(
+    all_keys: &Range<Key>,
+    values: &Vec<Modification::Result>,
+    keys: &Range<Key>,
+) -> Modification::Result {
+    let mut result = Modification::Result::new_for_empty_range();
+    let mut result_range = keys.start.clone()..keys.start.clone();
+    for index in
+        IndexableKey::index(&all_keys, &keys.start)..IndexableKey::index(&all_keys, &keys.end)
+    {
+        let element_range = IndexableKey::element_range(&all_keys, index);
+        Modification::Result::add(&mut result, &result_range, &values[index], &element_range);
+        result_range.end = element_range.end;
+    }
+    result
+}
+
+impl<Modification: RangeModification<Key>, Key: IndexableKey> VecReadableVersion<Modification, Key>
+    for NaiveFrozenVersion<Modification, Key>
+{
+    fn get(&self, keys: &Range<Key>) -> Modification::Result {
+        get::<Modification, Key>(&self.all_keys, &self.values, keys)
+    }
+}
+
+// Manual implementation of `Clone` becase `derive` requires `Modification: Clone`
+impl<Modification: RangeModification<Key>, Key: Clone> Clone
+    for NaiveFrozenVersion<Modification, Key>
+{
+    fn clone(&self) -> Self {
+        Self {
+            all_keys: self.all_keys.clone(),
+            values: self.values.clone(),
+        }
+    }
+}
+
+// TODO: is it at all possible to store previous versions in this struct,
+// without any Rc<>?
+pub struct NaiveVecStorage<
+    Modification: RangeModification<Key>,
+    Initializer: LazyRangeInitializer<Modification::Result, Key>,
+    Key: IndexableKey,
+> {
+    all_keys: Range<Key>,
+    last_version: Vec<Modification::Result>,
+    _initializer: PhantomData<Initializer>,
+}
+
+impl<
+        Modification: RangeModification<Key>,
+        Initializer: LazyRangeInitializer<Modification::Result, Key>,
+        Key: IndexableKey,
+    > VecReadableVersion<Modification, Key> for NaiveVecStorage<Modification, Initializer, Key>
+{
+    fn get(&self, keys: &Range<Key>) -> Modification::Result {
+        get::<Modification, Key>(&self.all_keys, &self.last_version, keys)
+    }
+}
+
+impl<
+        Modification: RangeModification<Key>,
+        Initializer: LazyRangeInitializer<Modification::Result, Key>,
+        Key: IndexableKey,
+    > PersistentVecStorage<Modification, Initializer, Key>
+    for NaiveVecStorage<Modification, Initializer, Key>
+{
+    fn new(all_keys: Range<Key>, initializer: Initializer) -> Self {
+        let mut values = Vec::with_capacity(IndexableKey::index(&all_keys, &all_keys.end));
+        for index in 0..values.capacity() {
+            values.push(initializer.get(&IndexableKey::element_range(&all_keys, index)));
+        }
+        NaiveVecStorage {
+            all_keys,
+            last_version: values,
+            _initializer: PhantomData,
+        }
+    }
+
+    type FrozenVersion = NaiveFrozenVersion<Modification, Key>;
+
+    fn modify(&mut self, keys: &Range<Key>, modification: &Modification) {
+        for index in IndexableKey::index(&self.all_keys, &keys.start)
+            ..IndexableKey::index(&self.all_keys, &keys.end)
+        {
+            let element_range = IndexableKey::element_range(&self.all_keys, index);
+            modification.apply(&mut self.last_version[index], &element_range);
+        }
+    }
+
+    fn freeze(&mut self) -> Self::FrozenVersion {
+        NaiveFrozenVersion::<Modification, Key> {
+            all_keys: self.all_keys.clone(),
+            values: Rc::new(Box::new(self.last_version.clone())),
+        }
+    }
+}

libs/persistent_range_query/src/ops/mod.rs

@@ -0,0 +1,14 @@
+pub mod rsq;
+
+#[derive(Copy, Clone, Debug)]
+pub struct SameElementsInitializer<T> {
+    initial_element_value: T,
+}
+
+impl<T> SameElementsInitializer<T> {
+    pub fn new(initial_element_value: T) -> Self {
+        SameElementsInitializer {
+            initial_element_value,
+        }
+    }
+}

libs/persistent_range_query/src/ops/rsq.rs

@@ -0,0 +1,118 @@
+//! # Range Sum Query
+
+use crate::ops::SameElementsInitializer;
+use crate::{LazyRangeInitializer, RangeModification, RangeQueryResult};
+use std::borrow::Borrow;
+use std::ops::{Add, AddAssign, Range};
+
+// TODO: commutative Add
+
+#[derive(Clone, Copy, Debug)]
+pub struct SumResult<T> {
+    sum: T,
+}
+
+impl<T> SumResult<T> {
+    pub fn sum(&self) -> &T {
+        &self.sum
+    }
+}
+
+impl<T: Clone + for<'a> AddAssign<&'a T> + From<u8>, Key> RangeQueryResult<Key> for SumResult<T>
+where
+    for<'a> &'a T: Add<&'a T, Output = T>,
+{
+    fn new_for_empty_range() -> Self {
+        SumResult { sum: 0.into() }
+    }
+
+    fn combine(
+        left: &Self,
+        _left_range: &Range<Key>,
+        right: &Self,
+        _right_range: &Range<Key>,
+    ) -> Self {
+        SumResult {
+            sum: &left.sum + &right.sum,
+        }
+    }
+
+    fn add(left: &mut Self, _left_range: &Range<Key>, right: &Self, _right_range: &Range<Key>) {
+        left.sum += &right.sum
+    }
+}
+
+pub trait SumOfSameElements<Key> {
+    fn sum(initial_element_value: &Self, keys: &Range<Key>) -> Self;
+}
+
+impl<T: SumOfSameElements<Key>, TB: Borrow<T>, Key> LazyRangeInitializer<SumResult<T>, Key>
+    for SameElementsInitializer<TB>
+where
+    SumResult<T>: RangeQueryResult<Key>,
+{
+    fn get(&self, range: &Range<Key>) -> SumResult<T> {
+        SumResult {
+            sum: SumOfSameElements::sum(self.initial_element_value.borrow(), range),
+        }
+    }
+}
+
+#[derive(Copy, Clone, Debug)]
+pub enum AddAssignModification<T> {
+    None,
+    Add(T),
+    Assign(T),
+}
+
+impl<T: Clone + for<'a> AddAssign<&'a T>, Key> RangeModification<Key> for AddAssignModification<T>
+where
+    SumResult<T>: RangeQueryResult<Key>,
+    for<'a> SameElementsInitializer<&'a T>: LazyRangeInitializer<SumResult<T>, Key>,
+{
+    type Result = SumResult<T>;
+
+    fn no_op() -> Self {
+        AddAssignModification::None
+    }
+
+    fn is_no_op(&self) -> bool {
+        match self {
+            AddAssignModification::None => true,
+            _ => false,
+        }
+    }
+
+    fn is_reinitialization(&self) -> bool {
+        match self {
+            AddAssignModification::Assign(_) => true,
+            _ => false,
+        }
+    }
+
+    fn apply(&self, result: &mut SumResult<T>, range: &Range<Key>) {
+        use AddAssignModification::*;
+        match self {
+            None => {}
+            Add(x) | Assign(x) => {
+                let to_add = SameElementsInitializer::new(x).get(range).sum;
+                if let Assign(_) = self {
+                    result.sum = to_add;
+                } else {
+                    result.sum += &to_add;
+                }
+            }
+        }
+    }
+
+    fn compose(later: &Self, earlier: &mut Self) {
+        use AddAssignModification::*;
+        match (later, earlier) {
+            (_, e @ None) => *e = later.clone(),
+            (None, _) => {}
+            (Assign(_), e) => *e = later.clone(),
+            (Add(x), Add(y)) => *y += x,
+            (Add(x), Assign(value)) => *value += x,
+        }
+    }
+}

libs/persistent_range_query/src/segment_tree.rs

@@ -0,0 +1,255 @@
+//! # Segment Tree
+//! It is a competitive programming folklore data structure. Do not confuse with the interval tree.
+
+use crate::{LazyRangeInitializer, PersistentVecStorage, RangeQueryResult, VecReadableVersion};
+use std::ops::Range;
+use std::rc::Rc;
+
+pub trait MidpointableKey: Clone + Ord + Sized {
+    fn midpoint(range: &Range<Self>) -> Self;
+}
+
+pub trait RangeModification<Key>: Clone + crate::RangeModification<Key> {}
+
+// TODO: use trait alias when stabilized
+impl<T: Clone + crate::RangeModification<Key>, Key> RangeModification<Key> for T {}
+
+#[derive(Debug)]
+struct Node<Modification: RangeModification<Key>, Key> {
+    result: Modification::Result,
+    modify_children: Modification,
+    left: Option<Rc<Self>>,
+    right: Option<Rc<Self>>,
+}
+
+// Manual implementation because we don't need `Key: Clone` for this, unlike with `derive`.
+impl<Modification: RangeModification<Key>, Key> Clone for Node<Modification, Key> {
+    fn clone(&self) -> Self {
+        Node {
+            result: self.result.clone(),
+            modify_children: self.modify_children.clone(),
+            left: self.left.clone(),
+            right: self.right.clone(),
+        }
+    }
+}
+
+impl<Modification: RangeModification<Key>, Key> Node<Modification, Key> {
+    fn new<Initializer: LazyRangeInitializer<Modification::Result, Key>>(
+        range: &Range<Key>,
+        initializer: &Initializer,
+    ) -> Self {
+        Node {
+            result: initializer.get(range),
+            modify_children: Modification::no_op(),
+            left: None,
+            right: None,
+        }
+    }
+
+    pub fn apply(&mut self, modification: &Modification, range: &Range<Key>) {
+        modification.apply(&mut self.result, range);
+        Modification::compose(modification, &mut self.modify_children);
+        if self.modify_children.is_reinitialization() {
+            self.left = None;
+            self.right = None;
+        }
+    }
+
+    pub fn force_children<Initializer: LazyRangeInitializer<Modification::Result, Key>>(
+        &mut self,
+        initializer: &Initializer,
+        range_left: &Range<Key>,
+        range_right: &Range<Key>,
+    ) {
+        let left = Rc::make_mut(
+            self.left
+                .get_or_insert_with(|| Rc::new(Node::new(&range_left, initializer))),
+        );
+        let right = Rc::make_mut(
+            self.right
+                .get_or_insert_with(|| Rc::new(Node::new(&range_right, initializer))),
+        );
+        left.apply(&self.modify_children, &range_left);
+        right.apply(&self.modify_children, &range_right);
+        self.modify_children = Modification::no_op();
+    }
+
+    pub fn recalculate_from_children(&mut self, range_left: &Range<Key>, range_right: &Range<Key>) {
+        assert!(self.modify_children.is_no_op());
+        assert!(self.left.is_some());
+        assert!(self.right.is_some());
+        self.result = Modification::Result::combine(
+            &self.left.as_ref().unwrap().result,
+            &range_left,
+            &self.right.as_ref().unwrap().result,
+            &range_right,
+        );
+    }
+}
+
+fn split_range<Key: MidpointableKey>(range: &Range<Key>) -> (Range<Key>, Range<Key>) {
+    let range_left = range.start.clone()..MidpointableKey::midpoint(range);
+    let range_right = range_left.end.clone()..range.end.clone();
+    (range_left, range_right)
+}
+
+pub struct PersistentSegmentTreeVersion<
+    Modification: RangeModification<Key>,
+    Initializer: LazyRangeInitializer<Modification::Result, Key>,
+    Key: Clone,
+> {
+    root: Rc<Node<Modification, Key>>,
+    all_keys: Range<Key>,
+    initializer: Rc<Initializer>,
+}
+
+// Manual implementation because we don't need `Key: Clone` for this, unlike with `derive`.
+impl<
+        Modification: RangeModification<Key>,
+        Initializer: LazyRangeInitializer<Modification::Result, Key>,
+        Key: Clone,
+    > Clone for PersistentSegmentTreeVersion<Modification, Initializer, Key>
+{
+    fn clone(&self) -> Self {
+        Self {
+            root: self.root.clone(),
+            all_keys: self.all_keys.clone(),
+            initializer: self.initializer.clone(),
+        }
+    }
+}
+
+fn get<
+    Modification: RangeModification<Key>,
+    Initializer: LazyRangeInitializer<Modification::Result, Key>,
+    Key: MidpointableKey,
+>(
+    node: &mut Rc<Node<Modification, Key>>,
+    node_keys: &Range<Key>,
+    initializer: &Initializer,
+    keys: &Range<Key>,
+) -> Modification::Result {
+    if node_keys.end <= keys.start || keys.end <= node_keys.start {
+        return Modification::Result::new_for_empty_range();
+    }
+    if keys.start <= node_keys.start && node_keys.end <= keys.end {
+        return node.result.clone();
+    }
+    let node = Rc::make_mut(node);
+    let (left_keys, right_keys) = split_range(node_keys);
+    node.force_children(initializer, &left_keys, &right_keys);
+    let mut result = get(node.left.as_mut().unwrap(), &left_keys, initializer, keys);
+    Modification::Result::add(
+        &mut result,
+        &left_keys,
+        &get(node.right.as_mut().unwrap(), &right_keys, initializer, keys),
+        &right_keys,
+    );
+    result
+}
+
+fn modify<
+    Modification: RangeModification<Key>,
+    Initializer: LazyRangeInitializer<Modification::Result, Key>,
+    Key: MidpointableKey,
+>(
+    node: &mut Rc<Node<Modification, Key>>,
+    node_keys: &Range<Key>,
+    initializer: &Initializer,
+    keys: &Range<Key>,
+    modification: &Modification,
+) {
+    if modification.is_no_op() || node_keys.end <= keys.start || keys.end <= node_keys.start {
+        return;
+    }
+    let node = Rc::make_mut(node);
+    if keys.start <= node_keys.start && node_keys.end <= keys.end {
+        node.apply(modification, node_keys);
+        return;
+    }
+    let (left_keys, right_keys) = split_range(node_keys);
+    node.force_children(initializer, &left_keys, &right_keys);
+    modify(
+        node.left.as_mut().unwrap(),
+        &left_keys,
+        initializer,
+        keys,
+        &modification,
+    );
+    modify(
+        node.right.as_mut().unwrap(),
+        &right_keys,
+        initializer,
+        keys,
+        &modification,
+    );
+    node.recalculate_from_children(&left_keys, &right_keys);
+}
+
+impl<
+        Modification: RangeModification<Key>,
+        Initializer: LazyRangeInitializer<Modification::Result, Key>,
+        Key: MidpointableKey,
+    > VecReadableVersion<Modification, Key>
+    for PersistentSegmentTreeVersion<Modification, Initializer, Key>
+{
+    fn get(&self, keys: &Range<Key>) -> Modification::Result {
+        get(
+            &mut self.root.clone(), // TODO: do not always force a branch
+            &self.all_keys,
+            self.initializer.as_ref(),
+            keys,
+        )
+    }
+}
+
+pub struct PersistentSegmentTree<
+    Modification: RangeModification<Key>,
+    Initializer: LazyRangeInitializer<Modification::Result, Key>,
+    Key: MidpointableKey,
+>(PersistentSegmentTreeVersion<Modification, Initializer, Key>);
+
+impl<
+        Modification: RangeModification<Key>,
+        Initializer: LazyRangeInitializer<Modification::Result, Key>,
+        Key: MidpointableKey,
+    > VecReadableVersion<Modification, Key>
+    for PersistentSegmentTree<Modification, Initializer, Key>
+{
+    fn get(&self, keys: &Range<Key>) -> Modification::Result {
+        self.0.get(keys)
+    }
+}
+
+impl<
+        Modification: RangeModification<Key>,
+        Initializer: LazyRangeInitializer<Modification::Result, Key>,
+        Key: MidpointableKey,
+    > PersistentVecStorage<Modification, Initializer, Key>
+    for PersistentSegmentTree<Modification, Initializer, Key>
+{
+    fn new(all_keys: Range<Key>, initializer: Initializer) -> Self {
+        PersistentSegmentTree(PersistentSegmentTreeVersion {
+            root: Rc::new(Node::new(&all_keys, &initializer)),
+            all_keys: all_keys,
+            initializer: Rc::new(initializer),
+        })
+    }
+
+    type FrozenVersion = PersistentSegmentTreeVersion<Modification, Initializer, Key>;
+
+    fn modify(&mut self, keys: &Range<Key>, modification: &Modification) {
+        modify(
+            &mut self.0.root, // TODO: do not always force a branch
+            &self.0.all_keys,
+            self.0.initializer.as_ref(),
+            keys,
+            modification,
+        )
+    }
+
+    fn freeze(&mut self) -> Self::FrozenVersion {
+        self.0.clone()
+    }
+}

libs/persistent_range_query/tests/layer_map_test.rs

@@ -0,0 +1,295 @@
+use persistent_range_query::naive::{IndexableKey, NaiveVecStorage};
+use persistent_range_query::ops::SameElementsInitializer;
+use persistent_range_query::segment_tree::{MidpointableKey, PersistentSegmentTree};
+use persistent_range_query::{
+    LazyRangeInitializer, PersistentVecStorage, RangeModification, RangeQueryResult,
+    VecReadableVersion,
+};
+use std::cmp::Ordering;
+use std::ops::Range;
+
+#[derive(Clone, Copy, Debug, Eq, PartialEq, Ord, PartialOrd)]
+struct PageIndex(u32);
+type LayerId = String;
+
+impl IndexableKey for PageIndex {
+    fn index(all_keys: &Range<Self>, key: &Self) -> usize {
+        (key.0 as usize) - (all_keys.start.0 as usize)
+    }
+
+    fn element_range(all_keys: &Range<Self>, index: usize) -> Range<Self> {
+        PageIndex(all_keys.start.0 + index as u32)..PageIndex(all_keys.start.0 + index as u32 + 1)
+    }
+}
+
+impl MidpointableKey for PageIndex {
+    fn midpoint(range: &Range<Self>) -> Self {
+        PageIndex(range.start.0 + (range.end.0 - range.start.0) / 2)
+    }
+}
+
+#[derive(Clone, Debug, Eq, PartialEq)]
+struct LayerMapInformation {
+    // Only make sense for a range of length 1.
+    last_layer: Option<LayerId>,
+    last_image_layer: Option<LayerId>,
+    // Work for all ranges
+    max_delta_layers: (usize, Range<PageIndex>),
+}
+
+impl LayerMapInformation {
+    fn last_layers(&self) -> (&Option<LayerId>, &Option<LayerId>) {
+        (&self.last_layer, &self.last_image_layer)
+    }
+
+    fn max_delta_layers(&self) -> &(usize, Range<PageIndex>) {
+        &self.max_delta_layers
+    }
+}
+
+fn merge_ranges(left: &Range<PageIndex>, right: &Range<PageIndex>) -> Range<PageIndex> {
+    if left.is_empty() {
+        right.clone()
+    } else if right.is_empty() {
+        left.clone()
+    } else if left.end == right.start {
+        left.start..right.end
+    } else {
+        left.clone()
+    }
+}
+
+impl RangeQueryResult<PageIndex> for LayerMapInformation {
+    fn new_for_empty_range() -> Self {
+        LayerMapInformation {
+            last_layer: None,
+            last_image_layer: None,
+            max_delta_layers: (0, PageIndex(0)..PageIndex(0)),
+        }
+    }
+
+    fn combine(
+        left: &Self,
+        _left_range: &Range<PageIndex>,
+        right: &Self,
+        _right_range: &Range<PageIndex>,
+    ) -> Self {
+        // Note that either range may be empty.
+        LayerMapInformation {
+            last_layer: left
+                .last_layer
+                .as_ref()
+                .or_else(|| right.last_layer.as_ref())
+                .cloned(),
+            last_image_layer: left
+                .last_image_layer
+                .as_ref()
+                .or_else(|| right.last_image_layer.as_ref())
+                .cloned(),
+            max_delta_layers: match left.max_delta_layers.0.cmp(&right.max_delta_layers.0) {
+                Ordering::Less => right.max_delta_layers.clone(),
+                Ordering::Greater => left.max_delta_layers.clone(),
+                Ordering::Equal => (
+                    left.max_delta_layers.0,
+                    merge_ranges(&left.max_delta_layers.1, &right.max_delta_layers.1),
+                ),
+            },
+        }
+    }
+
+    fn add(
+        left: &mut Self,
+        left_range: &Range<PageIndex>,
+        right: &Self,
+        right_range: &Range<PageIndex>,
+    ) {
+        *left = Self::combine(&left, left_range, right, right_range);
+    }
+}
+
+#[derive(Clone, Debug)]
+struct AddDeltaLayers {
+    last_layer: LayerId,
+    count: usize,
+}
+
+#[derive(Clone, Debug)]
+struct LayerMapModification {
+    add_image_layer: Option<LayerId>,
+    add_delta_layers: Option<AddDeltaLayers>,
+}
+
+impl LayerMapModification {
+    fn add_image_layer(layer: impl Into<LayerId>) -> Self {
+        LayerMapModification {
+            add_image_layer: Some(layer.into()),
+            add_delta_layers: None,
+        }
+    }
+
+    fn add_delta_layer(layer: impl Into<LayerId>) -> Self {
+        LayerMapModification {
+            add_image_layer: None,
+            add_delta_layers: Some(AddDeltaLayers {
+                last_layer: layer.into(),
+                count: 1,
+            }),
+        }
+    }
+}
+
+impl RangeModification<PageIndex> for LayerMapModification {
+    type Result = LayerMapInformation;
+
+    fn no_op() -> Self {
+        LayerMapModification {
+            add_image_layer: None,
+            add_delta_layers: None,
+        }
+    }
+
+    fn is_no_op(&self) -> bool {
+        self.add_image_layer.is_none() && self.add_delta_layers.is_none()
+    }
+
+    fn is_reinitialization(&self) -> bool {
+        self.add_image_layer.is_some()
+    }
+
+    fn apply(&self, result: &mut Self::Result, range: &Range<PageIndex>) {
+        if let Some(layer) = &self.add_image_layer {
+            result.last_layer = Some(layer.clone());
+            result.last_image_layer = Some(layer.clone());
+            result.max_delta_layers = (0, range.clone());
+        }
+        if let Some(AddDeltaLayers { last_layer, count }) = &self.add_delta_layers {
+            result.last_layer = Some(last_layer.clone());
+            result.max_delta_layers.0 += count;
+        }
+    }
+
+    fn compose(later: &Self, earlier: &mut Self) {
+        if later.add_image_layer.is_some() {
+            *earlier = later.clone();
+            return;
+        }
+        if let Some(AddDeltaLayers { last_layer, count }) = &later.add_delta_layers {
+            let res = earlier.add_delta_layers.get_or_insert(AddDeltaLayers {
+                last_layer: LayerId::default(),
+                count: 0,
+            });
+            res.last_layer = last_layer.clone();
+            res.count += count;
+        }
+    }
+}
+
+impl LazyRangeInitializer<LayerMapInformation, PageIndex> for SameElementsInitializer<()> {
+    fn get(&self, range: &Range<PageIndex>) -> LayerMapInformation {
+        LayerMapInformation {
+            last_layer: None,
+            last_image_layer: None,
+            max_delta_layers: (0, range.clone()),
+        }
+    }
+}
+
+fn test_layer_map<
+    S: PersistentVecStorage<LayerMapModification, SameElementsInitializer<()>, PageIndex>,
+>() {
+    let mut s = S::new(
+        PageIndex(0)..PageIndex(100),
+        SameElementsInitializer::new(()),
+    );
+    s.modify(
+        &(PageIndex(0)..PageIndex(70)),
+        &LayerMapModification::add_image_layer("Img0..70"),
+    );
+    s.modify(
+        &(PageIndex(50)..PageIndex(100)),
+        &LayerMapModification::add_image_layer("Img50..100"),
+    );
+    s.modify(
+        &(PageIndex(10)..PageIndex(60)),
+        &LayerMapModification::add_delta_layer("Delta10..60"),
+    );
+    let s_before_last_delta = s.freeze();
+    s.modify(
+        &(PageIndex(20)..PageIndex(80)),
+        &LayerMapModification::add_delta_layer("Delta20..80"),
+    );
+
+    assert_eq!(
+        s.get(&(PageIndex(5)..PageIndex(6))).last_layers(),
+        (&Some("Img0..70".to_owned()), &Some("Img0..70".to_owned()))
+    );
+    assert_eq!(
+        s.get(&(PageIndex(15)..PageIndex(16))).last_layers(),
+        (
+            &Some("Delta10..60".to_owned()),
+            &Some("Img0..70".to_owned())
+        )
+    );
+    assert_eq!(
+        s.get(&(PageIndex(25)..PageIndex(26))).last_layers(),
+        (
+            &Some("Delta20..80".to_owned()),
+            &Some("Img0..70".to_owned())
+        )
+    );
+    assert_eq!(
+        s.get(&(PageIndex(65)..PageIndex(66))).last_layers(),
+        (
+            &Some("Delta20..80".to_owned()),
+            &Some("Img50..100".to_owned())
+        )
+    );
+    assert_eq!(
+        s.get(&(PageIndex(95)..PageIndex(96))).last_layers(),
+        (
+            &Some("Img50..100".to_owned()),
+            &Some("Img50..100".to_owned())
+        )
+    );
+
+    assert_eq!(
+        s.get(&(PageIndex(0)..PageIndex(100))).max_delta_layers(),
+        &(2, PageIndex(20)..PageIndex(60)),
+    );
+    assert_eq!(
+        *s_before_last_delta
+            .get(&(PageIndex(0)..PageIndex(100)))
+            .max_delta_layers(),
+        (1, PageIndex(10)..PageIndex(60)),
+    );
+
+    assert_eq!(
+        *s.get(&(PageIndex(10)..PageIndex(30))).max_delta_layers(),
+        (2, PageIndex(20)..PageIndex(30))
+    );
+    assert_eq!(
+        *s.get(&(PageIndex(10)..PageIndex(20))).max_delta_layers(),
+        (1, PageIndex(10)..PageIndex(20))
+    );
+
+    assert_eq!(
+        *s.get(&(PageIndex(70)..PageIndex(80))).max_delta_layers(),
+        (1, PageIndex(70)..PageIndex(80))
+    );
+    assert_eq!(
+        *s_before_last_delta
+            .get(&(PageIndex(70)..PageIndex(80)))
+            .max_delta_layers(),
+        (0, PageIndex(70)..PageIndex(80))
+    );
+}
+
+#[test]
+fn test_naive() {
+    test_layer_map::<NaiveVecStorage<_, _, _>>();
+}
+
+#[test]
+fn test_segment_tree() {
+    test_layer_map::<PersistentSegmentTree<_, _, _>>();
+}

libs/persistent_range_query/tests/rsq_test.rs

@@ -0,0 +1,116 @@
+use persistent_range_query::naive::*;
+use persistent_range_query::ops::rsq::AddAssignModification::Add;
+use persistent_range_query::ops::rsq::*;
+use persistent_range_query::ops::SameElementsInitializer;
+use persistent_range_query::segment_tree::{MidpointableKey, PersistentSegmentTree};
+use persistent_range_query::{PersistentVecStorage, VecReadableVersion};
+use rand::{Rng, SeedableRng};
+use std::ops::Range;
+
+#[derive(Copy, Clone, Debug, Eq, PartialEq, Ord, PartialOrd)]
+struct K(u16);
+
+impl IndexableKey for K {
+    fn index(all_keys: &Range<Self>, key: &Self) -> usize {
+        (key.0 as usize) - (all_keys.start.0 as usize)
+    }
+
+    fn element_range(all_keys: &Range<Self>, index: usize) -> Range<Self> {
+        K(all_keys.start.0 + index as u16)..K(all_keys.start.0 + index as u16 + 1)
+    }
+}
+
+impl SumOfSameElements<K> for i32 {
+    fn sum(initial_element_value: &Self, keys: &Range<K>) -> Self {
+        initial_element_value * (keys.end.0 - keys.start.0) as Self
+    }
+}
+
+impl MidpointableKey for K {
+    fn midpoint(range: &Range<Self>) -> Self {
+        K(range.start.0 + (range.end.0 - range.start.0) / 2)
+    }
+}
+
+fn test_storage<
+    S: PersistentVecStorage<AddAssignModification<i32>, SameElementsInitializer<i32>, K>,
+>() {
+    let mut s = S::new(K(0)..K(12), SameElementsInitializer::new(0i32));
+    assert_eq!(*s.get(&(K(0)..K(12))).sum(), 0);
+
+    s.modify(&(K(2)..K(5)), &AddAssignModification::Add(3));
+    assert_eq!(*s.get(&(K(0)..K(12))).sum(), 3 + 3 + 3);
+    let s_old = s.freeze();
+
+    s.modify(&(K(3)..K(6)), &AddAssignModification::Assign(10));
+    assert_eq!(*s.get(&(K(0)..K(12))).sum(), 3 + 10 + 10 + 10);
+
+    s.modify(&(K(4)..K(7)), &AddAssignModification::Add(2));
+    assert_eq!(*s.get(&(K(0)..K(12))).sum(), 3 + 10 + 12 + 12 + 2);
+
+    assert_eq!(*s.get(&(K(4)..K(6))).sum(), 12 + 12);
+    assert_eq!(*s_old.get(&(K(4)..K(6))).sum(), 3);
+}
+
+#[test]
+fn test_naive() {
+    test_storage::<NaiveVecStorage<_, _, _>>();
+}
+
+#[test]
+fn test_segment_tree() {
+    test_storage::<PersistentSegmentTree<_, _, _>>();
+}
+
+#[test]
+fn test_stress() {
+    const LEN: u16 = 17_238;
+    const OPERATIONS: i32 = 20_000;
+
+    let mut rng = rand::rngs::StdRng::seed_from_u64(0);
+    let mut naive: NaiveVecStorage<AddAssignModification<i32>, _, _> =
+        NaiveVecStorage::new(K(0)..K(LEN), SameElementsInitializer::new(2i32));
+    let mut segm_tree: PersistentSegmentTree<AddAssignModification<i32>, _, _> =
+        PersistentSegmentTree::new(K(0)..K(LEN), SameElementsInitializer::new(2i32));
+
+    fn gen_range(rng: &mut impl Rng) -> Range<K> {
+        let l: u16 = rng.gen_range(0..LEN);
+        let r: u16 = rng.gen_range(0..LEN);
+        if l <= r {
+            K(l)..K(r)
+        } else {
+            K(r)..K(l)
+        }
+    }
+
+    for _ in 0..2 {
+        let checksum_range = gen_range(&mut rng);
+        let checksum_before: i32 = *naive.get(&checksum_range).sum();
+        assert_eq!(checksum_before, *segm_tree.get(&checksum_range).sum());
+
+        let naive_before = naive.freeze();
+        let segm_tree_before = segm_tree.freeze();
+        assert_eq!(checksum_before, *naive_before.get(&checksum_range).sum());
+        assert_eq!(checksum_before, *segm_tree.get(&checksum_range).sum());
+
+        for _ in 0..OPERATIONS {
+            {
+                let range = gen_range(&mut rng);
+                assert_eq!(naive.get(&range).sum(), segm_tree.get(&range).sum());
+            }
+            {
+                let range = gen_range(&mut rng);
+                let val = rng.gen_range(-10i32..=10i32);
+                let op = Add(val);
+                naive.modify(&range, &op);
+                segm_tree.modify(&range, &op);
+            }
+        }
+
+        assert_eq!(checksum_before, *naive_before.get(&checksum_range).sum());
+        assert_eq!(
+            checksum_before,
+            *segm_tree_before.get(&checksum_range).sum()
+        );
+    }
+}

libs/postgres_ffi/Cargo.toml

@@ -13,7 +13,7 @@ crc32c = "0.6.0"
 hex = "0.4.3"
 once_cell = "1.13.0"
 log = "0.4.14"
-memoffset = "0.6.2"
+memoffset = "0.7"
 thiserror = "1.0"
 serde = { version = "1.0", features = ["derive"] }
 utils = { path = "../utils" }
@@ -26,4 +26,4 @@ wal_craft = { path = "wal_craft" }
 
 [build-dependencies]
 anyhow = "1.0"
-bindgen = "0.60.1"
+bindgen = "0.61"

libs/postgres_ffi/wal_craft/Cargo.toml

@@ -7,7 +7,7 @@ edition = "2021"
 
 [dependencies]
 anyhow = "1.0"
-clap = "3.0"
+clap = "4.0"
 env_logger = "0.9"
 log = "0.4"
 once_cell = "1.13.0"

libs/postgres_ffi/wal_craft/src/bin/wal_craft.rs

@@ -1,68 +1,19 @@
 use anyhow::*;
-use clap::{App, Arg, ArgMatches};
-use std::str::FromStr;
+use clap::{value_parser, Arg, ArgMatches, Command};
+use std::{path::PathBuf, str::FromStr};
 use wal_craft::*;
 
 fn main() -> Result<()> {
     env_logger::Builder::from_env(env_logger::Env::default().default_filter_or("wal_craft=info"))
         .init();
-    let type_arg = &Arg::new("type")
-        .takes_value(true)
-        .help("Type of WAL to craft")
-        .possible_values([
-            Simple::NAME,
-            LastWalRecordXlogSwitch::NAME,
-            LastWalRecordXlogSwitchEndsOnPageBoundary::NAME,
-            WalRecordCrossingSegmentFollowedBySmallOne::NAME,
-            LastWalRecordCrossingSegment::NAME,
-        ])
-        .required(true);
-    let arg_matches = App::new("Postgres WAL crafter")
-        .about("Crafts Postgres databases with specific WAL properties")
-        .subcommand(
-            App::new("print-postgres-config")
-                .about("Print the configuration required for PostgreSQL server before running this script")
-        )
-        .subcommand(
-            App::new("with-initdb")
-                .about("Craft WAL in a new data directory first initialized with initdb")
-                .arg(type_arg)
-                .arg(
-                    Arg::new("datadir")
-                        .takes_value(true)
-                        .help("Data directory for the Postgres server")
-                        .required(true)
-                )
-                .arg(
-                    Arg::new("pg-distrib-dir")
-                        .long("pg-distrib-dir")
-                        .takes_value(true)
-                        .help("Directory with Postgres distributions (bin and lib directories, e.g. pg_install containing subpath `v14/bin/postgresql`)")
-                        .default_value("/usr/local")
-                )
-                .arg(
-                    Arg::new("pg-version")
-                    .long("pg-version")
-                    .help("Postgres version to use for the initial tenant")
-                    .required(true)
-                    .takes_value(true)
-                )
-        )
-        .subcommand(
-            App::new("in-existing")
-                .about("Craft WAL at an existing recently created Postgres database. Note that server may append new WAL entries on shutdown.")
-                .arg(type_arg)
-                .arg(
-                    Arg::new("connection")
-                        .takes_value(true)
-                        .help("Connection string to the Postgres database to populate")
-                        .required(true)
-                )
-        )
-        .get_matches();
+    let arg_matches = cli().get_matches();
 
     let wal_craft = |arg_matches: &ArgMatches, client| {
-        let (intermediate_lsns, end_of_wal_lsn) = match arg_matches.value_of("type").unwrap() {
+        let (intermediate_lsns, end_of_wal_lsn) = match arg_matches
+            .get_one::<String>("type")
+            .map(|s| s.as_str())
+            .context("'type' is required")?
+        {
             Simple::NAME => Simple::craft(client)?,
             LastWalRecordXlogSwitch::NAME => LastWalRecordXlogSwitch::craft(client)?,
             LastWalRecordXlogSwitchEndsOnPageBoundary::NAME => {
@@ -72,12 +23,12 @@ fn main() -> Result<()> {
                 WalRecordCrossingSegmentFollowedBySmallOne::craft(client)?
             }
             LastWalRecordCrossingSegment::NAME => LastWalRecordCrossingSegment::craft(client)?,
-            a => panic!("Unknown --type argument: {}", a),
+            a => panic!("Unknown --type argument: {a}"),
         };
         for lsn in intermediate_lsns {
-            println!("intermediate_lsn = {}", lsn);
+            println!("intermediate_lsn = {lsn}");
         }
-        println!("end_of_wal = {}", end_of_wal_lsn);
+        println!("end_of_wal = {end_of_wal_lsn}");
         Ok(())
     };
 
@@ -85,20 +36,24 @@ fn main() -> Result<()> {
         None => panic!("No subcommand provided"),
         Some(("print-postgres-config", _)) => {
             for cfg in REQUIRED_POSTGRES_CONFIG.iter() {
-                println!("{}", cfg);
+                println!("{cfg}");
             }
             Ok(())
         }
 
         Some(("with-initdb", arg_matches)) => {
             let cfg = Conf {
-                pg_version: arg_matches
-                    .value_of("pg-version")
-                    .unwrap()
-                    .parse::<u32>()
-                    .context("Failed to parse postgres version from the argument string")?,
-                pg_distrib_dir: arg_matches.value_of("pg-distrib-dir").unwrap().into(),
-                datadir: arg_matches.value_of("datadir").unwrap().into(),
+                pg_version: *arg_matches
+                    .get_one::<u32>("pg-version")
+                    .context("'pg-version' is required")?,
+                pg_distrib_dir: arg_matches
+                    .get_one::<PathBuf>("pg-distrib-dir")
+                    .context("'pg-distrib-dir' is required")?
+                    .to_owned(),
+                datadir: arg_matches
+                    .get_one::<PathBuf>("datadir")
+                    .context("'datadir' is required")?
+                    .to_owned(),
             };
             cfg.initdb()?;
             let srv = cfg.start_server()?;
@@ -108,9 +63,77 @@ fn main() -> Result<()> {
         }
         Some(("in-existing", arg_matches)) => wal_craft(
             arg_matches,
-            &mut postgres::Config::from_str(arg_matches.value_of("connection").unwrap())?
-                .connect(postgres::NoTls)?,
+            &mut postgres::Config::from_str(
+                arg_matches
+                    .get_one::<String>("connection")
+                    .context("'connection' is required")?,
+            )
+            .context(
+                "'connection' argument value could not be parsed as a postgres connection string",
+            )?
+            .connect(postgres::NoTls)?,
         ),
         Some(_) => panic!("Unknown subcommand"),
     }
 }
+
+fn cli() -> Command {
+    let type_arg = &Arg::new("type")
+        .help("Type of WAL to craft")
+        .value_parser([
+            Simple::NAME,
+            LastWalRecordXlogSwitch::NAME,
+            LastWalRecordXlogSwitchEndsOnPageBoundary::NAME,
+            WalRecordCrossingSegmentFollowedBySmallOne::NAME,
+            LastWalRecordCrossingSegment::NAME,
+        ])
+        .required(true);
+
+    Command::new("Postgres WAL crafter")
+        .about("Crafts Postgres databases with specific WAL properties")
+        .subcommand(
+            Command::new("print-postgres-config")
+                .about("Print the configuration required for PostgreSQL server before running this script")
+        )
+        .subcommand(
+            Command::new("with-initdb")
+                .about("Craft WAL in a new data directory first initialized with initdb")
+                .arg(type_arg)
+                .arg(
+                    Arg::new("datadir")
+                        .help("Data directory for the Postgres server")
+                        .value_parser(value_parser!(PathBuf))
+                        .required(true)
+                )
+                .arg(
+                    Arg::new("pg-distrib-dir")
+                        .long("pg-distrib-dir")
+                        .value_parser(value_parser!(PathBuf))
+                        .help("Directory with Postgres distributions (bin and lib directories, e.g. pg_install containing subpath `v14/bin/postgresql`)")
+                        .default_value("/usr/local")
+                )
+                .arg(
+                    Arg::new("pg-version")
+                    .long("pg-version")
+                    .help("Postgres version to use for the initial tenant")
+                    .value_parser(value_parser!(u32))
+                    .required(true)
+
+                )
+        )
+        .subcommand(
+            Command::new("in-existing")
+                .about("Craft WAL at an existing recently created Postgres database. Note that server may append new WAL entries on shutdown.")
+                .arg(type_arg)
+                .arg(
+                    Arg::new("connection")
+                        .help("Connection string to the Postgres database to populate")
+                        .required(true)
+                )
+        )
+}
+
+#[test]
+fn verify_cli() {
+    cli().debug_assert();
+}

libs/postgres_ffi/wal_craft/src/lib.rs

@@ -37,22 +37,22 @@ pub static REQUIRED_POSTGRES_CONFIG: Lazy<Vec<&'static str>> = Lazy::new(|| {
 });
 
 impl Conf {
-    pub fn pg_distrib_dir(&self) -> PathBuf {
+    pub fn pg_distrib_dir(&self) -> anyhow::Result<PathBuf> {
         let path = self.pg_distrib_dir.clone();
 
         match self.pg_version {
-            14 => path.join(format!("v{}", self.pg_version)),
-            15 => path.join(format!("v{}", self.pg_version)),
-            _ => panic!("Unsupported postgres version: {}", self.pg_version),
+            14 => Ok(path.join(format!("v{}", self.pg_version))),
+            15 => Ok(path.join(format!("v{}", self.pg_version))),
+            _ => bail!("Unsupported postgres version: {}", self.pg_version),
         }
     }
 
-    fn pg_bin_dir(&self) -> PathBuf {
-        self.pg_distrib_dir().join("bin")
+    fn pg_bin_dir(&self) -> anyhow::Result<PathBuf> {
+        Ok(self.pg_distrib_dir()?.join("bin"))
     }
 
-    fn pg_lib_dir(&self) -> PathBuf {
-        self.pg_distrib_dir().join("lib")
+    fn pg_lib_dir(&self) -> anyhow::Result<PathBuf> {
+        Ok(self.pg_distrib_dir()?.join("lib"))
     }
 
     pub fn wal_dir(&self) -> PathBuf {
@@ -60,12 +60,12 @@ impl Conf {
     }
 
     fn new_pg_command(&self, command: impl AsRef<Path>) -> Result<Command> {
-        let path = self.pg_bin_dir().join(command);
+        let path = self.pg_bin_dir()?.join(command);
         ensure!(path.exists(), "Command {:?} does not exist", path);
         let mut cmd = Command::new(path);
         cmd.env_clear()
-            .env("LD_LIBRARY_PATH", self.pg_lib_dir())
-            .env("DYLD_LIBRARY_PATH", self.pg_lib_dir());
+            .env("LD_LIBRARY_PATH", self.pg_lib_dir()?)
+            .env("DYLD_LIBRARY_PATH", self.pg_lib_dir()?);
         Ok(cmd)
     }
 

libs/pq_proto/Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "pq_proto"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+anyhow = "1.0"
+bytes = "1.0.1"
+pin-project-lite = "0.2.7"
+postgres-protocol = { git = "https://github.com/neondatabase/rust-postgres.git", rev="d052ee8b86fff9897c77b0fe89ea9daba0e1fa38" }
+rand = "0.8.3"
+serde = { version = "1.0", features = ["derive"] }
+tokio = { version = "1.17", features = ["macros"] }
+tracing = "0.1"
+
+workspace_hack = { version = "0.1", path = "../../workspace_hack" }

libs/pq_proto/src/lib.rs

@@ -2,7 +2,9 @@
 //! <https://www.postgresql.org/docs/devel/protocol-message-formats.html>
 //! on message formats.
 
-use crate::sync::{AsyncishRead, SyncFuture};
+// Tools for calling certain async methods in sync contexts.
+pub mod sync;
+
 use anyhow::{bail, ensure, Context, Result};
 use bytes::{Buf, BufMut, Bytes, BytesMut};
 use postgres_protocol::PG_EPOCH;
@@ -10,11 +12,13 @@ use serde::{Deserialize, Serialize};
 use std::{
     borrow::Cow,
     collections::HashMap,
+    fmt,
     future::Future,
     io::{self, Cursor},
     str,
     time::{Duration, SystemTime},
 };
+use sync::{AsyncishRead, SyncFuture};
 use tokio::io::AsyncReadExt;
 use tracing::{trace, warn};
 
@@ -124,6 +128,19 @@ pub struct CancelKeyData {
     pub cancel_key: i32,
 }
 
+impl fmt::Display for CancelKeyData {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        let hi = (self.backend_pid as u64) << 32;
+        let lo = self.cancel_key as u64;
+        let id = hi | lo;
+
+        // This format is more compact and might work better for logs.
+        f.debug_tuple("CancelKeyData")
+            .field(&format_args!("{:x}", id))
+            .finish()
+    }
+}
+
 use rand::distributions::{Distribution, Standard};
 impl Distribution<CancelKeyData> for Standard {
     fn sample<R: rand::Rng + ?Sized>(&self, rng: &mut R) -> CancelKeyData {
@@ -184,7 +201,7 @@ impl FeMessage {
     ///
     /// ```
     /// # use std::io;
-    /// # use utils::pq_proto::FeMessage;
+    /// # use pq_proto::FeMessage;
     /// #
     /// # fn process_message(msg: FeMessage) -> anyhow::Result<()> {
     /// #     Ok(())
@@ -288,6 +305,7 @@ impl FeStartupPacket {
                 Err(e) => return Err(e.into()),
             };
 
+            #[allow(clippy::manual_range_contains)]
             if len < 4 || len > MAX_STARTUP_PACKET_LENGTH {
                 bail!("invalid message length");
             }

libs/pq_proto/src/sync.rs

@@ -29,7 +29,7 @@ impl<S, T: Future> SyncFuture<S, T> {
     /// Example:
     ///
     /// ```
-    /// # use utils::sync::SyncFuture;
+    /// # use pq_proto::sync::SyncFuture;
     /// # use std::future::Future;
     /// # use tokio::io::AsyncReadExt;
     /// #

libs/remote_storage/Cargo.toml

@@ -15,7 +15,7 @@ serde = { version = "1.0", features = ["derive"] }
 serde_json = "1"
 tokio = { version = "1.17", features = ["sync", "macros", "fs", "io-util"] }
 tokio-util = { version = "0.7", features = ["io"] }
-toml_edit = { version = "0.13", features = ["easy"] }
+toml_edit = { version = "0.14", features = ["easy"] }
 tracing = "0.1.27"
 
 workspace_hack = { version = "0.1", path = "../../workspace_hack" }

libs/remote_storage/src/local_fs.rs

@@ -16,7 +16,7 @@ use tokio::{
     io::{self, AsyncReadExt, AsyncSeekExt, AsyncWriteExt},
 };
 use tracing::*;
-use utils::crashsafe_dir::path_with_suffix_extension;
+use utils::crashsafe::path_with_suffix_extension;
 
 use crate::{Download, DownloadError, RemoteObjectId};
 

libs/safekeeper_api/Cargo.toml

@@ -5,7 +5,7 @@ edition = "2021"
 
 [dependencies]
 serde = { version = "1.0", features = ["derive"] }
-serde_with = "1.12.0"
+serde_with = "2.0"
 const_format = "0.2.21"
 
 utils = { path = "../utils" }