doc: mention about spawn_blocking canceling

needed to use the github editor to have fast preview. I don't know how else to get the "Submitting Patches" look for the ol/li elements.

.github/ansible/.gitignore

@@ -1,5 +0,0 @@
-neon_install.tar.gz
-.neon_current_version
-
-collections/*
-!collections/.keep

.github/ansible/ansible.cfg

@@ -1,12 +0,0 @@
-[defaults]
-
-localhost_warning = False
-host_key_checking = False
-timeout = 30
-
-[ssh_connection]
-ssh_args   = -F ./ansible.ssh.cfg
-# teleport doesn't support sftp yet https://github.com/gravitational/teleport/issues/7127
-# and scp neither worked for me
-transfer_method = piped
-pipelining = True

.github/ansible/ansible.ssh.cfg

@@ -1,15 +0,0 @@
-# Remove this once https://github.com/gravitational/teleport/issues/10918 is fixed
-# (use pre 8.5 option name to cope with old ssh in CI)
-PubkeyAcceptedKeyTypes +ssh-rsa-cert-v01@openssh.com
-
-Host tele.zenith.tech
-    User admin
-    Port 3023
-    StrictHostKeyChecking no
-    UserKnownHostsFile /dev/null
-
-Host * !tele.zenith.tech
-    User admin
-    StrictHostKeyChecking no
-    UserKnownHostsFile /dev/null
-    ProxyJump tele.zenith.tech

.github/ansible/deploy.yaml

@@ -1,211 +0,0 @@
-- name: Upload Neon binaries
-  hosts: storage
-  gather_facts: False
-  remote_user: "{{ remote_user }}"
-
-  tasks:
-
-    - name: get latest version of Neon binaries
-      register: current_version_file
-      set_fact:
-        current_version: "{{ lookup('file', '.neon_current_version') | trim }}"
-      tags:
-      - pageserver
-      - safekeeper
-
-    - name: inform about versions
-      debug:
-        msg: "Version to deploy - {{ current_version }}"
-      tags:
-      - pageserver
-      - safekeeper
-
-    - name: upload and extract Neon binaries to /usr/local
-      ansible.builtin.unarchive:
-        owner: root
-        group: root
-        src: neon_install.tar.gz
-        dest: /usr/local
-      become: true
-      tags:
-      - pageserver
-      - safekeeper
-      - binaries
-      - putbinaries
-
-- name: Deploy pageserver
-  hosts: pageservers
-  gather_facts: False
-  remote_user: "{{ remote_user }}"
-
-  tasks:
-
-    - name: upload init script
-      when: console_mgmt_base_url is defined
-      ansible.builtin.template:
-        src: scripts/init_pageserver.sh
-        dest: /tmp/init_pageserver.sh
-        owner: root
-        group: root
-        mode: '0755'
-      become: true
-      tags:
-      - pageserver
-
-    - name: init pageserver
-      shell:
-        cmd: /tmp/init_pageserver.sh
-      args:
-        creates: "/storage/pageserver/data/tenants"
-      environment:
-        NEON_REPO_DIR: "/storage/pageserver/data"
-        LD_LIBRARY_PATH: "/usr/local/v14/lib"
-      become: true
-      tags:
-      - pageserver
-
-    - 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
-
-    # used in `pageserver.service` template
-    - name: learn current availability_zone
-      shell:
-        cmd: "curl -s http://169.254.169.254/latest/meta-data/placement/availability-zone"
-      register: ec2_availability_zone
-
-    - set_fact: 
-        ec2_availability_zone={{ ec2_availability_zone.stdout }}
-
-    - name: upload systemd service definition
-      ansible.builtin.template:
-        src: systemd/pageserver.service
-        dest: /etc/systemd/system/pageserver.service
-        owner: root
-        group: root
-        mode: '0644'
-      become: true
-      tags:
-      - pageserver
-
-    - name: start systemd service
-      ansible.builtin.systemd:
-        daemon_reload: yes
-        name: pageserver
-        enabled: yes
-        state: restarted
-      become: true
-      tags:
-      - pageserver
-
-    - name: post version to console
-      when: console_mgmt_base_url is defined
-      shell:
-        cmd: |
-          INSTANCE_ID=$(curl -s http://169.254.169.254/latest/meta-data/instance-id)
-          curl -sfS -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" {{ console_mgmt_base_url }}/management/api/v2/pageservers/$INSTANCE_ID | jq '.version = {{ current_version }}' > /tmp/new_version
-          curl -sfS -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" -H "Content-Type: application/json" -X POST -d@/tmp/new_version {{ console_mgmt_base_url }}/management/api/v2/pageservers
-      tags:
-      - pageserver
-
-- name: Deploy safekeeper
-  hosts: safekeepers
-  gather_facts: False
-  remote_user: "{{ remote_user }}"
-
-  tasks:
-
-    - name: upload init script
-      when: console_mgmt_base_url is defined
-      ansible.builtin.template:
-        src: scripts/init_safekeeper.sh
-        dest: /tmp/init_safekeeper.sh
-        owner: root
-        group: root
-        mode: '0755'
-      become: true
-      tags:
-      - safekeeper
-
-    - name: init safekeeper
-      shell:
-        cmd: /tmp/init_safekeeper.sh
-      args:
-        creates: "/storage/safekeeper/data/safekeeper.id"
-      environment:
-        NEON_REPO_DIR: "/storage/safekeeper/data"
-        LD_LIBRARY_PATH: "/usr/local/v14/lib"
-      become: true
-      tags:
-      - safekeeper
-
-    # used in `safekeeper.service` template
-    - name: learn current availability_zone
-      shell:
-        cmd: "curl -s http://169.254.169.254/latest/meta-data/placement/availability-zone"
-      register: ec2_availability_zone
-
-    - set_fact: 
-        ec2_availability_zone={{ ec2_availability_zone.stdout }}
-
-    # in the future safekeepers should discover pageservers byself
-    # but currently use first pageserver that was discovered
-    - name: set first pageserver var for safekeepers
-      set_fact:
-        first_pageserver: "{{ hostvars[groups['pageservers'][0]]['inventory_hostname'] }}"
-      tags:
-      - safekeeper
-
-    - name: upload systemd service definition
-      ansible.builtin.template:
-        src: systemd/safekeeper.service
-        dest: /etc/systemd/system/safekeeper.service
-        owner: root
-        group: root
-        mode: '0644'
-      become: true
-      tags:
-      - safekeeper
-
-    - name: start systemd service
-      ansible.builtin.systemd:
-        daemon_reload: yes
-        name: safekeeper
-        enabled: yes
-        state: restarted
-      become: true
-      tags:
-      - safekeeper
-
-    - name: post version to console
-      when: console_mgmt_base_url is defined
-      shell:
-        cmd: |
-          INSTANCE_ID=$(curl -s http://169.254.169.254/latest/meta-data/instance-id)
-          curl -sfS -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" {{ console_mgmt_base_url }}/management/api/v2/safekeepers/$INSTANCE_ID | jq '.version = {{ current_version }}' > /tmp/new_version
-          curl -sfS -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" -H "Content-Type: application/json" -X POST -d@/tmp/new_version {{ console_mgmt_base_url }}/management/api/v2/safekeepers
-      tags:
-      - safekeeper

.github/ansible/get_binaries.sh

@@ -1,42 +0,0 @@
-#!/bin/bash
-
-set -e
-
-if [ -n "${DOCKER_TAG}" ]; then
-  # Verson is DOCKER_TAG but without prefix
-  VERSION=$(echo $DOCKER_TAG | sed 's/^.*-//g')
-else
-  echo "Please set DOCKER_TAG environment variable"
-  exit 1
-fi
-
-
-# do initial cleanup
-rm -rf neon_install postgres_install.tar.gz neon_install.tar.gz .neon_current_version
-mkdir neon_install
-
-# retrieve binaries from docker image
-echo "getting binaries from docker image"
-docker pull --quiet neondatabase/neon:${DOCKER_TAG}
-ID=$(docker create neondatabase/neon:${DOCKER_TAG})
-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/storage_broker neon_install/bin/
-docker cp ${ID}:/usr/local/bin/proxy neon_install/bin/
-docker cp ${ID}:/usr/local/v14/bin/ neon_install/v14/bin/
-docker cp ${ID}:/usr/local/v15/bin/ neon_install/v15/bin/
-docker cp ${ID}:/usr/local/v14/lib/ neon_install/v14/lib/
-docker cp ${ID}:/usr/local/v15/lib/ neon_install/v15/lib/
-docker rm -vf ${ID}
-
-# store version to file (for ansible playbooks) and create binaries tarball
-echo ${VERSION} > neon_install/.neon_current_version
-echo ${VERSION} > .neon_current_version
-tar -czf neon_install.tar.gz -C neon_install .
-
-# do final cleaup
-rm -rf neon_install postgres_install.tar.gz

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

@@ -1,48 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-prod-storage-ap-southeast-1
-    bucket_region: ap-southeast-1
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.tech
-    broker_endpoint: http://storage-broker-lb.epsilon.ap-southeast-1.internal.aws.neon.tech:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 85 # TODO: decrease to 80 after all pageservers are below 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "10m"
-          threshold: &default_eviction_threshold "24h"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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
-    sentry_environment: production
-
-  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-2.ap-southeast-1.aws.neon.tech:
-          ansible_host:  i-04fb63634e4679eb9
-        safekeeper-3.ap-southeast-1.aws.neon.tech:
-          ansible_host:  i-05481f3bc88cfc2d4

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

@@ -1,50 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-prod-storage-eu-central-1
-    bucket_region: eu-central-1
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.tech
-    broker_endpoint: http://storage-broker-lb.gamma.eu-central-1.internal.aws.neon.tech:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 85 # TODO: decrease to 80 after all pageservers are below 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "10m"
-          threshold: &default_eviction_threshold "24h"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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
-    sentry_environment: production
-
-  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
-        pageserver-2.eu-central-1.aws.neon.tech:
-          ansible_host:  i-033584edf3f4b6742
-
-    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-1.hosts.yaml

@@ -1,50 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-prod-storage-us-east-1
-    bucket_region: us-east-1
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.tech
-    broker_endpoint: http://storage-broker-lb.theta.us-east-1.internal.aws.neon.tech:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 85 # TODO: decrease to 80 after all pageservers are below 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "10m"
-          threshold: &default_eviction_threshold "24h"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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-1
-    ansible_aws_ssm_bucket_name: neon-prod-storage-us-east-1
-    console_region_id: aws-us-east-1
-    sentry_environment: production
-
-  children:
-    pageservers:
-      hosts:
-        pageserver-0.us-east-1.aws.neon.tech:
-          ansible_host: i-085222088b0d2e0c7
-        pageserver-1.us-east-1.aws.neon.tech:
-          ansible_host: i-0969d4f684d23a21e
-        pageserver-2.us-east-1.aws.neon.tech:
-          ansible_host: i-05dee87895da58dad
-
-    safekeepers:
-      hosts:
-        safekeeper-0.us-east-1.aws.neon.tech:
-          ansible_host: i-04ce739e88793d864
-        safekeeper-1.us-east-1.aws.neon.tech:
-          ansible_host: i-0e9e6c9227fb81410
-        safekeeper-2.us-east-1.aws.neon.tech:
-          ansible_host: i-072f4dd86a327d52f

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

@@ -1,51 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-prod-storage-us-east-2
-    bucket_region: us-east-2
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.tech
-    broker_endpoint: http://storage-broker-lb.delta.us-east-2.internal.aws.neon.tech:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 85 # TODO: decrease to 80 after all pageservers are below 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "10m"
-          threshold: &default_eviction_threshold "24h"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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
-    sentry_environment: production
-
-  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
-        pageserver-2.us-east-2.aws.neon.tech:
-          ansible_host:  i-0d7a1c4325e71421d
-
-    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/prod.us-west-2.hosts.yaml

@@ -1,72 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-prod-storage-us-west-2
-    bucket_region: us-west-2
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.tech
-    broker_endpoint: http://storage-broker-lb.eta.us-west-2.internal.aws.neon.tech:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 85 # TODO: decrease to 80 after all pageservers are below 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "10m"
-          threshold: &default_eviction_threshold "24h"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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-west-2
-    ansible_aws_ssm_bucket_name: neon-prod-storage-us-west-2
-    console_region_id: aws-us-west-2-new
-    sentry_environment: production
-
-  children:
-    pageservers:
-      hosts:
-        pageserver-0.us-west-2.aws.neon.tech:
-          ansible_host: i-0d9f6dfae0e1c780d
-        pageserver-1.us-west-2.aws.neon.tech:
-          ansible_host: i-0c834be1dddba8b3f
-        pageserver-2.us-west-2.aws.neon.tech:
-          ansible_host: i-051642d372c0a4f32
-        pageserver-3.us-west-2.aws.neon.tech:
-          ansible_host: i-00c3844beb9ad1c6b
-        pageserver-4.us-west-2.aws.neon.tech:
-          ansible_host: i-013263dd1c239adcc
-        pageserver-5.us-west-2.aws.neon.tech:
-          ansible_host: i-00ca6417c7bf96820
-        pageserver-6.us-west-2.aws.neon.tech:
-          ansible_host: i-01cdf7d2bc1433b6a
-        pageserver-7.us-west-2.aws.neon.tech:
-          ansible_host: i-02eec9b40617db5bc
-
-    safekeepers:
-      hosts:
-        safekeeper-0.us-west-2.aws.neon.tech:
-          ansible_host: i-00719d8a74986fda6
-        safekeeper-1.us-west-2.aws.neon.tech:
-          ansible_host: i-074682f9d3c712e7c
-        safekeeper-2.us-west-2.aws.neon.tech:
-          ansible_host: i-042b7efb1729d7966
-        safekeeper-3.us-west-2.aws.neon.tech:
-          ansible_host: i-089f6b9ef426dff76
-        safekeeper-4.us-west-2.aws.neon.tech:
-          ansible_host: i-0fe6bf912c4710c82
-        safekeeper-5.us-west-2.aws.neon.tech:
-          ansible_host: i-0a83c1c46d2b4e409
-        safekeeper-6.us-west-2.aws.neon.tech:
-          ansible_host: i-0fef5317b8fdc9f8d
-        safekeeper-7.us-west-2.aws.neon.tech:
-          ansible_host: i-0be739190d4289bf9
-        safekeeper-8.us-west-2.aws.neon.tech:
-          ansible_host: i-00e851803669e5cfe                    

.github/ansible/scripts/init_pageserver.sh

@@ -1,37 +0,0 @@
-#!/bin/sh
-
-# fetch params from meta-data service
-INSTANCE_ID=$(curl -s http://169.254.169.254/latest/meta-data/instance-id)
-AZ_ID=$(curl -s http://169.254.169.254/latest/meta-data/placement/availability-zone)
-INSTANCE_TYPE=$(curl -s http://169.254.169.254/latest/meta-data/instance-type)
-DISK_SIZE=$(df -B1 /storage | tail -1 | awk '{print $2}')
-
-# store fqdn hostname in var
-HOST=$(hostname -f)
-
-
-cat <<EOF | tee /tmp/payload
-{
-  "version": 1,
-  "host": "${HOST}",
-  "port": 6400,
-  "region_id": "{{ console_region_id }}",
-  "instance_id": "${INSTANCE_ID}",
-  "http_host": "${HOST}",
-  "http_port": 9898,
-  "active": false,
-  "availability_zone_id": "${AZ_ID}",
-  "disk_size": ${DISK_SIZE},
-  "instance_type": "${INSTANCE_TYPE}"
-}
-EOF
-
-# check if pageserver already registered or not
-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 -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" -H "Content-Type: application/json" {{ 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
-fi

.github/ansible/scripts/init_safekeeper.sh

@@ -1,31 +0,0 @@
-#!/bin/sh
-
-# fetch params from meta-data service
-INSTANCE_ID=$(curl -s http://169.254.169.254/latest/meta-data/instance-id)
-AZ_ID=$(curl -s http://169.254.169.254/latest/meta-data/placement/availability-zone)
-
-# store fqdn hostname in var
-HOST=$(hostname -f)
-
-
-cat <<EOF | tee /tmp/payload
-{
-  "version": 1,
-  "host": "${HOST}",
-  "port": 6500,
-  "http_port": 7676,
-  "region_id": "{{ console_region_id }}",
-  "instance_id": "${INSTANCE_ID}",
-  "availability_zone_id": "${AZ_ID}",
-  "active": false
-}
-EOF
-
-# check if safekeeper already registered or not
-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 -H "Authorization: Bearer {{ CONSOLE_API_TOKEN }}" -H "Content-Type: application/json" {{ 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

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

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

@@ -1,47 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-dev-storage-eu-central-1
-    bucket_region: eu-central-1
-    # We only register/update storage in one preview console and manually copy to other instances
-    console_mgmt_base_url: http://neon-internal-api.helium.aws.neon.build
-    broker_endpoint: http://storage-broker-lb.alpha.eu-central-1.internal.aws.neon.build:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.helium.aws.neon.build/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "20m"
-          threshold: &default_eviction_threshold "20m"
-      evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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-dev-storage-eu-central-1
-    console_region_id: aws-eu-central-1
-    sentry_environment: staging
-
-  children:
-    pageservers:
-      hosts:
-        pageserver-0.eu-central-1.aws.neon.build:
-          ansible_host: i-011f93ec26cfba2d4
-
-    safekeepers:
-      hosts:
-        safekeeper-0.eu-central-1.aws.neon.build:
-          ansible_host: i-0ff026d27babf8ddd
-        safekeeper-1.eu-central-1.aws.neon.build:
-          ansible_host: i-03983a49ee54725d9
-        safekeeper-2.eu-central-1.aws.neon.build:
-          ansible_host: i-0bd025ecdb61b0db3

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

@@ -1,60 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-dev-storage-eu-west-1
-    bucket_region: eu-west-1
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.build
-    broker_endpoint: http://storage-broker-lb.zeta.eu-west-1.internal.aws.neon.build:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.build/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "20m"
-          threshold: &default_eviction_threshold "20m"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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
-    sentry_environment: staging
-
-  children:
-    pageservers:
-      hosts:
-        pageserver-0.eu-west-1.aws.neon.build:
-          ansible_host: i-01d496c5041c7f34c
-        pageserver-1.eu-west-1.aws.neon.build:
-          ansible_host: i-0e8013e239ce3928c
-
-    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
-        safekeeper-3.eu-west-1.aws.neon.build:
-          ansible_host: i-0a6f91660e99e8891
-        safekeeper-4.eu-west-1.aws.neon.build:
-          ansible_host: i-0012e309e28e7c249
-        safekeeper-5.eu-west-1.aws.neon.build:
-          ansible_host: i-085a2b1193287b32e
-        safekeeper-6.eu-west-1.aws.neon.build:
-          ansible_host: i-0c713248465ed0fbd
-        safekeeper-7.eu-west-1.aws.neon.build:
-          ansible_host: i-02ad231aed2a80b7a
-        safekeeper-8.eu-west-1.aws.neon.build:
-          ansible_host: i-0dbbd8ffef66efda8

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

@@ -1,56 +0,0 @@
-storage:
-  vars:
-    bucket_name: neon-staging-storage-us-east-2
-    bucket_region: us-east-2
-    console_mgmt_base_url: http://neon-internal-api.aws.neon.build
-    broker_endpoint: http://storage-broker-lb.beta.us-east-2.internal.aws.neon.build:50051
-    pageserver_config_stub:
-      pg_distrib_dir: /usr/local
-      metric_collection_endpoint: http://neon-internal-api.aws.neon.build/billing/api/v1/usage_events
-      metric_collection_interval: 10min
-      disk_usage_based_eviction:
-        max_usage_pct: 80
-        min_avail_bytes: 0
-        period: "10s"
-      tenant_config:
-        eviction_policy:
-          kind: "LayerAccessThreshold"
-          period: "20m"
-          threshold: &default_eviction_threshold "20m"
-        evictions_low_residence_duration_metric_threshold: *default_eviction_threshold
-      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
-    sentry_environment: staging
-
-  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
-        pageserver-2.us-east-2.aws.neon.build:
-          ansible_host: i-01e31cdf7e970586a
-        pageserver-3.us-east-2.aws.neon.build:
-          ansible_host: i-0602a0291365ef7cc
-        pageserver-99.us-east-2.aws.neon.build:
-          ansible_host: i-0c39491109bb88824
-
-    safekeepers:
-      hosts:
-        safekeeper-0.us-east-2.aws.neon.build:
-          ansible_host: i-027662bd552bf5db0
-        safekeeper-2.us-east-2.aws.neon.build:
-          ansible_host: i-0de0b03a51676a6ce
-        safekeeper-3.us-east-2.aws.neon.build:
-          ansible_host: i-05f8ba2cda243bd18
-        safekeeper-99.us-east-2.aws.neon.build:
-          ansible_host: i-0d61b6a2ea32028d5

.github/ansible/systemd/pageserver.service

@@ -1,18 +0,0 @@
-[Unit]
-Description=Neon pageserver
-After=network.target auditd.service
-
-[Service]
-Type=simple
-User=pageserver
-Environment=RUST_BACKTRACE=1 NEON_REPO_DIR=/storage/pageserver LD_LIBRARY_PATH=/usr/local/v14/lib SENTRY_DSN={{ SENTRY_URL_PAGESERVER }} SENTRY_ENVIRONMENT={{ sentry_environment }}
-ExecStart=/usr/local/bin/pageserver -c "pg_distrib_dir='/usr/local'" -c "listen_pg_addr='0.0.0.0:6400'" -c "listen_http_addr='0.0.0.0:9898'" -c "broker_endpoint='{{ broker_endpoint }}'" -c "availability_zone='{{ ec2_availability_zone }}'" -D /storage/pageserver/data
-ExecReload=/bin/kill -HUP $MAINPID
-KillMode=mixed
-KillSignal=SIGINT
-Restart=on-failure
-TimeoutSec=10
-LimitNOFILE=30000000
-
-[Install]
-WantedBy=multi-user.target

.github/ansible/systemd/safekeeper.service

@@ -1,18 +0,0 @@
-[Unit]
-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 SENTRY_DSN={{ SENTRY_URL_SAFEKEEPER }} SENTRY_ENVIRONMENT={{ sentry_environment }}
-ExecStart=/usr/local/bin/safekeeper -l {{ inventory_hostname }}{{ hostname_suffix }}:6500 --listen-http {{ inventory_hostname }}{{ hostname_suffix }}:7676 -D /storage/safekeeper/data --broker-endpoint={{ broker_endpoint }} --remote-storage='{bucket_name="{{bucket_name}}", bucket_region="{{bucket_region}}", prefix_in_bucket="{{ safekeeper_s3_prefix }}"}' --availability-zone={{ ec2_availability_zone }}
-ExecReload=/bin/kill -HUP $MAINPID
-KillMode=mixed
-KillSignal=SIGINT
-Restart=on-failure
-TimeoutSec=10
-LimitNOFILE=30000000
-
-[Install]
-WantedBy=multi-user.target

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

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

.github/helm-values/dev-eu-central-1-alpha.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: staging
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.alpha.eu-central-1.internal.aws.neon.build
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "staging"

.github/helm-values/dev-eu-central-1-alpha.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.alpha.eu-central-1.internal.aws.neon.build"
-
-settings:
-  domain: "*.snirouter.alpha.eu-central-1.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

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

@@ -1,76 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.build/management/api/v2"
-  domain: "*.eu-west-1.aws.neon.build"
-  otelExporterOtlpEndpoint: "https://otel-collector.zeta.eu-west-1.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.build/billing/api/v1/usage_events"
-  metricCollectionInterval: "1min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: dev
-  neon_region: 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
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/dev-eu-west-1-zeta.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: staging
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.zeta.eu-west-1.internal.aws.neon.build
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "staging"

.github/helm-values/dev-eu-west-1-zeta.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.zeta.eu-west-1.internal.aws.neon.build"
-
-settings:
-  domain: "*.snirouter.zeta.eu-west-1.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

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

@@ -1,68 +0,0 @@
-# Helm chart values for neon-proxy-link.
-# This is a YAML-formatted file.
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "link"
-  authEndpoint: "https://console.stage.neon.tech/authenticate_proxy_request/"
-  uri: "https://console.stage.neon.tech/psql_session/"
-  domain: "pg.neon.build"
-  otelExporterOtlpEndpoint: "https://otel-collector.beta.us-east-2.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.build/billing/api/v1/usage_events"
-  metricCollectionInterval: "1min"
-
-# -- Additional labels for neon-proxy-link pods
-podLabels:
-  neon_service: proxy
-  neon_env: dev
-  neon_region: us-east-2
-
-service:
-  type: LoadBalancer
-  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: internal
-    external-dns.alpha.kubernetes.io/hostname: neon-proxy-link-mgmt.beta.us-east-2.aws.neon.build
-
-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: neon-proxy-link.beta.us-east-2.aws.neon.build
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

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

@@ -1,77 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.build/management/api/v2"
-  domain: "*.cloud.stage.neon.tech"
-  otelExporterOtlpEndpoint: "https://otel-collector.beta.us-east-2.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.build/billing/api/v1/usage_events"
-  metricCollectionInterval: "1min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram-legacy
-  neon_env: dev
-  neon_region: 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: neon-proxy-scram-legacy.beta.us-east-2.aws.neon.build
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

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

@@ -1,78 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.build/management/api/v2"
-  domain: "*.us-east-2.aws.neon.build"
-  extraDomains: ["*.us-east-2.postgres.zenith.tech", "*.us-east-2.retooldb-staging.com"]
-  otelExporterOtlpEndpoint: "https://otel-collector.beta.us-east-2.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.build/billing/api/v1/usage_events"
-  metricCollectionInterval: "1min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: dev
-  neon_region: 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
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/dev-us-east-2-beta.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: staging
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.beta.us-east-2.internal.aws.neon.build
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "staging"

.github/helm-values/dev-us-east-2-beta.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.beta.us-east-2.internal.aws.neon.build"
-
-settings:
-  domain: "*.snirouter.beta.us-east-2.internal.aws.neon.build"
-  sentryEnvironment: "staging"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

.github/helm-values/preview-template.neon-proxy-scram.yaml

@@ -1,67 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.${PREVIEW_NAME}.aws.neon.build/management/api/v2"
-  domain: "*.cloud.${PREVIEW_NAME}.aws.neon.build"
-  sentryEnvironment: "staging"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.${PREVIEW_NAME}.aws.neon.build/billing/api/v1/usage_events"
-  metricCollectionInterval: "1min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: test
-  neon_region: ${PREVIEW_NAME}.eu-central-1
-
-
-exposedService:
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internet-facing
-    external-dns.alpha.kubernetes.io/hostname: cloud.${PREVIEW_NAME}.aws.neon.build
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

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

@@ -1,77 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.tech/management/api/v2"
-  domain: "*.ap-southeast-1.aws.neon.tech"
-  extraDomains: ["*.ap-southeast-1.retooldb.com", "*.ap-southeast-1.postgres.vercel-storage.com"]
-  sentryEnvironment: "production"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events"
-  metricCollectionInterval: "10min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: prod
-  neon_region: 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
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/prod-ap-southeast-1-epsilon.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: production
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.epsilon.ap-southeast-1.internal.aws.neon.tech
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "production"

.github/helm-values/prod-ap-southeast-1-epsilon.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.epsilon.ap-southeast-1.internal.aws.neon.tech"
-
-settings:
-  domain: "*.snirouter.epsilon.ap-southeast-1.internal.aws.neon.tech"
-  sentryEnvironment: "production"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

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

@@ -1,77 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.tech/management/api/v2"
-  domain: "*.eu-central-1.aws.neon.tech"
-  extraDomains: ["*.eu-central-1.retooldb.com", "*.eu-central-1.postgres.vercel-storage.com"]
-  sentryEnvironment: "production"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events"
-  metricCollectionInterval: "10min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: prod
-  neon_region: 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
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/prod-eu-central-1-gamma.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: production
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.gamma.eu-central-1.internal.aws.neon.tech
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "production"

.github/helm-values/prod-eu-central-1-gamma.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.gamma.eu-central-1.internal.aws.neon.tech"
-
-settings:
-  domain: "*.snirouter.gamma.eu-central-1.internal.aws.neon.tech"
-  sentryEnvironment: "production"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

.github/helm-values/prod-us-east-1-theta.neon-proxy-scram.yaml

@@ -1,69 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.tech/management/api/v2"
-  domain: "*.us-east-1.aws.neon.tech"
-  # *.us-east-1.retooldb.com hasn't been delegated yet.
-  extraDomains: ["*.us-east-1.postgres.vercel-storage.com"]
-  sentryEnvironment: "production"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events"
-  metricCollectionInterval: "10min"
-
-podLabels:
-  neon_service: proxy-scram
-  neon_env: prod
-  neon_region: us-east-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: us-east-1.aws.neon.tech
-  httpsPort: 443
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/prod-us-east-1-theta.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: production
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.theta.us-east-1.internal.aws.neon.tech
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "production"

.github/helm-values/prod-us-east-1-theta.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.theta.us-east-1.internal.aws.neon.tech"
-
-settings:
-  domain: "*.snirouter.theta.us-east-1.internal.aws.neon.tech"
-  sentryEnvironment: "production"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

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

@@ -1,58 +0,0 @@
-# Helm chart values for neon-proxy-link.
-# This is a YAML-formatted file.
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "link"
-  authEndpoint: "https://console.neon.tech/authenticate_proxy_request/"
-  uri: "https://console.neon.tech/psql_session/"
-  domain: "pg.neon.tech"
-  sentryEnvironment: "production"
-
-# -- Additional labels for zenith-proxy pods
-podLabels:
-  neon_service: proxy
-  neon_env: production
-  neon_region: us-east-2
-
-service:
-  type: LoadBalancer
-  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: internal
-    external-dns.alpha.kubernetes.io/hostname: neon-proxy-link-mgmt.delta.us-east-2.aws.neon.tech
-
-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: neon-proxy-link.delta.us-east-2.aws.neon.tech
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

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

@@ -1,77 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.tech/management/api/v2"
-  domain: "*.us-east-2.aws.neon.tech"
-  extraDomains: ["*.us-east-2.retooldb.com", "*.us-east-2.postgres.vercel-storage.com"]
-  sentryEnvironment: "production"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events"
-  metricCollectionInterval: "10min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: prod
-  neon_region: 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
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/prod-us-east-2-delta.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: production
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.delta.us-east-2.internal.aws.neon.tech
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "production"

.github/helm-values/prod-us-east-2-delta.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.delta.us-east-2.internal.aws.neon.tech"
-
-settings:
-  domain: "*.snirouter.delta.us-east-2.internal.aws.neon.tech"
-  sentryEnvironment: "production"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

.github/helm-values/prod-us-west-2-eta.neon-proxy-scram-legacy.yaml

@@ -1,76 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.tech/management/api/v2"
-  domain: "*.cloud.neon.tech"
-  sentryEnvironment: "production"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events"
-  metricCollectionInterval: "10min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: prod
-  neon_region: us-west-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: neon-proxy-scram-legacy.eta.us-west-2.aws.neon.tech
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/prod-us-west-2-eta.neon-proxy-scram.yaml

@@ -1,77 +0,0 @@
-# Helm chart values for neon-proxy-scram.
-# This is a YAML-formatted file.
-
-deploymentStrategy:
-  type: RollingUpdate
-  rollingUpdate:
-    maxSurge: 100%
-    maxUnavailable: 50%
-
-# Delay the kill signal by 5 minutes (5 * 60)
-# The pod(s) will stay in Terminating, keeps the existing connections
-# but doesn't receive new ones
-containerLifecycle:
-  preStop:
-    exec:
-      command: ["/bin/sh", "-c", "sleep 300"]
-terminationGracePeriodSeconds: 604800
-
-
-image:
-  repository: neondatabase/neon
-
-settings:
-  authBackend: "console"
-  authEndpoint: "http://neon-internal-api.aws.neon.tech/management/api/v2"
-  domain: "*.us-west-2.aws.neon.tech"
-  extraDomains: ["*.us-west-2.retooldb.com", "*.us-west-2.postgres.vercel-storage.com"]
-  sentryEnvironment: "production"
-  wssPort: 8443
-  metricCollectionEndpoint: "http://neon-internal-api.aws.neon.tech/billing/api/v1/usage_events"
-  metricCollectionInterval: "10min"
-
-# -- Additional labels for neon-proxy pods
-podLabels:
-  neon_service: proxy-scram
-  neon_env: prod
-  neon_region: us-west-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-west-2.aws.neon.tech
-  httpsPort: 443
-
-#metrics:
-#  enabled: true
-#  serviceMonitor:
-#    enabled: true
-#    selector:
-#      release: kube-prometheus-stack
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-proxy.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-proxy-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-proxy
-        app.kubernetes.io/instance: "{{ include \"neon-proxy.fullname\" . }}"
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-proxy"
-      endpoints:
-        - port: http
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"

.github/helm-values/prod-us-west-2-eta.neon-storage-broker.yaml

@@ -1,52 +0,0 @@
-# Helm chart values for neon-storage-broker
-podLabels:
-  neon_env: production
-  neon_service: storage-broker
-
-# Use L4 LB
-service:
-  # service.annotations -- Annotations to add to the service
-  annotations:
-    service.beta.kubernetes.io/aws-load-balancer-type: external  # use newer AWS Load Balancer Controller
-    service.beta.kubernetes.io/aws-load-balancer-nlb-target-type: ip
-    service.beta.kubernetes.io/aws-load-balancer-scheme: internal  # deploy LB to private subnet
-    # assign service to this name at external-dns
-    external-dns.alpha.kubernetes.io/hostname: storage-broker-lb.eta.us-west-2.internal.aws.neon.tech
-  # service.type -- Service type
-  type: LoadBalancer
-  # service.port -- broker listen port
-  port: 50051
-
-ingress:
-  enabled: false
-
-metrics:
-  enabled: false
-
-extraManifests:
-  - apiVersion: operator.victoriametrics.com/v1beta1
-    kind: VMServiceScrape
-    metadata:
-      name: "{{ include \"neon-storage-broker.fullname\" . }}"
-      labels:
-        helm.sh/chart: neon-storage-broker-{{ .Chart.Version }}
-        app.kubernetes.io/name: neon-storage-broker
-        app.kubernetes.io/instance: neon-storage-broker
-        app.kubernetes.io/version: "{{ .Chart.AppVersion }}"
-        app.kubernetes.io/managed-by: Helm
-      namespace: "{{ .Release.Namespace }}"
-    spec:
-      selector:
-        matchLabels:
-          app.kubernetes.io/name: "neon-storage-broker"
-      endpoints:
-        - port: broker
-          path: /metrics
-          interval: 10s
-          scrapeTimeout: 10s
-      namespaceSelector:
-        matchNames:
-          - "{{ .Release.Namespace }}"
-
-settings:
-  sentryEnvironment: "production"

.github/helm-values/prod-us-west-2-eta.pg-sni-router.yaml

@@ -1,19 +0,0 @@
-useCertManager: true
-
-replicaCount: 3
-
-exposedService:
-  # exposedService.port -- Exposed Service proxy port
-  port: 4432
-  annotations:
-    external-dns.alpha.kubernetes.io/hostname: "*.snirouter.eta.us-west-2.internal.aws.neon.tech"
-
-settings:
-  domain: "*.snirouter.eta.us-west-2.internal.aws.neon.tech"
-  sentryEnvironment: "production"
-
-imagePullSecrets:
-  - name: docker-hub-neon
-
-metrics:
-  enabled: false

.github/workflows/build_and_test.yml

@@ -630,6 +630,7 @@ jobs:
                            --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache
                            --context .
                            --build-arg GIT_VERSION=${{ github.sha }}
+                           --build-arg REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
                            --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}}
                            --destination neondatabase/neon:${{needs.tag.outputs.build-tag}}
 
@@ -715,6 +716,7 @@ jobs:
                            --cache-repo 369495373322.dkr.ecr.eu-central-1.amazonaws.com/cache
                            --context .
                            --build-arg GIT_VERSION=${{ github.sha }}
+                           --build-arg REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
                            --dockerfile Dockerfile.compute-tools
                            --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{needs.tag.outputs.build-tag}}
                            --destination neondatabase/compute-tools:${{needs.tag.outputs.build-tag}}
@@ -767,6 +769,7 @@ jobs:
                            --context .
                            --build-arg GIT_VERSION=${{ github.sha }}
                            --build-arg PG_VERSION=${{ matrix.version }}
+                           --build-arg REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
                            --dockerfile Dockerfile.compute-node
                            --destination 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
                            --destination neondatabase/compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
@@ -932,42 +935,6 @@ jobs:
       - name: Cleanup ECR folder
         run: rm -rf ~/.ecr
 
-  deploy-pr-test-new:
-    runs-on: [ self-hosted, gen3, small ]
-    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: [ promote-images, 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 }} -e SENTRY_URL_PAGESERVER=${{ secrets.SENTRY_URL_PAGESERVER }} -e SENTRY_URL_SAFEKEEPER=${{ secrets.SENTRY_URL_SAFEKEEPER }}
-          rm -f neon_install.tar.gz .neon_current_version
-
-      - name: Cleanup ansible folder
-        run: rm -rf ~/.ansible
-
   deploy:
     runs-on: [ self-hosted, gen3, small ]
     container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
@@ -992,12 +959,12 @@ jobs:
 
       - name: Trigger deploy workflow
         env:
-          GH_TOKEN: ${{ github.token }}
+          GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
         run: |
           if [[ "$GITHUB_REF_NAME" == "main" ]]; then
-            gh workflow run deploy-dev.yml --ref main -f branch=${{ github.sha }} -f dockerTag=${{needs.tag.outputs.build-tag}}
+            gh workflow --repo neondatabase/aws run deploy-dev.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}}
           elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
-            gh workflow run deploy-prod.yml --ref release -f branch=${{ github.sha }} -f dockerTag=${{needs.tag.outputs.build-tag}} -f disclamerAcknowledged=true
+            gh workflow --repo neondatabase/aws run deploy-prod.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}} -f disclamerAcknowledged=true
           else
             echo "GITHUB_REF_NAME (value '$GITHUB_REF_NAME') is not set to either 'main' or 'release'"
             exit 1

.github/workflows/deploy-dev.yml

@@ -1,280 +0,0 @@
-name: Neon Deploy dev
-
-on:
-  workflow_dispatch:
-    inputs:
-      dockerTag:
-        description: 'Docker tag to deploy'
-        required: true
-        type: string
-      branch:
-        description: 'Branch or commit used for deploy scripts and configs'
-        required: true
-        type: string
-        default: 'main'
-      deployStorage:
-        description: 'Deploy storage'
-        required: true
-        type: boolean
-        default: true
-      deployProxy:
-        description: 'Deploy proxy'
-        required: true
-        type: boolean
-        default: true
-      deployStorageBroker:
-        description: 'Deploy storage-broker'
-        required: true
-        type: boolean
-        default: true
-      deployPgSniRouter:
-        description: 'Deploy pg-sni-router'
-        required: true
-        type: boolean
-        default: true
-
-env:
-  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-concurrency:
-  group: deploy-dev
-  cancel-in-progress: false
-
-jobs:
-  deploy-storage-new:
-    runs-on: [ self-hosted, gen3, small ]
-    container:
-      image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
-      options: --user root --privileged
-    if: inputs.deployStorage
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        # TODO(sergey): Fix storage deploy in eu-central-1
-        target_region: [ eu-west-1, us-east-2]
-    environment:
-      name: dev-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-
-      - name: Redeploy
-        run: |
-          export DOCKER_TAG=${{ inputs.dockerTag }}
-          cd "$(pwd)/.github/ansible"
-
-          ./get_binaries.sh
-
-          ansible-galaxy collection install sivel.toiletwater
-          ansible-playbook -v deploy.yaml -i staging.${{ matrix.target_region }}.hosts.yaml -e @ssm_config -e CONSOLE_API_TOKEN=${{ secrets.NEON_STAGING_API_KEY }} -e SENTRY_URL_PAGESERVER=${{ secrets.SENTRY_URL_PAGESERVER }} -e SENTRY_URL_SAFEKEEPER=${{ secrets.SENTRY_URL_SAFEKEEPER }}
-          rm -f neon_install.tar.gz .neon_current_version
-
-      - name: Cleanup ansible folder
-        run: rm -rf ~/.ansible
-
-  deploy-proxy-new:
-    runs-on: [ self-hosted, gen3, small ]
-    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
-    if: inputs.deployProxy
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        include:
-          - target_region:  us-east-2
-            target_cluster: dev-us-east-2-beta
-            deploy_link_proxy: true
-            deploy_legacy_scram_proxy: true
-          - target_region:  eu-west-1
-            target_cluster: dev-eu-west-1-zeta
-            deploy_link_proxy: false
-            deploy_legacy_scram_proxy: false
-    environment:
-      name: dev-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-  
-      - name: Configure AWS Credentials
-        uses: aws-actions/configure-aws-credentials@v1-node16
-        with:
-          role-to-assume: arn:aws:iam::369495373322:role/github-runner
-          aws-region: eu-central-1
-          role-skip-session-tagging: true
-          role-duration-seconds: 1800
-  
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-  
-      - name: Re-deploy scram proxy
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          helm upgrade neon-proxy-scram neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-scram.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-  
-      - name: Re-deploy link proxy
-        if: matrix.deploy_link_proxy
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          helm upgrade neon-proxy-link neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-link.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-  
-      - name: Re-deploy legacy scram proxy
-        if: matrix.deploy_legacy_scram_proxy
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          helm upgrade neon-proxy-scram-legacy neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-scram-legacy.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-  
-      - name: Cleanup helm folder
-        run: rm -rf ~/.cache
-
-  deploy-preview-proxy-new:
-    runs-on: [ self-hosted, gen3, small ]
-    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
-    if: inputs.deployProxy
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        include:
-          - target_region:  eu-central-1
-            target_cluster: dev-eu-central-1-alpha
-    environment:
-      name: dev-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-  
-      - name: Configure AWS Credentials
-        uses: aws-actions/configure-aws-credentials@v1-node16
-        with:
-          role-to-assume: arn:aws:iam::369495373322:role/github-runner
-          aws-region: eu-central-1
-          role-skip-session-tagging: true
-          role-duration-seconds: 1800
-  
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-  
-      - name: Re-deploy preview proxies
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          for PREVIEW_NAME in helium argon krypton xenon radon oganesson hydrogen nitrogen oxygen fluorine chlorine; do
-            export PREVIEW_NAME
-            envsubst <.github/helm-values/preview-template.neon-proxy-scram.yaml >preview-${PREVIEW_NAME}.neon-proxy-scram.yaml
-            helm upgrade neon-proxy-scram-${PREVIEW_NAME} neondatabase/neon-proxy --namespace neon-proxy-${PREVIEW_NAME} --create-namespace --install --atomic -f preview-${PREVIEW_NAME}.neon-proxy-scram.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-          done
-
-      - name: Cleanup helm folder
-        run: rm -rf ~/.cache
-  
-  deploy-storage-broker-new:
-    runs-on: [ self-hosted, gen3, small ]
-    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
-    if: inputs.deployStorageBroker
-    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
-          - target_region:  eu-central-1
-            target_cluster: dev-eu-central-1-alpha
-    environment:
-      name: dev-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-  
-      - name: Configure AWS Credentials
-        uses: aws-actions/configure-aws-credentials@v1-node16
-        with:
-          role-to-assume: arn:aws:iam::369495373322:role/github-runner
-          aws-region: eu-central-1
-          role-skip-session-tagging: true
-          role-duration-seconds: 1800
-  
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-  
-      - name: Deploy storage-broker
-        run:
-          helm upgrade neon-storage-broker-lb neondatabase/neon-storage-broker --namespace neon-storage-broker-lb --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-storage-broker.yaml --set image.tag=${{ inputs.dockerTag }} --set settings.sentryUrl=${{ secrets.SENTRY_URL_BROKER }} --wait --timeout 5m0s
-  
-      - name: Cleanup helm folder
-        run: rm -rf ~/.cache
-
-  deploy-pg-sni-router:
-    runs-on: [ self-hosted, gen3, small ]
-    container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:pinned
-    if: inputs.deployPgSniRouter
-    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
-          - target_region:  eu-central-1
-            target_cluster: dev-eu-central-1-alpha
-    environment:
-      name: dev-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-  
-      - name: Configure AWS Credentials
-        uses: aws-actions/configure-aws-credentials@v1-node16
-        with:
-          role-to-assume: arn:aws:iam::369495373322:role/github-runner
-          aws-region: eu-central-1
-          role-skip-session-tagging: true
-          role-duration-seconds: 1800
-  
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-  
-      - name: Deploy pg-sni-router
-        run:
-          helm upgrade neon-pg-sni-router neondatabase/neon-pg-sni-router --namespace neon-pg-sni-router --create-namespace --install --debug --atomic -f .github/helm-values/${{ matrix.target_cluster }}.pg-sni-router.yaml --set image.tag=${{ inputs.dockerTag }} --set settings.sentryUrl=${{ secrets.SENTRY_URL_BROKER }} --wait --timeout 15m0s
-  
-      - name: Cleanup helm folder
-        run: rm -rf ~/.cache

.github/workflows/deploy-prod.yml

@@ -1,217 +0,0 @@
-name: Neon Deploy prod
-
-on:
-  workflow_dispatch:
-    inputs:
-      dockerTag:
-        description: 'Docker tag to deploy'
-        required: true
-        type: string
-      branch:
-        description: 'Branch or commit used for deploy scripts and configs'
-        required: true
-        type: string
-        default: 'release'
-      deployStorage:
-        description: 'Deploy storage'
-        required: true
-        type: boolean
-        default: true
-      deployProxy:
-        description: 'Deploy proxy'
-        required: true
-        type: boolean
-        default: true
-      deployStorageBroker:
-        description: 'Deploy storage-broker'
-        required: true
-        type: boolean
-        default: true
-      deployPgSniRouter:
-        description: 'Deploy pg-sni-router'
-        required: true
-        type: boolean
-        default: true
-      disclamerAcknowledged:
-        description: 'I confirm that there is an emergency and I can not use regular release workflow'
-        required: true
-        type: boolean
-        default: false
-
-concurrency:
-  group: deploy-prod
-  cancel-in-progress: false
-
-jobs:
-  deploy-prod-new:
-    runs-on: prod
-    container:
-      image: 093970136003.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
-      options: --user root --privileged
-    if: inputs.deployStorage && inputs.disclamerAcknowledged
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        target_region: [ us-east-2, us-west-2, eu-central-1, ap-southeast-1, us-east-1 ]
-    environment:
-      name: prod-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-
-      - name: Redeploy
-        run: |
-          export DOCKER_TAG=${{ inputs.dockerTag }}
-          cd "$(pwd)/.github/ansible"
-
-          ./get_binaries.sh
-
-          ansible-galaxy collection install sivel.toiletwater
-          ansible-playbook -v deploy.yaml -i prod.${{ matrix.target_region }}.hosts.yaml -e @ssm_config -e CONSOLE_API_TOKEN=${{ secrets.NEON_PRODUCTION_API_KEY }} -e SENTRY_URL_PAGESERVER=${{ secrets.SENTRY_URL_PAGESERVER }} -e SENTRY_URL_SAFEKEEPER=${{ secrets.SENTRY_URL_SAFEKEEPER }}
-          rm -f neon_install.tar.gz .neon_current_version
-
-  deploy-proxy-prod-new:
-    runs-on: prod
-    container: 093970136003.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
-    if: inputs.deployProxy && inputs.disclamerAcknowledged
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        include:
-          - target_region:  us-east-2
-            target_cluster: prod-us-east-2-delta
-            deploy_link_proxy: true
-            deploy_legacy_scram_proxy: false
-          - target_region:  us-west-2
-            target_cluster: prod-us-west-2-eta
-            deploy_link_proxy: false
-            deploy_legacy_scram_proxy: true
-          - target_region: eu-central-1
-            target_cluster: prod-eu-central-1-gamma
-            deploy_link_proxy: false
-            deploy_legacy_scram_proxy: false
-          - target_region: ap-southeast-1
-            target_cluster: prod-ap-southeast-1-epsilon
-            deploy_link_proxy: false
-            deploy_legacy_scram_proxy: false
-          - target_region: us-east-1
-            target_cluster: prod-us-east-1-theta
-            deploy_link_proxy: false
-            deploy_legacy_scram_proxy: false
-    environment:
-      name: prod-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-
-      - name: Re-deploy scram proxy
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          helm upgrade neon-proxy-scram neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-scram.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-
-      - name: Re-deploy link proxy
-        if: matrix.deploy_link_proxy
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          helm upgrade neon-proxy-link neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-link.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-
-      - name: Re-deploy legacy scram proxy
-        if: matrix.deploy_legacy_scram_proxy
-        run: |
-          DOCKER_TAG=${{ inputs.dockerTag }}
-          helm upgrade neon-proxy-scram-legacy neondatabase/neon-proxy --namespace neon-proxy --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-proxy-scram-legacy.yaml --set image.tag=${DOCKER_TAG} --set settings.sentryUrl=${{ secrets.SENTRY_URL_PROXY }} --wait --timeout 15m0s
-
-  deploy-storage-broker-prod-new:
-    runs-on: prod
-    container: 093970136003.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
-    if: inputs.deployStorageBroker && inputs.disclamerAcknowledged
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        include:
-          - target_region:  us-east-2
-            target_cluster: prod-us-east-2-delta
-          - target_region:  us-west-2
-            target_cluster: prod-us-west-2-eta
-          - target_region: eu-central-1
-            target_cluster: prod-eu-central-1-gamma
-          - target_region: ap-southeast-1
-            target_cluster: prod-ap-southeast-1-epsilon
-          - target_region: us-east-1
-            target_cluster: prod-us-east-1-theta
-    environment:
-      name: prod-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-
-      - name: Deploy storage-broker
-        run:
-          helm upgrade neon-storage-broker-lb neondatabase/neon-storage-broker --namespace neon-storage-broker-lb --create-namespace --install --atomic -f .github/helm-values/${{ matrix.target_cluster }}.neon-storage-broker.yaml --set image.tag=${{ inputs.dockerTag }} --set settings.sentryUrl=${{ secrets.SENTRY_URL_BROKER }} --wait --timeout 5m0s
-
-  deploy-pg-sni-router:
-    runs-on: prod
-    container: 093970136003.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
-    if: inputs.deployPgSniRouter && inputs.disclamerAcknowledged
-    defaults:
-      run:
-        shell: bash
-    strategy:
-      matrix:
-        include:
-          - target_region:  us-east-2
-            target_cluster: prod-us-east-2-delta
-          - target_region:  us-west-2
-            target_cluster: prod-us-west-2-eta
-          - target_region: eu-central-1
-            target_cluster: prod-eu-central-1-gamma
-          - target_region: ap-southeast-1
-            target_cluster: prod-ap-southeast-1-epsilon
-          - target_region: us-east-1
-            target_cluster: prod-us-east-1-theta
-    environment:
-      name: prod-${{ matrix.target_region }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-        with:
-          submodules: true
-          fetch-depth: 0
-          ref: ${{ inputs.branch }}
-  
-      - name: Configure environment
-        run: |
-          helm repo add neondatabase https://neondatabase.github.io/helm-charts
-          aws --region ${{ matrix.target_region }} eks update-kubeconfig --name  ${{ matrix.target_cluster }}
-  
-      - name: Deploy pg-sni-router
-        run:
-          helm upgrade neon-pg-sni-router neondatabase/neon-pg-sni-router --namespace neon-pg-sni-router --create-namespace --install --debug --atomic -f .github/helm-values/${{ matrix.target_cluster }}.pg-sni-router.yaml --set image.tag=${{ inputs.dockerTag }} --set settings.sentryUrl=${{ secrets.SENTRY_URL_BROKER }} --wait --timeout 15m0s

CONTRIBUTING.md

@@ -9,6 +9,32 @@ refactoring, additional comments, and so forth. Let's try to raise the
 bar, and clean things up as we go. Try to leave code in a better shape
 than it was before.
 
+## Rust pitfalls, async or not
+
+Please be aware of at least these common problems:
+
+1. Async and Cancellation
+
+    Any future can be cancelled at every await point unless it's a top
+    level spawned future. Most common reasons to be cancelled or
+    dropped is that a future is polled or raced within `tokio::select!`
+    with a cancellation token or the HTTP client drops the connection,
+    causing a cancellation or drop on a handler future.
+    
+    Additionally please note that blocking tasks spawned with
+    `tokio::spawn_blocking` directly or indirectly will not be
+    cancelled with the spawning future.
+
+2. When using `scopeguard` or generally within `Drop::drop`, the code
+must not panic
+
+    `scopeguard` can be handy for simple operations, such as
+    decrementing a metric, but it can easily obscure that the code will
+    run on `Drop::drop`. Should a panic happen and the implemented drop
+    also panic when unwinding, the rust runtime will abort the process.
+    Within a multi-tenant system, we don't want to abort, at least for
+    accidental `unwrap` within a `Drop::drop`.
+
 ## Submitting changes
 
 1. Get at least one +1 on your PR before you push.

Cargo.lock

@@ -1032,6 +1032,7 @@ dependencies = [
  "serde",
  "serde_json",
  "serde_with",
+ "utils",
  "workspace_hack",
 ]
 
@@ -1105,6 +1106,7 @@ dependencies = [
  "anyhow",
  "clap 4.2.2",
  "comfy-table",
+ "compute_api",
  "git-version",
  "nix",
  "once_cell",

Dockerfile

@@ -2,7 +2,7 @@
 ### The image itself is mainly used as a container for the binaries and for starting e2e tests with custom parameters.
 ### By default, the binaries inside the image have some mock parameters and can start, but are not intended to be used
 ### inside this image in the real deployments.
-ARG REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
+ARG REPOSITORY=neondatabase
 ARG IMAGE=rust
 ARG TAG=pinned
 

Dockerfile.compute-node

@@ -1,5 +1,5 @@
 ARG PG_VERSION
-ARG REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
+ARG REPOSITORY=neondatabase
 ARG IMAGE=rust
 ARG TAG=pinned
 
@@ -393,6 +393,28 @@ RUN case "${PG_VERSION}" in \
     make install -j $(getconf _NPROCESSORS_ONLN) && \
     echo "trusted = true" >> /usr/local/pgsql/share/extension/pg_hint_plan.control
 
+#########################################################################################
+#
+# Layer "kq-imcx-pg-build"
+# compile kq_imcx extension
+#
+#########################################################################################
+FROM build-deps AS kq-imcx-pg-build
+COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
+
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
+RUN apt-get update && \
+    apt-get install -y git libgtk2.0-dev libpq-dev libpam-dev libxslt-dev libkrb5-dev cmake && \
+    wget https://github.com/ketteq-neon/postgres-exts/archive/e0bd1a9d9313d7120c1b9c7bb15c48c0dede4c4e.tar.gz -O kq_imcx.tar.gz && \
+    echo "dc93a97ff32d152d32737ba7e196d9687041cda15e58ab31344c2f2de8855336 kq_imcx.tar.gz" | sha256sum --check && \
+    mkdir kq_imcx-src && cd kq_imcx-src && tar xvzf ../kq_imcx.tar.gz --strip-components=1 -C . && \
+    mkdir build && \
+    cd build && \
+    cmake .. && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
+    make -j $(getconf _NPROCESSORS_ONLN) install && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/kq_imcx.control
+
 #########################################################################################
 #
 # Layer "rust extensions"
@@ -506,6 +528,7 @@ COPY --from=hll-pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=plpgsql-check-pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=timescaledb-pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg-hint-plan-pg-build /usr/local/pgsql/ /usr/local/pgsql/
+COPY --from=kq-imcx-pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY pgxn/ pgxn/
 
 RUN make -j $(getconf _NPROCESSORS_ONLN) \

Dockerfile.compute-tools

@@ -1,6 +1,6 @@
 # First transient image to build compute_tools binaries
 # NB: keep in sync with rust image version in .github/workflows/build_and_test.yml
-ARG REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
+ARG REPOSITORY=neondatabase
 ARG IMAGE=rust
 ARG TAG=pinned
 

README.md

@@ -1,3 +1,5 @@
+[![Neon](https://user-images.githubusercontent.com/13738772/236813940-dcfdcb5b-69d3-449b-a686-013febe834d4.png)](https://neon.tech)
+
 # Neon
 
 Neon is a serverless open-source alternative to AWS Aurora Postgres. It separates storage and compute and substitutes the PostgreSQL storage layer by redistributing data across a cluster of nodes.

compute_tools/src/bin/compute_ctl.rs

@@ -30,6 +30,7 @@
 //!             -b /usr/local/bin/postgres
 //! ```
 //!
+use std::collections::HashMap;
 use std::fs::File;
 use std::panic;
 use std::path::Path;
@@ -67,6 +68,54 @@ fn main() -> Result<()> {
     let spec_json = matches.get_one::<String>("spec");
     let spec_path = matches.get_one::<String>("spec-path");
 
+    // Extract OpenTelemetry context for the startup actions from the
+    // TRACEPARENT and TRACESTATE env variables, and attach it to the current
+    // tracing context.
+    //
+    // This is used to propagate the context for the 'start_compute' operation
+    // from the neon control plane. This allows linking together the wider
+    // 'start_compute' operation that creates the compute container, with the
+    // startup actions here within the container.
+    //
+    // There is no standard for passing context in env variables, but a lot of
+    // tools use TRACEPARENT/TRACESTATE, so we use that convention too. See
+    // https://github.com/open-telemetry/opentelemetry-specification/issues/740
+    //
+    // Switch to the startup context here, and exit it once the startup has
+    // completed and Postgres is up and running.
+    //
+    // If this pod is pre-created without binding it to any particular endpoint
+    // yet, this isn't the right place to enter the startup context. In that
+    // case, the control plane should pass the tracing context as part of the
+    // /configure API call.
+    //
+    // NOTE: This is supposed to only cover the *startup* actions. Once
+    // postgres is configured and up-and-running, we exit this span. Any other
+    // actions that are performed on incoming HTTP requests, for example, are
+    // performed in separate spans.
+    //
+    // XXX: If the pod is restarted, we perform the startup actions in the same
+    // context as the original startup actions, which probably doesn't make
+    // sense.
+    let mut startup_tracing_carrier: HashMap<String, String> = HashMap::new();
+    if let Ok(val) = std::env::var("TRACEPARENT") {
+        startup_tracing_carrier.insert("traceparent".to_string(), val);
+    }
+    if let Ok(val) = std::env::var("TRACESTATE") {
+        startup_tracing_carrier.insert("tracestate".to_string(), val);
+    }
+    let startup_context_guard = if !startup_tracing_carrier.is_empty() {
+        use opentelemetry::propagation::TextMapPropagator;
+        use opentelemetry::sdk::propagation::TraceContextPropagator;
+        let guard = TraceContextPropagator::new()
+            .extract(&startup_tracing_carrier)
+            .attach();
+        info!("startup tracing context attached");
+        Some(guard)
+    } else {
+        None
+    };
+
     let compute_id = matches.get_one::<String>("compute-id");
     let control_plane_uri = matches.get_one::<String>("control-plane-uri");
 
@@ -148,8 +197,6 @@ fn main() -> Result<()> {
 
     // We got all we need, update the state.
     let mut state = compute.state.lock().unwrap();
-    let pspec = state.pspec.as_ref().expect("spec must be set");
-    let startup_tracing_context = pspec.spec.startup_tracing_context.clone();
 
     // Record for how long we slept waiting for the spec.
     state.metrics.wait_for_spec_ms = Utc::now()
@@ -165,29 +212,6 @@ fn main() -> Result<()> {
     compute.state_changed.notify_all();
     drop(state);
 
-    // Extract OpenTelemetry context for the startup actions from the spec, and
-    // attach it to the current tracing context.
-    //
-    // This is used to propagate the context for the 'start_compute' operation
-    // from the neon control plane. This allows linking together the wider
-    // 'start_compute' operation that creates the compute container, with the
-    // startup actions here within the container.
-    //
-    // Switch to the startup context here, and exit it once the startup has
-    // completed and Postgres is up and running.
-    //
-    // NOTE: This is supposed to only cover the *startup* actions. Once
-    // postgres is configured and up-and-running, we exit this span. Any other
-    // actions that are performed on incoming HTTP requests, for example, are
-    // performed in separate spans.
-    let startup_context_guard = if let Some(ref carrier) = startup_tracing_context {
-        use opentelemetry::propagation::TextMapPropagator;
-        use opentelemetry::sdk::propagation::TraceContextPropagator;
-        Some(TraceContextPropagator::new().extract(carrier).attach())
-    } else {
-        None
-    };
-
     // Launch remaining service threads
     let _monitor_handle = launch_monitor(&compute).expect("cannot launch compute monitor thread");
     let _configurator_handle =

compute_tools/src/compute.rs

@@ -30,7 +30,7 @@ use utils::id::{TenantId, TimelineId};
 use utils::lsn::Lsn;
 
 use compute_api::responses::{ComputeMetrics, ComputeStatus};
-use compute_api::spec::ComputeSpec;
+use compute_api::spec::{ComputeMode, ComputeSpec};
 
 use crate::config;
 use crate::pg_helpers::*;
@@ -67,8 +67,9 @@ pub struct ComputeNode {
 pub struct ComputeState {
     pub start_time: DateTime<Utc>,
     pub status: ComputeStatus,
-    /// Timestamp of the last Postgres activity
-    pub last_active: DateTime<Utc>,
+    /// Timestamp of the last Postgres activity. It could be `None` if
+    /// compute wasn't used since start.
+    pub last_active: Option<DateTime<Utc>>,
     pub error: Option<String>,
     pub pspec: Option<ParsedSpec>,
     pub metrics: ComputeMetrics,
@@ -79,7 +80,7 @@ impl ComputeState {
         Self {
             start_time: Utc::now(),
             status: ComputeStatus::Empty,
-            last_active: Utc::now(),
+            last_active: None,
             error: None,
             pspec: None,
             metrics: ComputeMetrics::default(),
@@ -249,38 +250,10 @@ impl ComputeNode {
     /// safekeepers sync, basebackup, etc.
     #[instrument(skip(self, compute_state))]
     pub fn prepare_pgdata(&self, compute_state: &ComputeState) -> Result<()> {
-        #[derive(Clone)]
-        enum Replication {
-            Primary,
-            Static { lsn: Lsn },
-            HotStandby,
-        }
-
         let pspec = compute_state.pspec.as_ref().expect("spec must be set");
         let spec = &pspec.spec;
         let pgdata_path = Path::new(&self.pgdata);
 
-        let hot_replica = if let Some(option) = spec.cluster.settings.find_ref("hot_standby") {
-            if let Some(value) = &option.value {
-                anyhow::ensure!(option.vartype == "bool");
-                matches!(value.as_str(), "on" | "yes" | "true")
-            } else {
-                false
-            }
-        } else {
-            false
-        };
-
-        let replication = if hot_replica {
-            Replication::HotStandby
-        } else if let Some(lsn) = spec.cluster.settings.find("recovery_target_lsn") {
-            Replication::Static {
-                lsn: Lsn::from_str(&lsn)?,
-            }
-        } else {
-            Replication::Primary
-        };
-
         // Remove/create an empty pgdata directory and put configuration there.
         self.create_pgdata()?;
         config::write_postgres_conf(&pgdata_path.join("postgresql.conf"), &pspec.spec)?;
@@ -288,8 +261,8 @@ impl ComputeNode {
         // Syncing safekeepers is only safe with primary nodes: if a primary
         // is already connected it will be kicked out, so a secondary (standby)
         // cannot sync safekeepers.
-        let lsn = match &replication {
-            Replication::Primary => {
+        let lsn = match spec.mode {
+            ComputeMode::Primary => {
                 info!("starting safekeepers syncing");
                 let lsn = self
                     .sync_safekeepers(pspec.storage_auth_token.clone())
@@ -297,11 +270,11 @@ impl ComputeNode {
                 info!("safekeepers synced at LSN {}", lsn);
                 lsn
             }
-            Replication::Static { lsn } => {
+            ComputeMode::Static(lsn) => {
                 info!("Starting read-only node at static LSN {}", lsn);
-                *lsn
+                lsn
             }
-            Replication::HotStandby => {
+            ComputeMode::Replica => {
                 info!("Initializing standby from latest Pageserver LSN");
                 Lsn(0)
             }
@@ -321,9 +294,9 @@ impl ComputeNode {
         // Update pg_hba.conf received with basebackup.
         update_pg_hba(pgdata_path)?;
 
-        match &replication {
-            Replication::Primary | Replication::Static { .. } => {}
-            Replication::HotStandby => {
+        match spec.mode {
+            ComputeMode::Primary | ComputeMode::Static(..) => {}
+            ComputeMode::Replica => {
                 add_standby_signal(pgdata_path)?;
             }
         }
@@ -430,11 +403,13 @@ impl ComputeNode {
         self.pg_reload_conf(&mut client)?;
 
         // Proceed with post-startup configuration. Note, that order of operations is important.
-        handle_roles(&spec, &mut client)?;
-        handle_databases(&spec, &mut client)?;
-        handle_role_deletions(&spec, self.connstr.as_str(), &mut client)?;
-        handle_grants(&spec, self.connstr.as_str(), &mut client)?;
-        handle_extensions(&spec, &mut client)?;
+        if spec.mode == ComputeMode::Primary {
+            handle_roles(&spec, &mut client)?;
+            handle_databases(&spec, &mut client)?;
+            handle_role_deletions(&spec, self.connstr.as_str(), &mut client)?;
+            handle_grants(&spec, self.connstr.as_str(), &mut client)?;
+            handle_extensions(&spec, &mut client)?;
+        }
 
         // 'Close' connection
         drop(client);
@@ -467,7 +442,9 @@ impl ComputeNode {
 
         let pg = self.start_postgres(spec.storage_auth_token.clone())?;
 
-        self.apply_config(&compute_state)?;
+        if spec.spec.mode == ComputeMode::Primary {
+            self.apply_config(&compute_state)?;
+        }
 
         let startup_end_time = Utc::now();
         {

compute_tools/src/config.rs

@@ -6,7 +6,7 @@ use std::path::Path;
 use anyhow::Result;
 
 use crate::pg_helpers::PgOptionsSerialize;
-use compute_api::spec::ComputeSpec;
+use compute_api::spec::{ComputeMode, ComputeSpec};
 
 /// Check that `line` is inside a text file and put it there if it is not.
 /// Create file if it doesn't exist.
@@ -34,17 +34,25 @@ pub fn line_in_file(path: &Path, line: &str) -> Result<bool> {
 /// Create or completely rewrite configuration file specified by `path`
 pub fn write_postgres_conf(path: &Path, spec: &ComputeSpec) -> Result<()> {
     // File::create() destroys the file content if it exists.
-    let mut postgres_conf = File::create(path)?;
+    let mut file = File::create(path)?;
 
-    write_auto_managed_block(&mut postgres_conf, &spec.cluster.settings.as_pg_settings())?;
-
-    Ok(())
-}
-
-// Write Postgres config block wrapped with generated comment section
-fn write_auto_managed_block(file: &mut File, buf: &str) -> Result<()> {
     writeln!(file, "# Managed by compute_ctl: begin")?;
-    writeln!(file, "{}", buf)?;
+
+    write!(file, "{}", &spec.cluster.settings.as_pg_settings())?;
+
+    match spec.mode {
+        ComputeMode::Primary => {}
+        ComputeMode::Static(lsn) => {
+            // hot_standby is 'on' by default, but let's be explicit
+            writeln!(file, "hot_standby=on")?;
+            writeln!(file, "recovery_target_lsn='{lsn}'")?;
+        }
+        ComputeMode::Replica => {
+            // hot_standby is 'on' by default, but let's be explicit
+            writeln!(file, "hot_standby=on")?;
+        }
+    }
+
     writeln!(file, "# Managed by compute_ctl: end")?;
 
     Ok(())

compute_tools/src/http/openapi_spec.yaml

@@ -181,8 +181,8 @@ components:
     ComputeState:
       type: object
       required:
+        - start_time
         - status
-        - last_active
       properties:
         start_time:
           type: string
@@ -195,11 +195,13 @@ components:
           $ref: '#/components/schemas/ComputeStatus'
         last_active:
           type: string
-          description: The last detected compute activity timestamp in UTC and RFC3339 format.
+          description: |
+            The last detected compute activity timestamp in UTC and RFC3339 format.
+            It could be empty if compute was never used by user since start.
           example: "2022-10-12T07:20:50.52Z"
         error:
           type: string
-          description: Text of the error during compute startup, if any.
+          description: Text of the error during compute startup or reconfiguration, if any.
           example: ""
         tenant:
           type: string
@@ -222,9 +224,12 @@ components:
     ComputeStatus:
       type: string
       enum:
+        - empty
         - init
         - failed
         - running
+        - configuration_pending
+        - configuration
       example: running
 
     #

compute_tools/src/monitor.rs

@@ -74,7 +74,7 @@ fn watch_compute_activity(compute: &ComputeNode) {
                             // Found non-idle backend, so the last activity is NOW.
                             // Save it and exit the for loop. Also clear the idle backend
                             // `state_change` timestamps array as it doesn't matter now.
-                            last_active = Utc::now();
+                            last_active = Some(Utc::now());
                             idle_backs.clear();
                             break;
                         }
@@ -82,15 +82,16 @@ fn watch_compute_activity(compute: &ComputeNode) {
 
                     // Get idle backend `state_change` with the max timestamp.
                     if let Some(last) = idle_backs.iter().max() {
-                        last_active = *last;
+                        last_active = Some(*last);
                     }
                 }
 
                 // Update the last activity in the shared state if we got a more recent one.
                 let mut state = compute.state.lock().unwrap();
+                // NB: `Some(<DateTime>)` is always greater than `None`.
                 if last_active > state.last_active {
                     state.last_active = last_active;
-                    debug!("set the last compute activity time to: {}", last_active);
+                    debug!("set the last compute activity time to: {:?}", last_active);
                 }
             }
             Err(e) => {

control_plane/Cargo.toml

@@ -30,4 +30,5 @@ postgres_connection.workspace = true
 storage_broker.workspace = true
 utils.workspace = true
 
+compute_api.workspace = true
 workspace_hack.workspace = true

control_plane/src/bin/neon_local.rs

@@ -7,8 +7,8 @@
 //!
 use anyhow::{anyhow, bail, Context, Result};
 use clap::{value_parser, Arg, ArgAction, ArgMatches, Command};
+use compute_api::spec::ComputeMode;
 use control_plane::endpoint::ComputeControlPlane;
-use control_plane::endpoint::ComputeMode;
 use control_plane::local_env::LocalEnv;
 use control_plane::pageserver::PageServerNode;
 use control_plane::safekeeper::SafekeeperNode;

control_plane/src/endpoint.rs

@@ -22,6 +22,8 @@ use crate::local_env::LocalEnv;
 use crate::pageserver::PageServerNode;
 use crate::postgresql_conf::PostgresConf;
 
+use compute_api::spec::ComputeMode;
+
 // contents of a endpoint.json file
 #[serde_as]
 #[derive(Serialize, Deserialize, PartialEq, Eq, Clone, Debug)]
@@ -122,26 +124,12 @@ impl ComputeControlPlane {
 
 ///////////////////////////////////////////////////////////////////////////////
 
-#[serde_as]
-#[derive(Serialize, Deserialize, Debug, Clone, Copy, Eq, PartialEq)]
-pub enum ComputeMode {
-    // Regular read-write node
-    Primary,
-    // if recovery_target_lsn is provided, and we want to pin the node to a specific LSN
-    Static(#[serde_as(as = "DisplayFromStr")] Lsn),
-    // Hot standby; read-only replica.
-    // Future versions may want to distinguish between replicas with hot standby
-    // feedback and other kinds of replication configurations.
-    Replica,
-}
-
 #[derive(Debug)]
 pub struct Endpoint {
     /// used as the directory name
     name: String,
     pub tenant_id: TenantId,
     pub timeline_id: TimelineId,
-    // Some(lsn) if this is a read-only endpoint anchored at 'lsn'. None for the primary.
     pub mode: ComputeMode,
 
     // port and address of the Postgres server

libs/compute_api/Cargo.toml

@@ -11,4 +11,5 @@ serde.workspace = true
 serde_with.workspace = true
 serde_json.workspace = true
 
+utils = { path = "../utils" }
 workspace_hack.workspace = true

libs/compute_api/src/responses.rs

@@ -19,7 +19,7 @@ pub struct ComputeStatusResponse {
     pub timeline: Option<String>,
     pub status: ComputeStatus,
     #[serde(serialize_with = "rfc3339_serialize")]
-    pub last_active: DateTime<Utc>,
+    pub last_active: Option<DateTime<Utc>>,
     pub error: Option<String>,
 }
 
@@ -29,7 +29,7 @@ pub struct ComputeState {
     pub status: ComputeStatus,
     /// Timestamp of the last Postgres activity
     #[serde(serialize_with = "rfc3339_serialize")]
-    pub last_active: DateTime<Utc>,
+    pub last_active: Option<DateTime<Utc>>,
     pub error: Option<String>,
 }
 
@@ -54,11 +54,15 @@ pub enum ComputeStatus {
     Failed,
 }
 
-fn rfc3339_serialize<S>(x: &DateTime<Utc>, s: S) -> Result<S::Ok, S::Error>
+fn rfc3339_serialize<S>(x: &Option<DateTime<Utc>>, s: S) -> Result<S::Ok, S::Error>
 where
     S: Serializer,
 {
-    x.to_rfc3339().serialize(s)
+    if let Some(x) = x {
+        x.to_rfc3339().serialize(s)
+    } else {
+        s.serialize_none()
+    }
 }
 
 /// Response of the /metrics.json API

libs/compute_api/src/spec.rs

@@ -3,8 +3,9 @@
 //! The spec.json file is used to pass information to 'compute_ctl'. It contains
 //! all the information needed to start up the right version of PostgreSQL,
 //! and connect it to the storage nodes.
-use serde::Deserialize;
-use std::collections::HashMap;
+use serde::{Deserialize, Serialize};
+use serde_with::{serde_as, DisplayFromStr};
+use utils::lsn::Lsn;
 
 /// String type alias representing Postgres identifier and
 /// intended to be used for DB / role names.
@@ -12,6 +13,7 @@ pub type PgIdent = String;
 
 /// Cluster spec or configuration represented as an optional number of
 /// delta operations + final cluster state description.
+#[serde_as]
 #[derive(Clone, Debug, Default, Deserialize)]
 pub struct ComputeSpec {
     pub format_version: f32,
@@ -24,9 +26,25 @@ pub struct ComputeSpec {
     pub cluster: Cluster,
     pub delta_operations: Option<Vec<DeltaOp>>,
 
-    pub storage_auth_token: Option<String>,
+    #[serde(default)]
+    pub mode: ComputeMode,
 
-    pub startup_tracing_context: Option<HashMap<String, String>>,
+    pub storage_auth_token: Option<String>,
+}
+
+#[serde_as]
+#[derive(Clone, Copy, Debug, Default, Eq, PartialEq, Deserialize, Serialize)]
+pub enum ComputeMode {
+    /// A read-write node
+    #[default]
+    Primary,
+    /// A read-only node, pinned at a particular LSN
+    Static(#[serde_as(as = "DisplayFromStr")] Lsn),
+    /// A read-only node that follows the tip of the branch in hot standby mode
+    ///
+    /// Future versions may want to distinguish between replicas with hot standby
+    /// feedback and other kinds of replication configurations.
+    Replica,
 }
 
 #[derive(Clone, Debug, Default, Deserialize)]

libs/pq_proto/src/lib.rs

@@ -613,7 +613,7 @@ pub struct XLogDataBody<'a> {
 
 #[derive(Debug)]
 pub struct WalSndKeepAlive {
-    pub sent_ptr: u64,
+    pub wal_end: u64, // current end of WAL on the server
     pub timestamp: i64,
     pub request_reply: bool,
 }
@@ -924,7 +924,7 @@ impl<'a> BeMessage<'a> {
                 buf.put_u8(b'd');
                 write_body(buf, |buf| {
                     buf.put_u8(b'k');
-                    buf.put_u64(req.sent_ptr);
+                    buf.put_u64(req.wal_end);
                     buf.put_i64(req.timestamp);
                     buf.put_u8(u8::from(req.request_reply));
                 });

libs/remote_storage/src/local_fs.rs

@@ -135,7 +135,7 @@ impl RemoteStorage for LocalFs {
         // T1: rename(temp, dst) -> succeeds
         // T2: rename(temp, dst) -> fails, temp no longet exists
         // This can be solved by supplying unique temp suffix every time, but this situation
-        // is not normall in the first place, the error can help (and helped at least once)
+        // is not normal in the first place, the error can help (and helped at least once)
         // to discover bugs in upper level synchronization.
         let temp_file_path =
             path_with_suffix_extension(&target_file_path, LOCAL_FS_TEMP_FILE_SUFFIX);

libs/utils/src/lib.rs

@@ -58,6 +58,8 @@ pub mod pageserver_feedback;
 
 pub mod tracing_span_assert;
 
+pub mod rate_limit;
+
 /// use with fail::cfg("$name", "return(2000)")
 #[macro_export]
 macro_rules! failpoint_sleep_millis_async {

libs/utils/src/rate_limit.rs

@@ -0,0 +1,66 @@
+//! A helper to rate limit operations.
+
+use std::time::{Duration, Instant};
+
+pub struct RateLimit {
+    last: Option<Instant>,
+    interval: Duration,
+}
+
+impl RateLimit {
+    pub fn new(interval: Duration) -> Self {
+        Self {
+            last: None,
+            interval,
+        }
+    }
+
+    /// Call `f` if the rate limit allows.
+    /// Don't call it otherwise.
+    pub fn call<F: FnOnce()>(&mut self, f: F) {
+        let now = Instant::now();
+        match self.last {
+            Some(last) if now - last <= self.interval => {
+                // ratelimit
+            }
+            _ => {
+                self.last = Some(now);
+                f();
+            }
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::sync::atomic::AtomicUsize;
+
+    #[test]
+    fn basics() {
+        use super::RateLimit;
+        use std::sync::atomic::Ordering::Relaxed;
+        use std::time::Duration;
+
+        let called = AtomicUsize::new(0);
+        let mut f = RateLimit::new(Duration::from_millis(100));
+
+        let cl = || {
+            called.fetch_add(1, Relaxed);
+        };
+
+        f.call(cl);
+        assert_eq!(called.load(Relaxed), 1);
+        f.call(cl);
+        assert_eq!(called.load(Relaxed), 1);
+        f.call(cl);
+        assert_eq!(called.load(Relaxed), 1);
+        std::thread::sleep(Duration::from_millis(100));
+        f.call(cl);
+        assert_eq!(called.load(Relaxed), 2);
+        f.call(cl);
+        assert_eq!(called.load(Relaxed), 2);
+        std::thread::sleep(Duration::from_millis(100));
+        f.call(cl);
+        assert_eq!(called.load(Relaxed), 3);
+    }
+}

pageserver/src/http/openapi_spec.yml

@@ -340,7 +340,29 @@ paths:
           format: hex
 
     post:
-      description: Schedules attach operation to happen in the background for given tenant
+      description: |
+        Schedules attach operation to happen in the background for the given tenant.
+        As soon as the caller sends this request, it must assume the pageserver
+        starts writing to the tenant's S3 state unless it receives one of the
+        distinguished errors below that state otherwise.
+
+        The method to identify whether a request has arrived at the pageserver, and
+        whether it has succeeded, is to poll for the tenant status to reach "Active"
+        or "Broken" state. These values are currently not explicitly documented in
+        the API spec.
+        Polling for `has_in_progress_downloads == false` is INCORRECT because that
+        value can turn `false` during shutdown while the Attach operation is still
+        unfinished.
+
+        There is no way to cancel an in-flight request.
+
+        If a client receives a not-distinguished response, e.g., a network timeout,
+        it MUST retry the /attach request and poll again for tenant status.
+
+        In any case, it must
+        * NOT ASSUME that the /attach request has been lost in the network,
+        * NOT ASSUME that the request has been lost based on a subsequent
+          tenant status request returning 404. (The request may still be in flight!)
       responses:
         "202":
           description: Tenant attaching scheduled

pageserver/src/http/routes.rs

@@ -143,7 +143,6 @@ impl From<crate::tenant::DeleteTimelineError> for ApiError {
             HasChildren => ApiError::BadRequest(anyhow::anyhow!(
                 "Cannot delete timeline which has child timelines"
             )),
-            StopUploadQueue(e) => ApiError::InternalServerError(e.into()),
             Other(e) => ApiError::InternalServerError(e),
         }
     }
@@ -692,16 +691,6 @@ pub fn html_response(status: StatusCode, data: String) -> Result<Response<Body>,
     Ok(response)
 }
 
-// Helper function to standardize the error messages we produce on bad durations
-//
-// Intended to be used with anyhow's `with_context`, e.g.:
-//
-//   let value = result.with_context(bad_duration("name", &value))?;
-//
-fn bad_duration<'a>(field_name: &'static str, value: &'a str) -> impl 'a + Fn() -> String {
-    move || format!("Cannot parse `{field_name}` duration {value:?}")
-}
-
 async fn tenant_create_handler(mut request: Request<Body>) -> Result<Response<Body>, ApiError> {
     check_permission(&request, None)?;
 
@@ -709,91 +698,7 @@ async fn tenant_create_handler(mut request: Request<Body>) -> Result<Response<Bo
 
     let request_data: TenantCreateRequest = json_request(&mut request).await?;
 
-    let mut tenant_conf = TenantConfOpt::default();
-    if let Some(gc_period) = request_data.gc_period {
-        tenant_conf.gc_period = Some(
-            humantime::parse_duration(&gc_period)
-                .with_context(bad_duration("gc_period", &gc_period))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    tenant_conf.gc_horizon = request_data.gc_horizon;
-    tenant_conf.image_creation_threshold = request_data.image_creation_threshold;
-
-    if let Some(pitr_interval) = request_data.pitr_interval {
-        tenant_conf.pitr_interval = Some(
-            humantime::parse_duration(&pitr_interval)
-                .with_context(bad_duration("pitr_interval", &pitr_interval))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-
-    if let Some(walreceiver_connect_timeout) = request_data.walreceiver_connect_timeout {
-        tenant_conf.walreceiver_connect_timeout = Some(
-            humantime::parse_duration(&walreceiver_connect_timeout)
-                .with_context(bad_duration(
-                    "walreceiver_connect_timeout",
-                    &walreceiver_connect_timeout,
-                ))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    if let Some(lagging_wal_timeout) = request_data.lagging_wal_timeout {
-        tenant_conf.lagging_wal_timeout = Some(
-            humantime::parse_duration(&lagging_wal_timeout)
-                .with_context(bad_duration("lagging_wal_timeout", &lagging_wal_timeout))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    if let Some(max_lsn_wal_lag) = request_data.max_lsn_wal_lag {
-        tenant_conf.max_lsn_wal_lag = Some(max_lsn_wal_lag);
-    }
-    if let Some(trace_read_requests) = request_data.trace_read_requests {
-        tenant_conf.trace_read_requests = Some(trace_read_requests);
-    }
-
-    tenant_conf.checkpoint_distance = request_data.checkpoint_distance;
-    if let Some(checkpoint_timeout) = request_data.checkpoint_timeout {
-        tenant_conf.checkpoint_timeout = Some(
-            humantime::parse_duration(&checkpoint_timeout)
-                .with_context(bad_duration("checkpoint_timeout", &checkpoint_timeout))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-
-    tenant_conf.compaction_target_size = request_data.compaction_target_size;
-    tenant_conf.compaction_threshold = request_data.compaction_threshold;
-
-    if let Some(compaction_period) = request_data.compaction_period {
-        tenant_conf.compaction_period = Some(
-            humantime::parse_duration(&compaction_period)
-                .with_context(bad_duration("compaction_period", &compaction_period))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-
-    if let Some(eviction_policy) = request_data.eviction_policy {
-        tenant_conf.eviction_policy = Some(
-            serde_json::from_value(eviction_policy)
-                .context("parse field `eviction_policy`")
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-
-    tenant_conf.min_resident_size_override = request_data.min_resident_size_override;
-
-    if let Some(evictions_low_residence_duration_metric_threshold) =
-        request_data.evictions_low_residence_duration_metric_threshold
-    {
-        tenant_conf.evictions_low_residence_duration_metric_threshold = Some(
-            humantime::parse_duration(&evictions_low_residence_duration_metric_threshold)
-                .with_context(bad_duration(
-                    "evictions_low_residence_duration_metric_threshold",
-                    &evictions_low_residence_duration_metric_threshold,
-                ))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
+    let tenant_conf = TenantConfOpt::try_from(&request_data).map_err(ApiError::BadRequest)?;
 
     let target_tenant_id = request_data
         .new_tenant_id
@@ -861,85 +766,7 @@ async fn update_tenant_config_handler(
     let tenant_id = request_data.tenant_id;
     check_permission(&request, Some(tenant_id))?;
 
-    let mut tenant_conf = TenantConfOpt::default();
-    if let Some(gc_period) = request_data.gc_period {
-        tenant_conf.gc_period = Some(
-            humantime::parse_duration(&gc_period)
-                .with_context(bad_duration("gc_period", &gc_period))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    tenant_conf.gc_horizon = request_data.gc_horizon;
-    tenant_conf.image_creation_threshold = request_data.image_creation_threshold;
-
-    if let Some(pitr_interval) = request_data.pitr_interval {
-        tenant_conf.pitr_interval = Some(
-            humantime::parse_duration(&pitr_interval)
-                .with_context(bad_duration("pitr_interval", &pitr_interval))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    if let Some(walreceiver_connect_timeout) = request_data.walreceiver_connect_timeout {
-        tenant_conf.walreceiver_connect_timeout = Some(
-            humantime::parse_duration(&walreceiver_connect_timeout)
-                .with_context(bad_duration(
-                    "walreceiver_connect_timeout",
-                    &walreceiver_connect_timeout,
-                ))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    if let Some(lagging_wal_timeout) = request_data.lagging_wal_timeout {
-        tenant_conf.lagging_wal_timeout = Some(
-            humantime::parse_duration(&lagging_wal_timeout)
-                .with_context(bad_duration("lagging_wal_timeout", &lagging_wal_timeout))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    tenant_conf.max_lsn_wal_lag = request_data.max_lsn_wal_lag;
-    tenant_conf.trace_read_requests = request_data.trace_read_requests;
-
-    tenant_conf.checkpoint_distance = request_data.checkpoint_distance;
-    if let Some(checkpoint_timeout) = request_data.checkpoint_timeout {
-        tenant_conf.checkpoint_timeout = Some(
-            humantime::parse_duration(&checkpoint_timeout)
-                .with_context(bad_duration("checkpoint_timeout", &checkpoint_timeout))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-    tenant_conf.compaction_target_size = request_data.compaction_target_size;
-    tenant_conf.compaction_threshold = request_data.compaction_threshold;
-
-    if let Some(compaction_period) = request_data.compaction_period {
-        tenant_conf.compaction_period = Some(
-            humantime::parse_duration(&compaction_period)
-                .with_context(bad_duration("compaction_period", &compaction_period))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-
-    if let Some(eviction_policy) = request_data.eviction_policy {
-        tenant_conf.eviction_policy = Some(
-            serde_json::from_value(eviction_policy)
-                .context("parse field `eviction_policy`")
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
-
-    tenant_conf.min_resident_size_override = request_data.min_resident_size_override;
-
-    if let Some(evictions_low_residence_duration_metric_threshold) =
-        request_data.evictions_low_residence_duration_metric_threshold
-    {
-        tenant_conf.evictions_low_residence_duration_metric_threshold = Some(
-            humantime::parse_duration(&evictions_low_residence_duration_metric_threshold)
-                .with_context(bad_duration(
-                    "evictions_low_residence_duration_metric_threshold",
-                    &evictions_low_residence_duration_metric_threshold,
-                ))
-                .map_err(ApiError::BadRequest)?,
-        );
-    }
+    let tenant_conf = TenantConfOpt::try_from(&request_data).map_err(ApiError::BadRequest)?;
 
     let state = get_state(&request);
     mgr::set_new_tenant_config(state.conf, tenant_conf, tenant_id)

pageserver/src/tenant.rs

@@ -59,6 +59,7 @@ use crate::tenant::config::TenantConfOpt;
 use crate::tenant::metadata::load_metadata;
 use crate::tenant::remote_timeline_client::index::IndexPart;
 use crate::tenant::remote_timeline_client::MaybeDeletedIndexPart;
+use crate::tenant::remote_timeline_client::PersistIndexPartWithDeletedFlagError;
 use crate::tenant::storage_layer::DeltaLayer;
 use crate::tenant::storage_layer::ImageLayer;
 use crate::tenant::storage_layer::Layer;
@@ -272,7 +273,10 @@ impl UninitializedTimeline<'_> {
             .await
             .context("Failed to flush after basebackup import")?;
 
-        self.initialize(ctx)
+        // Initialize without loading the layer map. We started with an empty layer map, and already
+        // updated it for the layers that we created during the import.
+        let mut timelines = self.owning_tenant.timelines.lock().unwrap();
+        self.initialize_with_lock(ctx, &mut timelines, false, true)
     }
 
     fn raw_timeline(&self) -> anyhow::Result<&Arc<Timeline>> {
@@ -448,8 +452,6 @@ pub enum DeleteTimelineError {
     NotFound,
     #[error("HasChildren")]
     HasChildren,
-    #[error("stop upload queue: {0:#}")]
-    StopUploadQueue(#[from] remote_timeline_client::StopError),
     #[error(transparent)]
     Other(#[from] anyhow::Error),
 }
@@ -1051,6 +1053,7 @@ impl Tenant {
     /// Subroutine of `load_tenant`, to load an individual timeline
     ///
     /// NB: The parent is assumed to be already loaded!
+    #[instrument(skip_all, fields(timeline_id))]
     async fn load_local_timeline(
         &self,
         timeline_id: TimelineId,
@@ -1362,6 +1365,8 @@ impl Tenant {
         timeline_id: TimelineId,
         _ctx: &RequestContext,
     ) -> Result<(), DeleteTimelineError> {
+        timeline::debug_assert_current_span_has_tenant_and_timeline_id();
+
         // Transition the timeline into TimelineState::Stopping.
         // This should prevent new operations from starting.
         let timeline = {
@@ -1405,32 +1410,39 @@ impl Tenant {
         // Prevent new uploads from starting.
         if let Some(remote_client) = timeline.remote_client.as_ref() {
             let res = remote_client.stop();
-            match &res {
+            match res {
                 Ok(()) => {}
                 Err(e) => match e {
-                    remote_timeline_client::StopError::QueueBroken => {
-                        // This happens if there's a panic inside above stop() call,
-                        // and we call stop() again after that.
-                        // The calling again can happen because we won't poison any
-                        // mutexes on the unwind path at the first panicking call.
-                    }
                     remote_timeline_client::StopError::QueueUninitialized => {
-                        // This could happen if the timeline is Broken, e.g., because it failed to fetch IndexPart when it was loaded.
+                        // This case shouldn't happen currently because the
+                        // load and attach code bails out if _any_ of the timeline fails to fetch its IndexPart.
+                        // That is, before we declare the Tenant as Active.
+                        // But we only allow calls to delete_timeline on Active tenants.
+                        return Err(DeleteTimelineError::Other(anyhow::anyhow!("upload queue is uninitialized, likely the timeline was in Broken state prior to this call because it failed to fetch IndexPart during load or attach, check the logs")));
                     }
                 },
             }
-            res?;
         }
 
         // Stop & wait for the remaining timeline tasks, including upload tasks.
+        // NB: This and other delete_timeline calls do not run as a task_mgr task,
+        //     so, they are not affected by this shutdown_tasks() call.
         info!("waiting for timeline tasks to shutdown");
         task_mgr::shutdown_tasks(None, Some(self.tenant_id), Some(timeline_id)).await;
 
-        // Mark timeline as deleted in S3 so we wont pick it up next time
+        // Mark timeline as deleted in S3 so we won't pick it up next time
         // during attach or pageserver restart.
         // See comment in persist_index_part_with_deleted_flag.
         if let Some(remote_client) = timeline.remote_client.as_ref() {
-            remote_client.persist_index_part_with_deleted_flag().await?;
+            match remote_client.persist_index_part_with_deleted_flag().await {
+                // If we (now, or already) marked it successfully as deleted, we can proceed
+                Ok(()) | Err(PersistIndexPartWithDeletedFlagError::AlreadyDeleted(_)) => (),
+                // Bail out otherwise
+                Err(e @ PersistIndexPartWithDeletedFlagError::AlreadyInProgress(_))
+                | Err(e @ PersistIndexPartWithDeletedFlagError::Other(_)) => {
+                    return Err(DeleteTimelineError::Other(anyhow::anyhow!(e)));
+                }
+            }
         }
 
         {
@@ -1467,17 +1479,43 @@ impl Tenant {
             // For configurations without remote storage, we tolerate that we're not crash-safe here.
             // The timeline may come up Active but with missing layer files, in such setups.
             // See https://github.com/neondatabase/neon/pull/3919#issuecomment-1531726720
-            std::fs::remove_dir_all(&local_timeline_directory).with_context(|| {
-                format!(
-                    "Failed to remove local timeline directory '{}'",
-                    local_timeline_directory.display()
-                )
-            })?;
+            match std::fs::remove_dir_all(&local_timeline_directory) {
+                Err(e) if e.kind() == std::io::ErrorKind::NotFound => {
+                    // This can happen if we're called a second time, e.g.,
+                    // because of a previous failure/cancellation at/after
+                    // failpoint timeline-delete-after-rm.
+                    //
+                    // It can also happen if we race with tenant detach, because,
+                    // it doesn't grab the layer_removal_cs lock.
+                    //
+                    // For now, log and continue.
+                    // warn! level is technically not appropriate for the
+                    // first case because we should expect retries to happen.
+                    // But the error is so rare, it seems better to get attention if it happens.
+                    let tenant_state = self.current_state();
+                    warn!(
+                        timeline_dir=?local_timeline_directory,
+                        ?tenant_state,
+                        "timeline directory not found, proceeding anyway"
+                    );
+                    // continue with the rest of the deletion
+                }
+                res => res.with_context(|| {
+                    format!(
+                        "Failed to remove local timeline directory '{}'",
+                        local_timeline_directory.display()
+                    )
+                })?,
+            }
 
             info!("finished deleting layer files, releasing layer_removal_cs.lock()");
             drop(layer_removal_guard);
         }
 
+        fail::fail_point!("timeline-delete-after-rm", |_| {
+            Err(anyhow::anyhow!("failpoint: timeline-delete-after-rm"))?
+        });
+
         // Remove the timeline from the map.
         let mut timelines = self.timelines.lock().unwrap();
         let children_exist = timelines
@@ -2420,6 +2458,8 @@ impl Tenant {
                 )
             })?;
 
+        // Initialize the timeline without loading the layer map, because we already updated the layer
+        // map above, when we imported the datadir.
         let timeline = {
             let mut timelines = self.timelines.lock().unwrap();
             raw_timeline.initialize_with_lock(ctx, &mut timelines, false, true)?

pageserver/src/tenant/config.rs

@@ -8,6 +8,8 @@
 //! We cannot use global or default config instead, because wrong settings
 //! may lead to a data loss.
 //!
+use anyhow::Context;
+use pageserver_api::models::{TenantConfigRequest, TenantCreateRequest};
 use serde::{Deserialize, Serialize};
 use std::num::NonZeroU64;
 use std::time::Duration;
@@ -280,6 +282,179 @@ impl Default for TenantConf {
     }
 }
 
+// Helper function to standardize the error messages we produce on bad durations
+//
+// Intended to be used with anyhow's `with_context`, e.g.:
+//
+//   let value = result.with_context(bad_duration("name", &value))?;
+//
+fn bad_duration<'a>(field_name: &'static str, value: &'a str) -> impl 'a + Fn() -> String {
+    move || format!("Cannot parse `{field_name}` duration {value:?}")
+}
+
+impl TryFrom<&'_ TenantCreateRequest> for TenantConfOpt {
+    type Error = anyhow::Error;
+
+    fn try_from(request_data: &TenantCreateRequest) -> Result<Self, Self::Error> {
+        let mut tenant_conf = TenantConfOpt::default();
+
+        if let Some(gc_period) = &request_data.gc_period {
+            tenant_conf.gc_period = Some(
+                humantime::parse_duration(gc_period)
+                    .with_context(bad_duration("gc_period", gc_period))?,
+            );
+        }
+        tenant_conf.gc_horizon = request_data.gc_horizon;
+        tenant_conf.image_creation_threshold = request_data.image_creation_threshold;
+
+        if let Some(pitr_interval) = &request_data.pitr_interval {
+            tenant_conf.pitr_interval = Some(
+                humantime::parse_duration(pitr_interval)
+                    .with_context(bad_duration("pitr_interval", pitr_interval))?,
+            );
+        }
+
+        if let Some(walreceiver_connect_timeout) = &request_data.walreceiver_connect_timeout {
+            tenant_conf.walreceiver_connect_timeout = Some(
+                humantime::parse_duration(walreceiver_connect_timeout).with_context(
+                    bad_duration("walreceiver_connect_timeout", walreceiver_connect_timeout),
+                )?,
+            );
+        }
+        if let Some(lagging_wal_timeout) = &request_data.lagging_wal_timeout {
+            tenant_conf.lagging_wal_timeout = Some(
+                humantime::parse_duration(lagging_wal_timeout)
+                    .with_context(bad_duration("lagging_wal_timeout", lagging_wal_timeout))?,
+            );
+        }
+        if let Some(max_lsn_wal_lag) = request_data.max_lsn_wal_lag {
+            tenant_conf.max_lsn_wal_lag = Some(max_lsn_wal_lag);
+        }
+        if let Some(trace_read_requests) = request_data.trace_read_requests {
+            tenant_conf.trace_read_requests = Some(trace_read_requests);
+        }
+
+        tenant_conf.checkpoint_distance = request_data.checkpoint_distance;
+        if let Some(checkpoint_timeout) = &request_data.checkpoint_timeout {
+            tenant_conf.checkpoint_timeout = Some(
+                humantime::parse_duration(checkpoint_timeout)
+                    .with_context(bad_duration("checkpoint_timeout", checkpoint_timeout))?,
+            );
+        }
+
+        tenant_conf.compaction_target_size = request_data.compaction_target_size;
+        tenant_conf.compaction_threshold = request_data.compaction_threshold;
+
+        if let Some(compaction_period) = &request_data.compaction_period {
+            tenant_conf.compaction_period = Some(
+                humantime::parse_duration(compaction_period)
+                    .with_context(bad_duration("compaction_period", compaction_period))?,
+            );
+        }
+
+        if let Some(eviction_policy) = &request_data.eviction_policy {
+            tenant_conf.eviction_policy = Some(
+                serde::Deserialize::deserialize(eviction_policy)
+                    .context("parse field `eviction_policy`")?,
+            );
+        }
+
+        tenant_conf.min_resident_size_override = request_data.min_resident_size_override;
+
+        if let Some(evictions_low_residence_duration_metric_threshold) =
+            &request_data.evictions_low_residence_duration_metric_threshold
+        {
+            tenant_conf.evictions_low_residence_duration_metric_threshold = Some(
+                humantime::parse_duration(evictions_low_residence_duration_metric_threshold)
+                    .with_context(bad_duration(
+                        "evictions_low_residence_duration_metric_threshold",
+                        evictions_low_residence_duration_metric_threshold,
+                    ))?,
+            );
+        }
+
+        Ok(tenant_conf)
+    }
+}
+
+impl TryFrom<&'_ TenantConfigRequest> for TenantConfOpt {
+    type Error = anyhow::Error;
+
+    fn try_from(request_data: &TenantConfigRequest) -> Result<Self, Self::Error> {
+        let mut tenant_conf = TenantConfOpt::default();
+        if let Some(gc_period) = &request_data.gc_period {
+            tenant_conf.gc_period = Some(
+                humantime::parse_duration(gc_period)
+                    .with_context(bad_duration("gc_period", gc_period))?,
+            );
+        }
+        tenant_conf.gc_horizon = request_data.gc_horizon;
+        tenant_conf.image_creation_threshold = request_data.image_creation_threshold;
+
+        if let Some(pitr_interval) = &request_data.pitr_interval {
+            tenant_conf.pitr_interval = Some(
+                humantime::parse_duration(pitr_interval)
+                    .with_context(bad_duration("pitr_interval", pitr_interval))?,
+            );
+        }
+        if let Some(walreceiver_connect_timeout) = &request_data.walreceiver_connect_timeout {
+            tenant_conf.walreceiver_connect_timeout = Some(
+                humantime::parse_duration(walreceiver_connect_timeout).with_context(
+                    bad_duration("walreceiver_connect_timeout", walreceiver_connect_timeout),
+                )?,
+            );
+        }
+        if let Some(lagging_wal_timeout) = &request_data.lagging_wal_timeout {
+            tenant_conf.lagging_wal_timeout = Some(
+                humantime::parse_duration(lagging_wal_timeout)
+                    .with_context(bad_duration("lagging_wal_timeout", lagging_wal_timeout))?,
+            );
+        }
+        tenant_conf.max_lsn_wal_lag = request_data.max_lsn_wal_lag;
+        tenant_conf.trace_read_requests = request_data.trace_read_requests;
+
+        tenant_conf.checkpoint_distance = request_data.checkpoint_distance;
+        if let Some(checkpoint_timeout) = &request_data.checkpoint_timeout {
+            tenant_conf.checkpoint_timeout = Some(
+                humantime::parse_duration(checkpoint_timeout)
+                    .with_context(bad_duration("checkpoint_timeout", checkpoint_timeout))?,
+            );
+        }
+        tenant_conf.compaction_target_size = request_data.compaction_target_size;
+        tenant_conf.compaction_threshold = request_data.compaction_threshold;
+
+        if let Some(compaction_period) = &request_data.compaction_period {
+            tenant_conf.compaction_period = Some(
+                humantime::parse_duration(compaction_period)
+                    .with_context(bad_duration("compaction_period", compaction_period))?,
+            );
+        }
+
+        if let Some(eviction_policy) = &request_data.eviction_policy {
+            tenant_conf.eviction_policy = Some(
+                serde::Deserialize::deserialize(eviction_policy)
+                    .context("parse field `eviction_policy`")?,
+            );
+        }
+
+        tenant_conf.min_resident_size_override = request_data.min_resident_size_override;
+
+        if let Some(evictions_low_residence_duration_metric_threshold) =
+            &request_data.evictions_low_residence_duration_metric_threshold
+        {
+            tenant_conf.evictions_low_residence_duration_metric_threshold = Some(
+                humantime::parse_duration(evictions_low_residence_duration_metric_threshold)
+                    .with_context(bad_duration(
+                        "evictions_low_residence_duration_metric_threshold",
+                        evictions_low_residence_duration_metric_threshold,
+                    ))?,
+            );
+        }
+
+        Ok(tenant_conf)
+    }
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;

pageserver/src/tenant/remote_timeline_client.rs

@@ -205,7 +205,7 @@ pub mod index;
 mod upload;
 
 use anyhow::Context;
-use chrono::Utc;
+use chrono::{NaiveDateTime, Utc};
 // re-export these
 pub use download::{is_temp_download_file, list_remote_timelines};
 use scopeguard::ScopeGuard;
@@ -243,6 +243,7 @@ use utils::id::{TenantId, TimelineId};
 use self::index::IndexPart;
 
 use super::storage_layer::LayerFileName;
+use super::upload_queue::SetDeletedFlagProgress;
 
 // Occasional network issues and such can cause remote operations to fail, and
 // that's expected. If a download fails, we log it at info-level, and retry.
@@ -268,8 +269,16 @@ pub enum StopError {
     /// See [`RemoteTimelineClient::init_upload_queue`] and [`RemoteTimelineClient::init_upload_queue_for_empty_remote`].
     #[error("queue is not initialized")]
     QueueUninitialized,
-    #[error("queue is broken")]
-    QueueBroken,
+}
+
+#[derive(Debug, thiserror::Error)]
+pub enum PersistIndexPartWithDeletedFlagError {
+    #[error("another task is already setting the deleted_flag, started at {0:?}")]
+    AlreadyInProgress(NaiveDateTime),
+    #[error("the deleted_flag was already set, value is {0:?}")]
+    AlreadyDeleted(NaiveDateTime),
+    #[error(transparent)]
+    Other(#[from] anyhow::Error),
 }
 
 /// A client for accessing a timeline's data in remote storage.
@@ -355,7 +364,6 @@ impl RemoteTimelineClient {
     pub fn last_uploaded_consistent_lsn(&self) -> Option<Lsn> {
         match &*self.upload_queue.lock().unwrap() {
             UploadQueue::Uninitialized => None,
-            UploadQueue::Broken => None, // could we return something?
             UploadQueue::Initialized(q) => Some(q.last_uploaded_consistent_lsn),
             UploadQueue::Stopped(q) => Some(q.last_uploaded_consistent_lsn),
         }
@@ -659,7 +667,7 @@ impl RemoteTimelineClient {
     /// Check method [`RemoteTimelineClient::stop`] for details.
     pub(crate) async fn persist_index_part_with_deleted_flag(
         self: &Arc<Self>,
-    ) -> anyhow::Result<()> {
+    ) -> Result<(), PersistIndexPartWithDeletedFlagError> {
         let index_part_with_deleted_at = {
             let mut locked = self.upload_queue.lock().unwrap();
 
@@ -667,20 +675,26 @@ impl RemoteTimelineClient {
             // we can have inprogress index part upload that can overwrite the file
             // with missing is_deleted flag that we going to set below
             let stopped = match &mut *locked {
-                UploadQueue::Uninitialized | UploadQueue::Initialized(_) | UploadQueue::Broken => {
-                    anyhow::bail!(
-                        "upload queue must be in state Stopped, but is in state {}",
-                        locked.as_str()
-                    );
+                UploadQueue::Uninitialized => {
+                    return Err(anyhow::anyhow!("is not Stopped but Uninitialized").into())
+                }
+                UploadQueue::Initialized(_) => {
+                    return Err(anyhow::anyhow!("is not Stopped but Initialized").into())
                 }
                 UploadQueue::Stopped(stopped) => stopped,
             };
 
-            if let Some(deleted_at) = stopped.deleted_at.as_ref() {
-                anyhow::bail!("timeline is deleting, deleted_at: {:?}", deleted_at);
-            }
+            match stopped.deleted_at {
+                SetDeletedFlagProgress::NotRunning => (), // proceed
+                SetDeletedFlagProgress::InProgress(at) => {
+                    return Err(PersistIndexPartWithDeletedFlagError::AlreadyInProgress(at));
+                }
+                SetDeletedFlagProgress::Successful(at) => {
+                    return Err(PersistIndexPartWithDeletedFlagError::AlreadyDeleted(at));
+                }
+            };
             let deleted_at = Utc::now().naive_utc();
-            stopped.deleted_at = Some(deleted_at);
+            stopped.deleted_at = SetDeletedFlagProgress::InProgress(deleted_at);
 
             let mut index_part = IndexPart::new(
                 stopped.latest_files.clone(),
@@ -697,17 +711,17 @@ impl RemoteTimelineClient {
         let undo_deleted_at = scopeguard::guard(Arc::clone(self), |self_clone| {
             let mut locked = self_clone.upload_queue.lock().unwrap();
             let stopped = match &mut *locked {
-                UploadQueue::Broken | UploadQueue::Uninitialized | UploadQueue::Initialized(_) => {
-                    unreachable!(
-                        "there's no way out of Stopping, and we checked it's Stopping above: {:?}",
-                        locked.as_str(),
-                    )
-                }
+                UploadQueue::Uninitialized | UploadQueue::Initialized(_) => unreachable!(
+                    "there's no way out of Stopping, and we checked it's Stopping above: {:?}",
+                    locked.as_str(),
+                ),
                 UploadQueue::Stopped(stopped) => stopped,
             };
-            stopped.deleted_at = None;
+            stopped.deleted_at = SetDeletedFlagProgress::NotRunning;
         });
 
+        // Have a failpoint that can use the `pause` failpoint action.
+        // We don't want to block the executor thread, hence, spawn_blocking + await.
         #[cfg(feature = "testing")]
         tokio::task::spawn_blocking({
             let current = tracing::Span::current();
@@ -733,8 +747,23 @@ impl RemoteTimelineClient {
         )
         .await?;
 
-        // all good, keep the deleted_at flag
+        // all good, disarm the guard and mark as success
         ScopeGuard::into_inner(undo_deleted_at);
+        {
+            let mut locked = self.upload_queue.lock().unwrap();
+            let stopped = match &mut *locked {
+                UploadQueue::Uninitialized | UploadQueue::Initialized(_) => unreachable!(
+                    "there's no way out of Stopping, and we checked it's Stopping above: {:?}",
+                    locked.as_str(),
+                ),
+                UploadQueue::Stopped(stopped) => stopped,
+            };
+            stopped.deleted_at = SetDeletedFlagProgress::Successful(
+                index_part_with_deleted_at
+                    .deleted_at
+                    .expect("we set it above"),
+            );
+        }
 
         Ok(())
     }
@@ -855,14 +884,9 @@ impl RemoteTimelineClient {
             // is cancellation safe, so we don't dare to do that. Hopefully, the
             // upload finishes or times out soon enough.
             if task_mgr::is_shutdown_requested() {
-                info!("upload task cancelled by shutdown request, stopping queue");
+                info!("upload task cancelled by shutdown request");
                 match self.stop() {
                     Ok(()) => {}
-                    Err(StopError::QueueBroken) => {
-                        warn!("stop() observed upload queue as broken");
-                        // In this case, it's still ok to proceed with balancing out the metric and returning.
-                        // (The metric has nothing to do with the queue state itself).
-                    }
                     Err(StopError::QueueUninitialized) => {
                         unreachable!("we never launch an upload task if the queue is uninitialized, and once it is initialized, we never go back")
                     }
@@ -992,10 +1016,6 @@ impl RemoteTimelineClient {
                     info!("another concurrent task already stopped the queue");
                     return;
                 }, // nothing to do
-                UploadQueue::Broken => {
-                    warn!("the upload queue became broken while the task was running");
-                    return;
-                }
                 UploadQueue::Initialized(qi) => { qi }
             };
 
@@ -1084,76 +1104,74 @@ impl RemoteTimelineClient {
         // into stopped state, thereby dropping all off the queued *ops* which haven't become *tasks* yet.
         // The other *tasks* will come here and observe an already shut down queue and hence simply wrap up their business.
         let mut guard = self.upload_queue.lock().unwrap();
-        // If any of the code below panics, the queue remains in Broken state.
-        // If we're coming from Initialized state, `queued_operations` will get dropped
-        // as part of the panic, because it sits in the local variable named `owned`.
-        // Any `wait_completion` operations against those queued operations
-        // will observe an error. That's exactly what we want.
-        // We don't need to care about in-progress operations because that responsibility
-        // lies with the caller. There's no point for them to try anything funky, like,
-        // catching the panic and retrying the stop() call. We will return QueueBroken in that case.
-        let owned = std::mem::replace(&mut *guard, UploadQueue::Broken);
-        let res;
-        *guard = match owned {
-            UploadQueue::Broken => {
-                res = Err(StopError::QueueBroken);
-                owned
-            }
-            UploadQueue::Uninitialized => {
-                res = Err(StopError::QueueUninitialized);
-                owned
-            }
+        match &mut *guard {
+            UploadQueue::Uninitialized => Err(StopError::QueueUninitialized),
             UploadQueue::Stopped(_) => {
                 // nothing to do
                 info!("another concurrent task already shut down the queue");
-                res = Ok(());
-                owned
+                Ok(())
             }
-            UploadQueue::Initialized(qi) => {
-                let UploadQueueInitialized {
-                    task_counter: _,
-                    latest_files,
-                    // XXX need to think about what it means if it's non-zero here
-                    latest_files_changes_since_metadata_upload_scheduled: _,
-                    latest_metadata,
-                    last_uploaded_consistent_lsn,
-                    num_inprogress_layer_uploads,
-                    num_inprogress_metadata_uploads,
-                    num_inprogress_deletions,
-                    inprogress_tasks,
-                    queued_operations,
-                } = qi;
+            UploadQueue::Initialized(UploadQueueInitialized {
+                latest_files,
+                latest_metadata,
+                last_uploaded_consistent_lsn,
+                ..
+            }) => {
+                info!("shutting down upload queue");
+
+                // Replace the queue with the Stopped state, taking ownership of the old
+                // Initialized queue. We will do some checks on it, and then drop it.
+                let qi = {
+                    // take or clone what we need
+                    let latest_files = std::mem::take(latest_files);
+                    let last_uploaded_consistent_lsn = *last_uploaded_consistent_lsn;
+                    // this could be Copy
+                    let latest_metadata = latest_metadata.clone();
+
+                    let stopped = UploadQueueStopped {
+                        latest_files,
+                        last_uploaded_consistent_lsn,
+                        latest_metadata,
+                        deleted_at: SetDeletedFlagProgress::NotRunning,
+                    };
+
+                    let upload_queue =
+                        std::mem::replace(&mut *guard, UploadQueue::Stopped(stopped));
+                    if let UploadQueue::Initialized(qi) = upload_queue {
+                        qi
+                    } else {
+                        unreachable!("we checked in the match above that it is Initialized");
+                    }
+                };
+
+                assert!(qi.latest_files.is_empty(), "do not use this anymore");
 
                 // consistency check
                 assert_eq!(
-                    num_inprogress_layer_uploads
-                        + num_inprogress_metadata_uploads
-                        + num_inprogress_deletions,
-                    inprogress_tasks.len()
+                    qi.num_inprogress_layer_uploads
+                        + qi.num_inprogress_metadata_uploads
+                        + qi.num_inprogress_deletions,
+                    qi.inprogress_tasks.len()
                 );
 
                 // We don't need to do anything here for in-progress tasks. They will finish
                 // on their own, decrement the unfinished-task counter themselves, and observe
                 // that the queue is Stopped.
-                drop(inprogress_tasks);
+                drop(qi.inprogress_tasks);
 
                 // Tear down queued ops
-                for op in queued_operations.into_iter() {
+                for op in qi.queued_operations.into_iter() {
                     self.calls_unfinished_metric_end(&op);
                     // Dropping UploadOp::Barrier() here will make wait_completion() return with an Err()
                     // which is exactly what we want to happen.
                     drop(op);
                 }
-                res = Ok(());
-                UploadQueue::Stopped(UploadQueueStopped {
-                    latest_files,
-                    last_uploaded_consistent_lsn,
-                    latest_metadata,
-                    deleted_at: None,
-                })
+
+                // We're done.
+                drop(guard);
+                Ok(())
             }
-        };
-        res
+        }
     }
 }
 

pageserver/src/tenant/storage_layer.rs

@@ -13,9 +13,9 @@ use crate::task_mgr::TaskKind;
 use crate::walrecord::NeonWalRecord;
 use anyhow::Result;
 use bytes::Bytes;
-use either::Either;
 use enum_map::EnumMap;
 use enumset::EnumSet;
+use once_cell::sync::Lazy;
 use pageserver_api::models::LayerAccessKind;
 use pageserver_api::models::{
     HistoricLayerInfo, LayerResidenceEvent, LayerResidenceEventReason, LayerResidenceStatus,
@@ -23,8 +23,10 @@ use pageserver_api::models::{
 use std::ops::Range;
 use std::path::PathBuf;
 use std::sync::{Arc, Mutex};
-use std::time::{SystemTime, UNIX_EPOCH};
+use std::time::{Duration, SystemTime, UNIX_EPOCH};
+use tracing::warn;
 use utils::history_buffer::HistoryBufferWithDropCounter;
+use utils::rate_limit::RateLimit;
 
 use utils::{
     id::{TenantId, TimelineId},
@@ -37,6 +39,8 @@ pub use image_layer::{ImageLayer, ImageLayerWriter};
 pub use inmemory_layer::InMemoryLayer;
 pub use remote_layer::RemoteLayer;
 
+use super::layer_map::BatchedUpdates;
+
 pub fn range_overlaps<T>(a: &Range<T>, b: &Range<T>) -> bool
 where
     T: PartialOrd<T>,
@@ -158,41 +162,82 @@ impl LayerAccessStatFullDetails {
 }
 
 impl LayerAccessStats {
-    pub(crate) fn for_loading_layer(status: LayerResidenceStatus) -> Self {
-        let new = LayerAccessStats(Mutex::new(LayerAccessStatsLocked::default()));
-        new.record_residence_event(status, LayerResidenceEventReason::LayerLoad);
-        new
+    /// Create an empty stats object.
+    ///
+    /// The caller is responsible for recording a residence event
+    /// using [`record_residence_event`] before calling `latest_activity`.
+    /// If they don't, [`latest_activity`] will return `None`.
+    pub(crate) fn empty_will_record_residence_event_later() -> Self {
+        LayerAccessStats(Mutex::default())
     }
 
-    pub(crate) fn for_new_layer_file() -> Self {
+    /// Create an empty stats object and record a [`LayerLoad`] event with the given residence status.
+    ///
+    /// See [`record_residence_event`] for why you need to do this while holding the layer map lock.
+    pub(crate) fn for_loading_layer<L>(
+        layer_map_lock_held_witness: &BatchedUpdates<'_, L>,
+        status: LayerResidenceStatus,
+    ) -> Self
+    where
+        L: ?Sized + Layer,
+    {
         let new = LayerAccessStats(Mutex::new(LayerAccessStatsLocked::default()));
         new.record_residence_event(
-            LayerResidenceStatus::Resident,
-            LayerResidenceEventReason::LayerCreate,
+            layer_map_lock_held_witness,
+            status,
+            LayerResidenceEventReason::LayerLoad,
         );
         new
     }
 
     /// Creates a clone of `self` and records `new_status` in the clone.
-    /// The `new_status` is not recorded in `self`
-    pub(crate) fn clone_for_residence_change(
+    ///
+    /// The `new_status` is not recorded in `self`.
+    ///
+    /// See [`record_residence_event`] for why you need to do this while holding the layer map lock.
+    pub(crate) fn clone_for_residence_change<L>(
         &self,
+        layer_map_lock_held_witness: &BatchedUpdates<'_, L>,
         new_status: LayerResidenceStatus,
-    ) -> LayerAccessStats {
+    ) -> LayerAccessStats
+    where
+        L: ?Sized + Layer,
+    {
         let clone = {
             let inner = self.0.lock().unwrap();
             inner.clone()
         };
         let new = LayerAccessStats(Mutex::new(clone));
-        new.record_residence_event(new_status, LayerResidenceEventReason::ResidenceChange);
+        new.record_residence_event(
+            layer_map_lock_held_witness,
+            new_status,
+            LayerResidenceEventReason::ResidenceChange,
+        );
         new
     }
 
-    fn record_residence_event(
+    /// Record a change in layer residency.
+    ///
+    /// Recording the event must happen while holding the layer map lock to
+    /// ensure that latest-activity-threshold-based layer eviction (eviction_task.rs)
+    /// can do an "imitate access" to this layer, before it observes `now-latest_activity() > threshold`.
+    ///
+    /// If we instead recorded the residence event with a timestamp from before grabbing the layer map lock,
+    /// the following race could happen:
+    ///
+    /// - Compact: Write out an L1 layer from several L0 layers. This records residence event LayerCreate with the current timestamp.
+    /// - Eviction: imitate access logical size calculation. This accesses the L0 layers because the L1 layer is not yet in the layer map.
+    /// - Compact: Grab layer map lock, add the new L1 to layer map and remove the L0s, release layer map lock.
+    /// - Eviction: observes the new L1 layer whose only activity timestamp is the LayerCreate event.
+    ///
+    pub(crate) fn record_residence_event<L>(
         &self,
+        _layer_map_lock_held_witness: &BatchedUpdates<'_, L>,
         status: LayerResidenceStatus,
         reason: LayerResidenceEventReason,
-    ) {
+    ) where
+        L: ?Sized + Layer,
+    {
         let mut locked = self.0.lock().unwrap();
         locked.iter_mut().for_each(|inner| {
             inner
@@ -255,24 +300,37 @@ impl LayerAccessStats {
         ret
     }
 
-    fn most_recent_access_or_residence_event(
-        &self,
-    ) -> Either<LayerAccessStatFullDetails, LayerResidenceEvent> {
+    /// Get the latest access timestamp, falling back to latest residence event.
+    ///
+    /// This function can only return `None` if there has not yet been a call to the
+    /// [`record_residence_event`] method. That would generally be considered an
+    /// implementation error. This function logs a rate-limited warning in that case.
+    ///
+    /// TODO: use type system to avoid the need for `fallback`.
+    /// The approach in https://github.com/neondatabase/neon/pull/3775
+    /// could be used to enforce that a residence event is recorded
+    /// before a layer is added to the layer map. We could also have
+    /// a layer wrapper type that holds the LayerAccessStats, and ensure
+    /// that that type can only be produced by inserting into the layer map.
+    pub(crate) fn latest_activity(&self) -> Option<SystemTime> {
         let locked = self.0.lock().unwrap();
         let inner = &locked.for_eviction_policy;
         match inner.last_accesses.recent() {
-            Some(a) => Either::Left(*a),
+            Some(a) => Some(a.when),
             None => match inner.last_residence_changes.recent() {
-                Some(e) => Either::Right(e.clone()),
-                None => unreachable!("constructors for LayerAccessStats ensure that there's always a residence change event"),
-            }
-        }
-    }
-
-    pub(crate) fn latest_activity(&self) -> SystemTime {
-        match self.most_recent_access_or_residence_event() {
-            Either::Left(mra) => mra.when,
-            Either::Right(re) => re.timestamp,
+                Some(e) => Some(e.timestamp),
+                None => {
+                    static WARN_RATE_LIMIT: Lazy<Mutex<(usize, RateLimit)>> =
+                        Lazy::new(|| Mutex::new((0, RateLimit::new(Duration::from_secs(10)))));
+                    let mut guard = WARN_RATE_LIMIT.lock().unwrap();
+                    guard.0 += 1;
+                    let occurences = guard.0;
+                    guard.1.call(move || {
+                        warn!(parent: None, occurences, "latest_activity not available, this is an implementation bug, using fallback value");
+                    });
+                    None
+                }
+            },
         }
     }
 }

pageserver/src/tenant/storage_layer/delta_layer.rs

@@ -57,7 +57,7 @@ use utils::{
 
 use super::{
     DeltaFileName, Layer, LayerAccessStats, LayerAccessStatsReset, LayerFileName, LayerIter,
-    LayerKeyIter, LayerResidenceStatus, PathOrConf,
+    LayerKeyIter, PathOrConf,
 };
 
 ///
@@ -637,7 +637,7 @@ impl DeltaLayer {
             key_range: summary.key_range,
             lsn_range: summary.lsn_range,
             file_size: metadata.len(),
-            access_stats: LayerAccessStats::for_loading_layer(LayerResidenceStatus::Resident),
+            access_stats: LayerAccessStats::empty_will_record_residence_event_later(),
             inner: RwLock::new(DeltaLayerInner {
                 loaded: false,
                 file: None,
@@ -808,7 +808,7 @@ impl DeltaLayerWriterInner {
             key_range: self.key_start..key_end,
             lsn_range: self.lsn_range.clone(),
             file_size: metadata.len(),
-            access_stats: LayerAccessStats::for_new_layer_file(),
+            access_stats: LayerAccessStats::empty_will_record_residence_event_later(),
             inner: RwLock::new(DeltaLayerInner {
                 loaded: false,
                 file: None,

pageserver/src/tenant/storage_layer/image_layer.rs

@@ -53,7 +53,7 @@ use utils::{
 };
 
 use super::filename::{ImageFileName, LayerFileName};
-use super::{Layer, LayerAccessStatsReset, LayerIter, LayerResidenceStatus, PathOrConf};
+use super::{Layer, LayerAccessStatsReset, LayerIter, PathOrConf};
 
 ///
 /// Header stored in the beginning of the file
@@ -438,7 +438,7 @@ impl ImageLayer {
             key_range: summary.key_range,
             lsn: summary.lsn,
             file_size: metadata.len(),
-            access_stats: LayerAccessStats::for_loading_layer(LayerResidenceStatus::Resident),
+            access_stats: LayerAccessStats::empty_will_record_residence_event_later(),
             inner: RwLock::new(ImageLayerInner {
                 file: None,
                 loaded: false,
@@ -598,7 +598,7 @@ impl ImageLayerWriterInner {
             key_range: self.key_range.clone(),
             lsn: self.lsn,
             file_size: metadata.len(),
-            access_stats: LayerAccessStats::for_new_layer_file(),
+            access_stats: LayerAccessStats::empty_will_record_residence_event_later(),
             inner: RwLock::new(ImageLayerInner {
                 loaded: false,
                 file: None,

pageserver/src/tenant/storage_layer/remote_layer.rs

@@ -4,6 +4,7 @@
 use crate::config::PageServerConf;
 use crate::context::RequestContext;
 use crate::repository::Key;
+use crate::tenant::layer_map::BatchedUpdates;
 use crate::tenant::remote_timeline_client::index::LayerFileMetadata;
 use crate::tenant::storage_layer::{Layer, ValueReconstructResult, ValueReconstructState};
 use anyhow::{bail, Result};
@@ -246,11 +247,15 @@ impl RemoteLayer {
     }
 
     /// Create a Layer struct representing this layer, after it has been downloaded.
-    pub fn create_downloaded_layer(
+    pub fn create_downloaded_layer<L>(
         &self,
+        layer_map_lock_held_witness: &BatchedUpdates<'_, L>,
         conf: &'static PageServerConf,
         file_size: u64,
-    ) -> Arc<dyn PersistentLayer> {
+    ) -> Arc<dyn PersistentLayer>
+    where
+        L: ?Sized + Layer,
+    {
         if self.is_delta {
             let fname = DeltaFileName {
                 key_range: self.key_range.clone(),
@@ -262,8 +267,10 @@ impl RemoteLayer {
                 self.tenantid,
                 &fname,
                 file_size,
-                self.access_stats
-                    .clone_for_residence_change(LayerResidenceStatus::Resident),
+                self.access_stats.clone_for_residence_change(
+                    layer_map_lock_held_witness,
+                    LayerResidenceStatus::Resident,
+                ),
             ))
         } else {
             let fname = ImageFileName {
@@ -276,8 +283,10 @@ impl RemoteLayer {
                 self.tenantid,
                 &fname,
                 file_size,
-                self.access_stats
-                    .clone_for_residence_change(LayerResidenceStatus::Resident),
+                self.access_stats.clone_for_residence_change(
+                    layer_map_lock_held_witness,
+                    LayerResidenceStatus::Resident,
+                ),
             ))
         }
     }

pageserver/src/tenant/timeline.rs

@@ -11,7 +11,8 @@ use itertools::Itertools;
 use once_cell::sync::OnceCell;
 use pageserver_api::models::{
     DownloadRemoteLayersTaskInfo, DownloadRemoteLayersTaskSpawnRequest,
-    DownloadRemoteLayersTaskState, LayerMapInfo, LayerResidenceStatus, TimelineState,
+    DownloadRemoteLayersTaskState, LayerMapInfo, LayerResidenceEventReason, LayerResidenceStatus,
+    TimelineState,
 };
 use remote_storage::GenericRemoteStorage;
 use storage_broker::BrokerClientChannel;
@@ -1111,7 +1112,7 @@ impl Timeline {
                 &layer_metadata,
                 local_layer
                     .access_stats()
-                    .clone_for_residence_change(LayerResidenceStatus::Evicted),
+                    .clone_for_residence_change(batch_updates, LayerResidenceStatus::Evicted),
             ),
             LayerFileName::Delta(delta_name) => RemoteLayer::new_delta(
                 self.tenant_id,
@@ -1120,7 +1121,7 @@ impl Timeline {
                 &layer_metadata,
                 local_layer
                     .access_stats()
-                    .clone_for_residence_change(LayerResidenceStatus::Evicted),
+                    .clone_for_residence_change(batch_updates, LayerResidenceStatus::Evicted),
             ),
         });
 
@@ -1489,7 +1490,7 @@ impl Timeline {
                     self.tenant_id,
                     &imgfilename,
                     file_size,
-                    LayerAccessStats::for_loading_layer(LayerResidenceStatus::Resident),
+                    LayerAccessStats::for_loading_layer(&updates, LayerResidenceStatus::Resident),
                 );
 
                 trace!("found layer {}", layer.path().display());
@@ -1521,7 +1522,7 @@ impl Timeline {
                     self.tenant_id,
                     &deltafilename,
                     file_size,
-                    LayerAccessStats::for_loading_layer(LayerResidenceStatus::Resident),
+                    LayerAccessStats::for_loading_layer(&updates, LayerResidenceStatus::Resident),
                 );
 
                 trace!("found layer {}", layer.path().display());
@@ -1657,7 +1658,10 @@ impl Timeline {
                         self.timeline_id,
                         imgfilename,
                         &remote_layer_metadata,
-                        LayerAccessStats::for_loading_layer(LayerResidenceStatus::Evicted),
+                        LayerAccessStats::for_loading_layer(
+                            &updates,
+                            LayerResidenceStatus::Evicted,
+                        ),
                     );
                     let remote_layer = Arc::new(remote_layer);
 
@@ -1682,7 +1686,10 @@ impl Timeline {
                         self.timeline_id,
                         deltafilename,
                         &remote_layer_metadata,
-                        LayerAccessStats::for_loading_layer(LayerResidenceStatus::Evicted),
+                        LayerAccessStats::for_loading_layer(
+                            &updates,
+                            LayerResidenceStatus::Evicted,
+                        ),
                     );
                     let remote_layer = Arc::new(remote_layer);
                     updates.insert_historic(remote_layer);
@@ -2729,11 +2736,16 @@ impl Timeline {
         ])?;
 
         // Add it to the layer map
-        self.layers
-            .write()
-            .unwrap()
-            .batch_update()
-            .insert_historic(Arc::new(new_delta));
+        let l = Arc::new(new_delta);
+        let mut layers = self.layers.write().unwrap();
+        let mut batch_updates = layers.batch_update();
+        l.access_stats().record_residence_event(
+            &batch_updates,
+            LayerResidenceStatus::Resident,
+            LayerResidenceEventReason::LayerCreate,
+        );
+        batch_updates.insert_historic(l);
+        batch_updates.flush();
 
         // update the timeline's physical size
         let sz = new_delta_path.metadata()?.len();
@@ -2938,7 +2950,13 @@ impl Timeline {
             self.metrics
                 .resident_physical_size_gauge
                 .add(metadata.len());
-            updates.insert_historic(Arc::new(l));
+            let l = Arc::new(l);
+            l.access_stats().record_residence_event(
+                &updates,
+                LayerResidenceStatus::Resident,
+                LayerResidenceEventReason::LayerCreate,
+            );
+            updates.insert_historic(l);
         }
         updates.flush();
         drop(layers);
@@ -3371,6 +3389,11 @@ impl Timeline {
 
             new_layer_paths.insert(new_delta_path, LayerFileMetadata::new(metadata.len()));
             let x: Arc<dyn PersistentLayer + 'static> = Arc::new(l);
+            x.access_stats().record_residence_event(
+                &updates,
+                LayerResidenceStatus::Resident,
+                LayerResidenceEventReason::LayerCreate,
+            );
             updates.insert_historic(x);
         }
 
@@ -3881,9 +3904,9 @@ impl Timeline {
 
                     // Download complete. Replace the RemoteLayer with the corresponding
                     // Delta- or ImageLayer in the layer map.
-                    let new_layer = remote_layer.create_downloaded_layer(self_clone.conf, *size);
                     let mut layers = self_clone.layers.write().unwrap();
                     let mut updates = layers.batch_update();
+                    let new_layer = remote_layer.create_downloaded_layer(&updates, self_clone.conf, *size);
                     {
                         use crate::tenant::layer_map::Replacement;
                         let l: Arc<dyn PersistentLayer> = remote_layer.clone();
@@ -4155,7 +4178,15 @@ impl Timeline {
                 continue;
             }
 
-            let last_activity_ts = l.access_stats().latest_activity();
+            let last_activity_ts = l
+                .access_stats()
+                .latest_activity()
+                .unwrap_or_else(|| {
+                    // We only use this fallback if there's an implementation error.
+                    // `latest_activity` already does rate-limited warn!() log.
+                    debug!(layer=%l.filename().file_name(), "last_activity returns None, using SystemTime::now");
+                    SystemTime::now()
+                });
 
             resident_layers.push(LocalLayerInfoForDiskUsageEviction {
                 layer: l,

pageserver/src/tenant/timeline/eviction_task.rs

@@ -22,7 +22,7 @@ use std::{
 
 use tokio::time::Instant;
 use tokio_util::sync::CancellationToken;
-use tracing::{debug, error, info, instrument, warn};
+use tracing::{debug, error, info, info_span, instrument, warn, Instrument};
 
 use crate::{
     context::{DownloadBehavior, RequestContext},
@@ -184,7 +184,14 @@ impl Timeline {
                 if hist_layer.is_remote_layer() {
                     continue;
                 }
-                let last_activity_ts = hist_layer.access_stats().latest_activity();
+
+                let last_activity_ts = hist_layer.access_stats().latest_activity().unwrap_or_else(|| {
+                    // We only use this fallback if there's an implementation error.
+                    // `latest_activity` already does rate-limited warn!() log.
+                    debug!(layer=%hist_layer.filename().file_name(), "last_activity returns None, using SystemTime::now");
+                    SystemTime::now()
+                });
+
                 let no_activity_for = match now.duration_since(last_activity_ts) {
                     Ok(d) => d,
                     Err(_e) => {
@@ -269,6 +276,7 @@ impl Timeline {
         ControlFlow::Continue(())
     }
 
+    #[instrument(skip_all)]
     async fn imitate_layer_accesses(
         &self,
         p: &EvictionPolicyLayerAccessThreshold,
@@ -317,6 +325,7 @@ impl Timeline {
     }
 
     /// Recompute the values which would cause on-demand downloads during restart.
+    #[instrument(skip_all)]
     async fn imitate_timeline_cached_layer_accesses(
         &self,
         cancel: &CancellationToken,
@@ -325,7 +334,10 @@ impl Timeline {
         let lsn = self.get_last_record_lsn();
 
         // imitiate on-restart initial logical size
-        let size = self.calculate_logical_size(lsn, cancel.clone(), ctx).await;
+        let size = self
+            .calculate_logical_size(lsn, cancel.clone(), ctx)
+            .instrument(info_span!("calculate_logical_size"))
+            .await;
 
         match &size {
             Ok(_size) => {
@@ -340,7 +352,11 @@ impl Timeline {
         }
 
         // imitiate repartiting on first compactation
-        if let Err(e) = self.collect_keyspace(lsn, ctx).await {
+        if let Err(e) = self
+            .collect_keyspace(lsn, ctx)
+            .instrument(info_span!("collect_keyspace"))
+            .await
+        {
             // if this failed, we probably failed logical size because these use the same keys
             if size.is_err() {
                 // ignore, see above comment
@@ -353,6 +369,7 @@ impl Timeline {
     }
 
     // Imitate the synthetic size calculation done by the consumption_metrics module.
+    #[instrument(skip_all)]
     async fn imitate_synthetic_size_calculation_worker(
         &self,
         tenant: &Arc<Tenant>,
@@ -391,7 +408,8 @@ impl Timeline {
 
         let mut throwaway_cache = HashMap::new();
         let gather =
-            crate::tenant::size::gather_inputs(tenant, limit, None, &mut throwaway_cache, ctx);
+            crate::tenant::size::gather_inputs(tenant, limit, None, &mut throwaway_cache, ctx)
+                .instrument(info_span!("gather_inputs"));
 
         tokio::select! {
             _ = cancel.cancelled() => {}

pageserver/src/tenant/upload_queue.rs

@@ -23,7 +23,6 @@ pub(super) enum UploadQueue {
     Uninitialized,
     Initialized(UploadQueueInitialized),
     Stopped(UploadQueueStopped),
-    Broken,
 }
 
 impl UploadQueue {
@@ -32,7 +31,6 @@ impl UploadQueue {
             UploadQueue::Uninitialized => "Uninitialized",
             UploadQueue::Initialized(_) => "Initialized",
             UploadQueue::Stopped(_) => "Stopped",
-            UploadQueue::Broken => "Broken",
         }
     }
 }
@@ -78,12 +76,18 @@ pub(crate) struct UploadQueueInitialized {
     pub(crate) queued_operations: VecDeque<UploadOp>,
 }
 
+#[derive(Clone, Copy)]
+pub(super) enum SetDeletedFlagProgress {
+    NotRunning,
+    InProgress(NaiveDateTime),
+    Successful(NaiveDateTime),
+}
+
 pub(super) struct UploadQueueStopped {
     pub(super) latest_files: HashMap<LayerFileName, LayerFileMetadata>,
     pub(super) last_uploaded_consistent_lsn: Lsn,
     pub(super) latest_metadata: TimelineMetadata,
-    /// If Some(), a call to `persist_index_part_with_deleted_flag` is ongoing or finished.
-    pub(super) deleted_at: Option<NaiveDateTime>,
+    pub(super) deleted_at: SetDeletedFlagProgress,
 }
 
 impl UploadQueue {
@@ -93,7 +97,7 @@ impl UploadQueue {
     ) -> anyhow::Result<&mut UploadQueueInitialized> {
         match self {
             UploadQueue::Uninitialized => (),
-            UploadQueue::Initialized(_) | UploadQueue::Stopped(_) | UploadQueue::Broken => {
+            UploadQueue::Initialized(_) | UploadQueue::Stopped(_) => {
                 anyhow::bail!("already initialized, state {}", self.as_str())
             }
         }
@@ -127,7 +131,7 @@ impl UploadQueue {
     ) -> anyhow::Result<&mut UploadQueueInitialized> {
         match self {
             UploadQueue::Uninitialized => (),
-            UploadQueue::Initialized(_) | UploadQueue::Stopped(_) | UploadQueue::Broken => {
+            UploadQueue::Initialized(_) | UploadQueue::Stopped(_) => {
                 anyhow::bail!("already initialized, state {}", self.as_str())
             }
         }
@@ -177,7 +181,7 @@ impl UploadQueue {
 
     pub(crate) fn initialized_mut(&mut self) -> anyhow::Result<&mut UploadQueueInitialized> {
         match self {
-            UploadQueue::Broken | UploadQueue::Uninitialized | UploadQueue::Stopped(_) => {
+            UploadQueue::Uninitialized | UploadQueue::Stopped(_) => {
                 anyhow::bail!("queue is in state {}", self.as_str())
             }
             UploadQueue::Initialized(x) => Ok(x),

safekeeper/src/bin/safekeeper.rs

@@ -108,6 +108,9 @@ struct Args {
     /// available to the system.
     #[arg(long)]
     wal_backup_threads: Option<usize>,
+    /// Number of max parallel WAL segments to be offloaded to remote storage.
+    #[arg(long, default_value = "5")]
+    wal_backup_parallel_jobs: usize,
     /// Disable WAL backup to s3. When disabled, safekeeper removes WAL ignoring
     /// WAL backup horizon.
     #[arg(long)]
@@ -182,6 +185,7 @@ fn main() -> anyhow::Result<()> {
         max_offloader_lag_bytes: args.max_offloader_lag,
         backup_runtime_threads: args.wal_backup_threads,
         wal_backup_enabled: !args.disable_wal_backup,
+        backup_parallel_jobs: args.wal_backup_parallel_jobs,
         auth,
     };
 

safekeeper/src/lib.rs

@@ -61,6 +61,7 @@ pub struct SafeKeeperConf {
     pub remote_storage: Option<RemoteStorageConfig>,
     pub max_offloader_lag_bytes: u64,
     pub backup_runtime_threads: Option<usize>,
+    pub backup_parallel_jobs: usize,
     pub wal_backup_enabled: bool,
     pub auth: Option<Arc<JwtAuth>>,
 }
@@ -93,6 +94,7 @@ impl SafeKeeperConf {
             broker_keepalive_interval: Duration::from_secs(5),
             backup_runtime_threads: None,
             wal_backup_enabled: true,
+            backup_parallel_jobs: 1,
             auth: None,
             heartbeat_timeout: Duration::new(5, 0),
             max_offloader_lag_bytes: defaults::DEFAULT_MAX_OFFLOADER_LAG_BYTES,

safekeeper/src/send_wal.rs

@@ -384,6 +384,8 @@ impl SafekeeperPostgresHandler {
             self.appname.clone(),
         ));
 
+        let commit_lsn_watch_rx = tli.get_commit_lsn_watch_rx();
+
         // Walproposer gets special handling: safekeeper must give proposer all
         // local WAL till the end, whether committed or not (walproposer will
         // hang otherwise). That's because walproposer runs the consensus and
@@ -399,7 +401,14 @@ impl SafekeeperPostgresHandler {
         } else {
             None
         };
-        let end_pos = stop_pos.unwrap_or(Lsn::INVALID);
+
+        // take the latest commit_lsn if don't have stop_pos
+        let mut end_pos = stop_pos.unwrap_or(*commit_lsn_watch_rx.borrow());
+
+        if end_pos < start_pos {
+            warn!("start_pos {} is ahead of end_pos {}", start_pos, end_pos);
+            end_pos = start_pos;
+        }
 
         info!(
             "starting streaming from {:?} till {:?}",
@@ -429,7 +438,7 @@ impl SafekeeperPostgresHandler {
             start_pos,
             end_pos,
             stop_pos,
-            commit_lsn_watch_rx: tli.get_commit_lsn_watch_rx(),
+            commit_lsn_watch_rx,
             ws_guard: ws_guard.clone(),
             wal_reader,
             send_buf: [0; MAX_SEND_SIZE],
@@ -456,6 +465,11 @@ struct WalSender<'a, IO> {
     // Position since which we are sending next chunk.
     start_pos: Lsn,
     // WAL up to this position is known to be locally available.
+    // Usually this is the same as the latest commit_lsn, but in case of
+    // walproposer recovery, this is flush_lsn.
+    //
+    // We send this LSN to the receiver as wal_end, so that it knows how much
+    // WAL this safekeeper has. This LSN should be as fresh as possible.
     end_pos: Lsn,
     // If present, terminate after reaching this position; used by walproposer
     // in recovery.
@@ -530,10 +544,18 @@ impl<IO: AsyncRead + AsyncWrite + Unpin> WalSender<'_, IO> {
     /// exit in the meanwhile
     async fn wait_wal(&mut self) -> Result<(), CopyStreamHandlerEnd> {
         loop {
+            self.end_pos = *self.commit_lsn_watch_rx.borrow();
+            if self.end_pos > self.start_pos {
+                // We have something to send.
+                return Ok(());
+            }
+
+            // Wait for WAL to appear, now self.end_pos == self.start_pos.
             if let Some(lsn) = wait_for_lsn(&mut self.commit_lsn_watch_rx, self.start_pos).await? {
                 self.end_pos = lsn;
                 return Ok(());
             }
+
             // Timed out waiting for WAL, check for termination and send KA
             if let Some(remote_consistent_lsn) = self
                 .ws_guard
@@ -551,7 +573,7 @@ impl<IO: AsyncRead + AsyncWrite + Unpin> WalSender<'_, IO> {
 
             self.pgb
                 .write_message(&BeMessage::KeepAlive(WalSndKeepAlive {
-                    sent_ptr: self.end_pos.0,
+                    wal_end: self.end_pos.0,
                     timestamp: get_current_timestamp(),
                     request_reply: true,
                 }))
@@ -618,11 +640,6 @@ const POLL_STATE_TIMEOUT: Duration = Duration::from_secs(1);
 /// - Ok(None) if timeout expired;
 /// - Err in case of error (if watch channel is in trouble, shouldn't happen).
 async fn wait_for_lsn(rx: &mut Receiver<Lsn>, lsn: Lsn) -> anyhow::Result<Option<Lsn>> {
-    let commit_lsn: Lsn = *rx.borrow();
-    if commit_lsn > lsn {
-        return Ok(Some(commit_lsn));
-    }
-
     let res = timeout(POLL_STATE_TIMEOUT, async move {
         let mut commit_lsn;
         loop {

safekeeper/src/wal_backup.rs

@@ -1,5 +1,7 @@
 use anyhow::{Context, Result};
 
+use futures::stream::FuturesOrdered;
+use futures::StreamExt;
 use tokio::task::JoinHandle;
 use utils::id::NodeId;
 
@@ -155,8 +157,14 @@ async fn update_task(
             let timeline_dir = conf.timeline_dir(&ttid);
 
             let handle = tokio::spawn(
-                backup_task_main(ttid, timeline_dir, conf.workdir.clone(), shutdown_rx)
-                    .instrument(info_span!("WAL backup task", ttid = %ttid)),
+                backup_task_main(
+                    ttid,
+                    timeline_dir,
+                    conf.workdir.clone(),
+                    conf.backup_parallel_jobs,
+                    shutdown_rx,
+                )
+                .instrument(info_span!("WAL backup task", ttid = %ttid)),
             );
 
             entry.handle = Some(WalBackupTaskHandle {
@@ -240,6 +248,7 @@ struct WalBackupTask {
     timeline_dir: PathBuf,
     workspace_dir: PathBuf,
     wal_seg_size: usize,
+    parallel_jobs: usize,
     commit_lsn_watch_rx: watch::Receiver<Lsn>,
 }
 
@@ -248,6 +257,7 @@ async fn backup_task_main(
     ttid: TenantTimelineId,
     timeline_dir: PathBuf,
     workspace_dir: PathBuf,
+    parallel_jobs: usize,
     mut shutdown_rx: Receiver<()>,
 ) {
     info!("started");
@@ -264,6 +274,7 @@ async fn backup_task_main(
         timeline: tli,
         timeline_dir,
         workspace_dir,
+        parallel_jobs,
     };
 
     // task is spinned up only when wal_seg_size already initialized
@@ -330,6 +341,7 @@ impl WalBackupTask {
                 self.wal_seg_size,
                 &self.timeline_dir,
                 &self.workspace_dir,
+                self.parallel_jobs,
             )
             .await
             {
@@ -356,20 +368,49 @@ pub async fn backup_lsn_range(
     wal_seg_size: usize,
     timeline_dir: &Path,
     workspace_dir: &Path,
+    parallel_jobs: usize,
 ) -> Result<()> {
+    if parallel_jobs < 1 {
+        anyhow::bail!("parallel_jobs must be >= 1");
+    }
+
     let start_lsn = *backup_lsn;
     let segments = get_segments(start_lsn, end_lsn, wal_seg_size);
-    for s in &segments {
-        backup_single_segment(s, timeline_dir, workspace_dir)
-            .await
-            .with_context(|| format!("offloading segno {}", s.seg_no))?;
 
-        let new_backup_lsn = s.end_lsn;
-        timeline
-            .set_wal_backup_lsn(new_backup_lsn)
-            .context("setting wal_backup_lsn")?;
-        *backup_lsn = new_backup_lsn;
+    // Pool of concurrent upload tasks. We use `FuturesOrdered` to
+    // preserve order of uploads, and update `backup_lsn` only after
+    // all previous uploads are finished.
+    let mut uploads = FuturesOrdered::new();
+    let mut iter = segments.iter();
+
+    loop {
+        let added_task = match iter.next() {
+            Some(s) => {
+                uploads.push_back(backup_single_segment(s, timeline_dir, workspace_dir));
+                true
+            }
+            None => false,
+        };
+
+        // Wait for the next segment to upload if we don't have any more segments,
+        // or if we have too many concurrent uploads.
+        if !added_task || uploads.len() >= parallel_jobs {
+            let next = uploads.next().await;
+            if let Some(res) = next {
+                // next segment uploaded
+                let segment = res?;
+                let new_backup_lsn = segment.end_lsn;
+                timeline
+                    .set_wal_backup_lsn(new_backup_lsn)
+                    .context("setting wal_backup_lsn")?;
+                *backup_lsn = new_backup_lsn;
+            } else {
+                // no more segments to upload
+                break;
+            }
+        }
     }
+
     info!(
         "offloaded segnos {:?} up to {}, previous backup_lsn {}",
         segments.iter().map(|&s| s.seg_no).collect::<Vec<_>>(),
@@ -383,7 +424,7 @@ async fn backup_single_segment(
     seg: &Segment,
     timeline_dir: &Path,
     workspace_dir: &Path,
-) -> Result<()> {
+) -> Result<Segment> {
     let segment_file_path = seg.file_path(timeline_dir)?;
     let remote_segment_path = segment_file_path
         .strip_prefix(workspace_dir)
@@ -404,7 +445,7 @@ async fn backup_single_segment(
     res?;
     debug!("Backup of {} done", segment_file_path.display());
 
-    Ok(())
+    Ok(*seg)
 }
 
 #[derive(Debug, Copy, Clone)]

test_runner/README.md

@@ -71,7 +71,8 @@ a subdirectory for each version with naming convention `v{PG_VERSION}/`.
 Inside that dir, a `bin/postgres` binary should be present.
 `DEFAULT_PG_VERSION`: The version of Postgres to use,
 This is used to construct full path to the postgres binaries.
-Format is 2-digit major version nubmer, i.e. `DEFAULT_PG_VERSION="14"`
+Format is 2-digit major version nubmer, i.e. `DEFAULT_PG_VERSION="14"`. Alternatively,
+you can use `--pg-version` argument.
 `TEST_OUTPUT`: Set the directory where test state and test output files
 should go.
 `TEST_SHARED_FIXTURES`: Try to re-use a single pageserver for all the tests.

test_runner/conftest.py

@@ -1,4 +1,5 @@
 pytest_plugins = (
+    "fixtures.pg_version",
     "fixtures.neon_fixtures",
     "fixtures.benchmark_fixture",
     "fixtures.pg_stats",

test_runner/fixtures/neon_fixtures.py

@@ -45,6 +45,7 @@ from typing_extensions import Literal
 from fixtures.log_helper import log
 from fixtures.pageserver.http import PageserverHttpClient
 from fixtures.pageserver.utils import wait_for_last_record_lsn, wait_for_upload
+from fixtures.pg_version import PgVersion
 from fixtures.types import Lsn, TenantId, TimelineId
 from fixtures.utils import (
     ATTACHMENT_NAME_REGEX,
@@ -75,7 +76,6 @@ Env = Dict[str, str]
 
 DEFAULT_OUTPUT_DIR: str = "test_output"
 DEFAULT_BRANCH_NAME: str = "main"
-DEFAULT_PG_VERSION_DEFAULT: str = "14"
 
 BASE_PORT: int = 15000
 WORKER_PORT_NUM: int = 1000
@@ -148,18 +148,7 @@ def top_output_dir(base_dir: Path) -> Iterator[Path]:
 
 
 @pytest.fixture(scope="session")
-def pg_version() -> Iterator[str]:
-    if env_default_pg_version := os.environ.get("DEFAULT_PG_VERSION"):
-        version = env_default_pg_version
-    else:
-        version = DEFAULT_PG_VERSION_DEFAULT
-
-    log.info(f"pg_version is {version}")
-    yield version
-
-
-@pytest.fixture(scope="session")
-def versioned_pg_distrib_dir(pg_distrib_dir: Path, pg_version: str) -> Iterator[Path]:
+def versioned_pg_distrib_dir(pg_distrib_dir: Path, pg_version: PgVersion) -> Iterator[Path]:
     versioned_dir = pg_distrib_dir / f"v{pg_version}"
 
     psql_bin_path = versioned_dir / "bin/psql"
@@ -292,6 +281,12 @@ def port_distributor(worker_base_port: int) -> PortDistributor:
     return PortDistributor(base_port=worker_base_port, port_number=WORKER_PORT_NUM)
 
 
+@pytest.fixture(scope="session")
+def httpserver_listen_address(port_distributor: PortDistributor):
+    port = port_distributor.get_port()
+    return ("localhost", port)
+
+
 @pytest.fixture(scope="function")
 def default_broker(
     port_distributor: PortDistributor,
@@ -586,7 +581,7 @@ class NeonEnvBuilder:
         mock_s3_server: MockS3Server,
         neon_binpath: Path,
         pg_distrib_dir: Path,
-        pg_version: str,
+        pg_version: PgVersion,
         remote_storage: Optional[RemoteStorage] = None,
         remote_storage_users: RemoteStorageUsers = RemoteStorageUsers.PAGESERVER,
         pageserver_config_override: Optional[str] = None,
@@ -1035,7 +1030,7 @@ def _shared_simple_env(
     top_output_dir: Path,
     neon_binpath: Path,
     pg_distrib_dir: Path,
-    pg_version: str,
+    pg_version: PgVersion,
 ) -> Iterator[NeonEnv]:
     """
     # Internal fixture backing the `neon_simple_env` fixture. If TEST_SHARED_FIXTURES
@@ -1092,7 +1087,7 @@ def neon_env_builder(
     mock_s3_server: MockS3Server,
     neon_binpath: Path,
     pg_distrib_dir: Path,
-    pg_version: str,
+    pg_version: PgVersion,
     default_broker: NeonBroker,
     run_id: uuid.UUID,
 ) -> Iterator[NeonEnvBuilder]:
@@ -1747,7 +1742,7 @@ def append_pageserver_param_overrides(
 class PgBin:
     """A helper class for executing postgres binaries"""
 
-    def __init__(self, log_dir: Path, pg_distrib_dir: Path, pg_version: str):
+    def __init__(self, log_dir: Path, pg_distrib_dir: Path, pg_version: PgVersion):
         self.log_dir = log_dir
         self.pg_version = pg_version
         self.pg_bin_path = pg_distrib_dir / f"v{pg_version}" / "bin"
@@ -1806,7 +1801,7 @@ class PgBin:
 
 
 @pytest.fixture(scope="function")
-def pg_bin(test_output_dir: Path, pg_distrib_dir: Path, pg_version: str) -> PgBin:
+def pg_bin(test_output_dir: Path, pg_distrib_dir: Path, pg_version: PgVersion) -> PgBin:
     return PgBin(test_output_dir, pg_distrib_dir, pg_version)
 
 
@@ -1894,7 +1889,7 @@ def vanilla_pg(
     test_output_dir: Path,
     port_distributor: PortDistributor,
     pg_distrib_dir: Path,
-    pg_version: str,
+    pg_version: PgVersion,
 ) -> Iterator[VanillaPostgres]:
     pgdatadir = test_output_dir / "pgdata-vanilla"
     pg_bin = PgBin(test_output_dir, pg_distrib_dir, pg_version)
@@ -1939,7 +1934,7 @@ class RemotePostgres(PgProtocol):
 
 @pytest.fixture(scope="function")
 def remote_pg(
-    test_output_dir: Path, pg_distrib_dir: Path, pg_version: str
+    test_output_dir: Path, pg_distrib_dir: Path, pg_version: PgVersion
 ) -> Iterator[RemotePostgres]:
     pg_bin = PgBin(test_output_dir, pg_distrib_dir, pg_version)
 
@@ -2623,7 +2618,7 @@ class Safekeeper:
 @dataclass
 class SafekeeperTimelineStatus:
     acceptor_epoch: int
-    pg_version: int
+    pg_version: int  # Not exactly a PgVersion, safekeeper returns version as int, for example 150002 for 15.2
     flush_lsn: Lsn
     commit_lsn: Lsn
     timeline_start_lsn: Lsn
@@ -2669,7 +2664,11 @@ class SafekeeperHttpClient(requests.Session):
         return res_json
 
     def timeline_create(
-        self, tenant_id: TenantId, timeline_id: TimelineId, pg_version: int, commit_lsn: Lsn
+        self,
+        tenant_id: TenantId,
+        timeline_id: TimelineId,
+        pg_version: int,  # Not exactly a PgVersion, safekeeper returns version as int, for example 150002 for 15.2
+        commit_lsn: Lsn,
     ):
         body = {
             "tenant_id": str(tenant_id),

test_runner/fixtures/pageserver/http.py

@@ -9,6 +9,7 @@ import requests
 
 from fixtures.log_helper import log
 from fixtures.metrics import Metrics, parse_metrics
+from fixtures.pg_version import PgVersion
 from fixtures.types import Lsn, TenantId, TimelineId
 from fixtures.utils import Fn
 
@@ -266,19 +267,21 @@ class PageserverHttpClient(requests.Session):
 
     def timeline_create(
         self,
+        pg_version: PgVersion,
         tenant_id: TenantId,
         new_timeline_id: Optional[TimelineId] = None,
         ancestor_timeline_id: Optional[TimelineId] = None,
         ancestor_start_lsn: Optional[Lsn] = None,
     ) -> Dict[Any, Any]:
-        res = self.post(
-            f"http://localhost:{self.port}/v1/tenant/{tenant_id}/timeline",
-            json={
-                "new_timeline_id": str(new_timeline_id) if new_timeline_id else None,
-                "ancestor_start_lsn": str(ancestor_start_lsn) if ancestor_start_lsn else None,
-                "ancestor_timeline_id": str(ancestor_timeline_id) if ancestor_timeline_id else None,
-            },
-        )
+        body: Dict[str, Any] = {
+            "new_timeline_id": str(new_timeline_id) if new_timeline_id else None,
+            "ancestor_start_lsn": str(ancestor_start_lsn) if ancestor_start_lsn else None,
+            "ancestor_timeline_id": str(ancestor_timeline_id) if ancestor_timeline_id else None,
+        }
+        if pg_version != PgVersion.NOT_SET:
+            body["pg_version"] = int(pg_version)
+
+        res = self.post(f"http://localhost:{self.port}/v1/tenant/{tenant_id}/timeline", json=body)
         self.verbose_error(res)
         if res.status_code == 409:
             raise Exception(f"could not create timeline: already exists for id {new_timeline_id}")

test_runner/fixtures/pg_version.py

@@ -0,0 +1,68 @@
+import enum
+import os
+from typing import Iterator, Optional
+
+import pytest
+from _pytest.config.argparsing import Parser
+from pytest import FixtureRequest
+
+from fixtures.log_helper import log
+
+"""
+This fixture is used to determine which version of Postgres to use for tests.
+"""
+
+
+# Inherit PgVersion from str rather than int to make it easier to pass as a command-line argument
+# TODO: use enum.StrEnum for Python >= 3.11
+@enum.unique
+class PgVersion(str, enum.Enum):
+    V14 = "14"
+    V15 = "15"
+    # Instead of making version an optional parameter in methods, we can use this fake entry
+    # to explicitly rely on the default server version (could be different from pg_version fixture value)
+    NOT_SET = "<-POSTRGRES VERSION IS NOT SET->"
+
+    # Make it less confusing in logs
+    def __repr__(self) -> str:
+        return f"'{self.value}'"
+
+    @classmethod
+    def _missing_(cls, value) -> Optional["PgVersion"]:
+        known_values = {v.value for _, v in cls.__members__.items()}
+
+        # Allow passing version as a string with "v" prefix (e.g. "v14")
+        if isinstance(value, str) and value.lower().startswith("v") and value[1:] in known_values:
+            return cls(value[1:])
+        # Allow passing version as an int (e.g. 15 or 150002, both will be converted to PgVersion.V15)
+        elif isinstance(value, int) and str(value)[:2] in known_values:
+            return cls(str(value)[:2])
+
+        # Make mypy happy
+        # See https://github.com/python/mypy/issues/3974
+        return None
+
+
+DEFAULT_VERSION: PgVersion = PgVersion.V14
+
+
+def pytest_addoption(parser: Parser):
+    parser.addoption(
+        "--pg-version",
+        action="store",
+        type=PgVersion,
+        help="Postgres version to use for tests",
+    )
+
+
+@pytest.fixture(scope="session")
+def pg_version(request: FixtureRequest) -> Iterator[PgVersion]:
+    if v := request.config.getoption("--pg-version"):
+        version, source = v, "from --pg-version commad-line argument"
+    elif v := os.environ.get("DEFAULT_PG_VERSION"):
+        version, source = PgVersion(v), "from DEFAULT_PG_VERSION environment variable"
+    else:
+        version, source = DEFAULT_VERSION, "default verson"
+
+    log.info(f"pg_version is {version} ({source})")
+    yield version

test_runner/pg_clients/csharp/npgsql/csharp-npgsql.csproj

@@ -8,7 +8,7 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="Npgsql" Version="7.0.0" />
+    <PackageReference Include="Npgsql" Version="7.0.4" />
   </ItemGroup>
 
 </Project>

test_runner/pg_clients/java/jdbc/.gitignore

@@ -1 +0,0 @@
-

test_runner/pg_clients/java/jdbc/Dockerfile

@@ -1,10 +1,10 @@
-FROM openjdk:17
+FROM openjdk:20
 WORKDIR /source
 
 COPY . .
 
 WORKDIR /app
-RUN curl --output postgresql.jar https://jdbc.postgresql.org/download/postgresql-42.5.1.jar && \
+RUN curl --output postgresql.jar https://jdbc.postgresql.org/download/postgresql-42.6.0.jar && \
     javac -d /app /source/Example.java
 
 CMD ["java", "-cp", "/app/postgresql.jar:.", "Example"]

test_runner/pg_clients/python/pg8000/requirements.txt

@@ -1 +1,2 @@
 pg8000==1.29.4
+scramp>=1.4.3

test_runner/pg_clients/rust/tokio-postgres/Cargo.lock

@@ -4,9 +4,9 @@ version = 3
 
 [[package]]
 name = "async-trait"
-version = "0.1.65"
+version = "0.1.68"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "095183a3539c7c7649b2beb87c2d3f0591f3a7fed07761cc546d244e27e0238c"
+checksum = "b9ccdd8f2a161be9bd5c023df56f1b2a0bd1d83872ae53b71a84a12c9bf6e842"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -21,9 +21,9 @@ checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa"
 
 [[package]]
 name = "base64"
-version = "0.13.1"
+version = "0.21.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9e1b586273c5702936fe7b7d6896644d8be71e6314cfe09d3167c95f712589e8"
+checksum = "a4a4ddaa51a5bc52a6948f74c06d20aaaddb71924eab79b8c97a8c556e942d6a"
 
 [[package]]
 name = "bitflags"
@@ -33,9 +33,9 @@ checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a"
 
 [[package]]
 name = "block-buffer"
-version = "0.10.3"
+version = "0.10.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "69cce20737498f97b993470a6e536b8523f0af7892a4f928cceb1ac5e52ebe7e"
+checksum = "3078c7629b62d3f0439517fa394996acacc5cbc91c5a20d8c658e77abd503a71"
 dependencies = [
  "generic-array",
 ]
@@ -76,15 +76,15 @@ dependencies = [
 
 [[package]]
 name = "core-foundation-sys"
-version = "0.8.3"
+version = "0.8.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5827cebf4670468b8772dd191856768aedcb1b0278a04f989f7766351917b9dc"
+checksum = "e496a50fda8aacccc86d7529e2c1e0892dbd0f898a6b5645b5561b89c3210efa"
 
 [[package]]
 name = "cpufeatures"
-version = "0.2.5"
+version = "0.2.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "28d997bd5e24a5928dd43e46dc529867e207907fe0b239c3477d924f7f2ca320"
+checksum = "3e4c1eaa2012c47becbbad2ab175484c2a84d1185b566fb2cc5b8707343dfe58"
 dependencies = [
  "libc",
 ]
@@ -112,13 +112,13 @@ dependencies = [
 
 [[package]]
 name = "errno"
-version = "0.2.8"
+version = "0.3.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f639046355ee4f37944e44f60642c6f3a7efa3cf6b78c78a0d989a8ce6c396a1"
+checksum = "4bcfec3a70f97c962c307b2d2c56e358cf1d00b558d74262b5f929ee8cc7e73a"
 dependencies = [
  "errno-dragonfly",
  "libc",
- "winapi",
+ "windows-sys 0.48.0",
 ]
 
 [[package]]
@@ -163,9 +163,9 @@ checksum = "00b0228411908ca8685dba7fc2cdd70ec9990a6e753e89b6ac91a84c40fbaf4b"
 
 [[package]]
 name = "futures"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "13e2792b0ff0340399d58445b88fd9770e3489eff258a4cbc1523418f12abf84"
+checksum = "23342abe12aba583913b2e62f22225ff9c950774065e4bfb61a19cd9770fec40"
 dependencies = [
  "futures-channel",
  "futures-core",
@@ -178,9 +178,9 @@ dependencies = [
 
 [[package]]
 name = "futures-channel"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2e5317663a9089767a1ec00a487df42e0ca174b61b4483213ac24448e4664df5"
+checksum = "955518d47e09b25bbebc7a18df10b81f0c766eaf4c4f1cccef2fca5f2a4fb5f2"
 dependencies = [
  "futures-core",
  "futures-sink",
@@ -188,15 +188,15 @@ dependencies = [
 
 [[package]]
 name = "futures-core"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ec90ff4d0fe1f57d600049061dc6bb68ed03c7d2fbd697274c41805dcb3f8608"
+checksum = "4bca583b7e26f571124fe5b7561d49cb2868d79116cfa0eefce955557c6fee8c"
 
 [[package]]
 name = "futures-executor"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e8de0a35a6ab97ec8869e32a2473f4b1324459e14c29275d14b10cb1fd19b50e"
+checksum = "ccecee823288125bd88b4d7f565c9e58e41858e47ab72e8ea2d64e93624386e0"
 dependencies = [
  "futures-core",
  "futures-task",
@@ -205,15 +205,15 @@ dependencies = [
 
 [[package]]
 name = "futures-io"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bfb8371b6fb2aeb2d280374607aeabfc99d95c72edfe51692e42d3d7f0d08531"
+checksum = "4fff74096e71ed47f8e023204cfd0aa1289cd54ae5430a9523be060cdb849964"
 
 [[package]]
 name = "futures-macro"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "95a73af87da33b5acf53acfebdc339fe592ecf5357ac7c0a7734ab9d8c876a70"
+checksum = "89ca545a94061b6365f2c7355b4b32bd20df3ff95f02da9329b34ccc3bd6ee72"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -222,21 +222,21 @@ dependencies = [
 
 [[package]]
 name = "futures-sink"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f310820bb3e8cfd46c80db4d7fb8353e15dfff853a127158425f31e0be6c8364"
+checksum = "f43be4fe21a13b9781a69afa4985b0f6ee0e1afab2c6f454a8cf30e2b2237b6e"
 
 [[package]]
 name = "futures-task"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dcf79a1bf610b10f42aea489289c5a2c478a786509693b80cd39c44ccd936366"
+checksum = "76d3d132be6c0e6aa1534069c705a74a5997a356c0dc2f86a47765e5617c5b65"
 
 [[package]]
 name = "futures-util"
-version = "0.3.26"
+version = "0.3.28"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9c1d6de3acfef38d2be4b1f543f553131788603495be83da675e180c8d6b7bd1"
+checksum = "26b01e40b772d54cf6c6d721c1d1abd0647a0106a12ecaa1c186273392a69533"
 dependencies = [
  "futures-channel",
  "futures-core",
@@ -252,9 +252,9 @@ dependencies = [
 
 [[package]]
 name = "generic-array"
-version = "0.14.6"
+version = "0.14.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bff49e947297f3312447abdca79f45f4738097cc82b06e72054d2223f601f1b9"
+checksum = "85649ca51fd72272d7821adaf274ad91c288277713d9c18820d8499a7ff69e9a"
 dependencies = [
  "typenum",
  "version_check",
@@ -262,15 +262,21 @@ dependencies = [
 
 [[package]]
 name = "getrandom"
-version = "0.2.8"
+version = "0.2.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c05aeb6a22b8f62540c194aac980f2115af067bfe15a0734d7277a768d396b31"
+checksum = "c85e1d9ab2eadba7e5040d4e09cbd6d072b76a557ad64e797c2cb9d4da21d7e4"
 dependencies = [
  "cfg-if",
  "libc",
  "wasi",
 ]
 
+[[package]]
+name = "hermit-abi"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fed44880c466736ef9a5c5b5facefb5ed0785676d0c02d612db14e54f0d84286"
+
 [[package]]
 name = "hmac"
 version = "0.12.1"
@@ -291,12 +297,13 @@ dependencies = [
 
 [[package]]
 name = "io-lifetimes"
-version = "1.0.5"
+version = "1.0.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1abeb7a0dd0f8181267ff8adc397075586500b81b28a73e8a0208b00fc170fb3"
+checksum = "9c66c74d2ae7e79a5a8f7ac924adbe38ee42a859c6539ad869eb51f0b52dc220"
 dependencies = [
+ "hermit-abi",
  "libc",
- "windows-sys 0.45.0",
+ "windows-sys 0.48.0",
 ]
 
 [[package]]
@@ -307,15 +314,15 @@ checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646"
 
 [[package]]
 name = "libc"
-version = "0.2.139"
+version = "0.2.142"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "201de327520df007757c1f0adce6e827fe8562fbc28bfd9c15571c66ca1f5f79"
+checksum = "6a987beff54b60ffa6d51982e1aa1146bc42f19bd26be28b0586f252fccf5317"
 
 [[package]]
 name = "linux-raw-sys"
-version = "0.1.4"
+version = "0.3.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f051f77a7c8e6957c0696eac88f26b0117e54f52d3fc682ab19397a8812846a4"
+checksum = "36eb31c1778188ae1e64398743890d0877fef36d11521ac60406b42016e8c2cf"
 
 [[package]]
 name = "lock_api"
@@ -389,9 +396,9 @@ checksum = "b7e5500299e16ebb147ae15a00a942af264cf3688f47923b8fc2cd5858f23ad3"
 
 [[package]]
 name = "openssl"
-version = "0.10.48"
+version = "0.10.52"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "518915b97df115dd36109bfa429a48b8f737bd05508cf9588977b599648926d2"
+checksum = "01b8574602df80f7b85fdfc5392fa884a4e3b3f4f35402c070ab34c3d3f78d56"
 dependencies = [
  "bitflags",
  "cfg-if",
@@ -404,9 +411,9 @@ dependencies = [
 
 [[package]]
 name = "openssl-macros"
-version = "0.1.0"
+version = "0.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b501e44f11665960c7e7fcf062c7d96a14ade4aa98116c004b2e37b5be7d736c"
+checksum = "a948666b637a0f465e8564c73e89d4dde00d72d4d473cc972f390fc3dcee7d9c"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -421,11 +428,10 @@ checksum = "ff011a302c396a5197692431fc1948019154afc178baf7d8e37367442a4601cf"
 
 [[package]]
 name = "openssl-sys"
-version = "0.9.83"
+version = "0.9.87"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "666416d899cf077260dac8698d60a60b435a46d57e82acb1be3d0dad87284e5b"
+checksum = "8e17f59264b2809d77ae94f0e1ebabc434773f370d6ca667bd223ea10e06cc7e"
 dependencies = [
- "autocfg",
  "cc",
  "libc",
  "pkg-config",
@@ -450,7 +456,7 @@ checksum = "9069cbb9f99e3a5083476ccb29ceb1de18b9118cafa53e90c9551235de2b9521"
 dependencies = [
  "cfg-if",
  "libc",
- "redox_syscall",
+ "redox_syscall 0.2.16",
  "smallvec",
  "windows-sys 0.45.0",
 ]
@@ -512,9 +518,9 @@ dependencies = [
 
 [[package]]
 name = "postgres-protocol"
-version = "0.6.4"
+version = "0.6.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "878c6cbf956e03af9aa8204b407b9cbf47c072164800aa918c516cd4b056c50c"
+checksum = "78b7fa9f396f51dffd61546fd8573ee20592287996568e6175ceb0f8699ad75d"
 dependencies = [
  "base64",
  "byteorder",
@@ -530,9 +536,9 @@ dependencies = [
 
 [[package]]
 name = "postgres-types"
-version = "0.2.4"
+version = "0.2.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "73d946ec7d256b04dfadc4e6a3292324e6f417124750fc5c0950f981b703a0f1"
+checksum = "f028f05971fe20f512bcc679e2c10227e57809a3af86a7606304435bc8896cd6"
 dependencies = [
  "bytes",
  "fallible-iterator",
@@ -547,18 +553,18 @@ checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"
 
 [[package]]
 name = "proc-macro2"
-version = "1.0.51"
+version = "1.0.56"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5d727cae5b39d21da60fa540906919ad737832fe0b1c165da3a34d6548c849d6"
+checksum = "2b63bdb0cd06f1f4dedf69b254734f9b45af66e4a031e42a7480257d9898b435"
 dependencies = [
  "unicode-ident",
 ]
 
 [[package]]
 name = "quote"
-version = "1.0.23"
+version = "1.0.26"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8856d8364d252a14d474036ea1358d63c9e6965c8e5c1885c18f73d70bff9c7b"
+checksum = "4424af4bf778aae2051a77b60283332f386554255d722233d09fbfc7e30da2fc"
 dependencies = [
  "proc-macro2",
 ]
@@ -602,6 +608,15 @@ dependencies = [
  "bitflags",
 ]
 
+[[package]]
+name = "redox_syscall"
+version = "0.3.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "567664f262709473930a4bf9e51bf2ebf3348f2e748ccc50dea20646858f8f29"
+dependencies = [
+ "bitflags",
+]
+
 [[package]]
 name = "rust-neon-example"
 version = "0.1.0"
@@ -614,16 +629,16 @@ dependencies = [
 
 [[package]]
 name = "rustix"
-version = "0.36.9"
+version = "0.37.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fd5c6ff11fecd55b40746d1995a02f2eb375bf8c00d192d521ee09f42bef37bc"
+checksum = "a0661814f891c57c930a610266415528da53c4933e6dea5fb350cbfe048a9ece"
 dependencies = [
  "bitflags",
  "errno",
  "io-lifetimes",
  "libc",
  "linux-raw-sys",
- "windows-sys 0.45.0",
+ "windows-sys 0.48.0",
 ]
 
 [[package]]
@@ -706,6 +721,16 @@ dependencies = [
  "winapi",
 ]
 
+[[package]]
+name = "socket2"
+version = "0.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6d283f86695ae989d1e18440a943880967156325ba025f05049946bff47bcc2b"
+dependencies = [
+ "libc",
+ "windows-sys 0.48.0",
+]
+
 [[package]]
 name = "stringprep"
 version = "0.1.2"
@@ -724,9 +749,9 @@ checksum = "6bdef32e8150c2a081110b42772ffe7d7c9032b606bc226c8260fd97e0976601"
 
 [[package]]
 name = "syn"
-version = "1.0.109"
+version = "2.0.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "72b64191b275b66ffe2469e8af2c1cfe3bafa67b529ead792a6d0160888b4237"
+checksum = "a34fcf3e8b60f57e6a14301a2e916d323af98b0ea63c599441eec8558660c822"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -735,15 +760,15 @@ dependencies = [
 
 [[package]]
 name = "tempfile"
-version = "3.4.0"
+version = "3.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "af18f7ae1acd354b992402e9ec5864359d693cd8a79dcbef59f76891701c1e95"
+checksum = "b9fbec84f381d5795b08656e4912bec604d162bff9291d6189a78f4c8ab87998"
 dependencies = [
  "cfg-if",
  "fastrand",
- "redox_syscall",
+ "redox_syscall 0.3.5",
  "rustix",
- "windows-sys 0.42.0",
+ "windows-sys 0.45.0",
 ]
 
 [[package]]
@@ -763,26 +788,25 @@ checksum = "1f3ccbac311fea05f86f61904b462b55fb3df8837a366dfc601a0161d0532f20"
 
 [[package]]
 name = "tokio"
-version = "1.26.0"
+version = "1.28.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "03201d01c3c27a29c8a5cee5b55a93ddae1ccf6f08f65365c2c918f8c1b76f64"
+checksum = "c3c786bf8134e5a3a166db9b29ab8f48134739014a3eca7bc6bfa95d673b136f"
 dependencies = [
  "autocfg",
  "bytes",
  "libc",
- "memchr",
  "mio",
  "pin-project-lite",
- "socket2",
+ "socket2 0.4.9",
  "tokio-macros",
- "windows-sys 0.45.0",
+ "windows-sys 0.48.0",
 ]
 
 [[package]]
 name = "tokio-macros"
-version = "1.8.2"
+version = "2.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d266c00fde287f55d3f1c3e96c500c362a2b8c695076ec180f27918820bc6df8"
+checksum = "630bdcf245f78637c13ec01ffae6187cca34625e8c63150d424b59e55af2675e"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -801,9 +825,9 @@ dependencies = [
 
 [[package]]
 name = "tokio-postgres"
-version = "0.7.7"
+version = "0.7.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "29a12c1b3e0704ae7dfc25562629798b29c72e6b1d0a681b6f29ab4ae5e7f7bf"
+checksum = "6e89f6234aa8fd43779746012fcf53603cdb91fdd8399aa0de868c2d56b6dde1"
 dependencies = [
  "async-trait",
  "byteorder",
@@ -818,16 +842,16 @@ dependencies = [
  "pin-project-lite",
  "postgres-protocol",
  "postgres-types",
- "socket2",
+ "socket2 0.5.2",
  "tokio",
  "tokio-util",
 ]
 
 [[package]]
 name = "tokio-util"
-version = "0.7.7"
+version = "0.7.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5427d89453009325de0d8f342c9490009f76e999cb7672d77e46267448f7e6b2"
+checksum = "806fe8c2c87eccc8b3267cbae29ed3ab2d0bd37fca70ab622e46aaa9375ddb7d"
 dependencies = [
  "bytes",
  "futures-core",
@@ -839,11 +863,10 @@ dependencies = [
 
 [[package]]
 name = "tracing"
-version = "0.1.37"
+version = "0.1.38"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8ce8c33a8d48bd45d624a6e523445fd21ec13d3653cd51f681abf67418f54eb8"
+checksum = "cf9cf6a813d3f40c88b0b6b6f29a5c95c6cdbf97c1f9cc53fb820200f5ad814d"
 dependencies = [
- "cfg-if",
  "pin-project-lite",
  "tracing-core",
 ]
@@ -865,15 +888,15 @@ checksum = "497961ef93d974e23eb6f433eb5fe1b7930b659f06d12dec6fc44a8f554c0bba"
 
 [[package]]
 name = "unicode-bidi"
-version = "0.3.10"
+version = "0.3.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d54675592c1dbefd78cbd98db9bacd89886e1ca50692a0692baefffdeb92dd58"
+checksum = "92888ba5573ff080736b3648696b70cafad7d250551175acbaa4e0385b3e1460"
 
 [[package]]
 name = "unicode-ident"
-version = "1.0.7"
+version = "1.0.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "775c11906edafc97bc378816b94585fbd9a054eabaf86fdd0ced94af449efab7"
+checksum = "e5464a87b239f13a63a501f2701565754bae92d243d4bb7eb12f6d57d2269bf4"
 
 [[package]]
 name = "unicode-normalization"
@@ -930,13 +953,13 @@ version = "0.42.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5a3e1820f08b8513f676f7ab6c1f99ff312fb97b553d30ff4dd86f9f15728aa7"
 dependencies = [
- "windows_aarch64_gnullvm",
- "windows_aarch64_msvc",
- "windows_i686_gnu",
- "windows_i686_msvc",
- "windows_x86_64_gnu",
- "windows_x86_64_gnullvm",
- "windows_x86_64_msvc",
+ "windows_aarch64_gnullvm 0.42.2",
+ "windows_aarch64_msvc 0.42.2",
+ "windows_i686_gnu 0.42.2",
+ "windows_i686_msvc 0.42.2",
+ "windows_x86_64_gnu 0.42.2",
+ "windows_x86_64_gnullvm 0.42.2",
+ "windows_x86_64_msvc 0.42.2",
 ]
 
 [[package]]
@@ -945,62 +968,128 @@ version = "0.45.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "75283be5efb2831d37ea142365f009c02ec203cd29a3ebecbc093d52315b66d0"
 dependencies = [
- "windows-targets",
+ "windows-targets 0.42.2",
+]
+
+[[package]]
+name = "windows-sys"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "677d2418bec65e3338edb076e806bc1ec15693c5d0104683f2efe857f61056a9"
+dependencies = [
+ "windows-targets 0.48.0",
 ]
 
 [[package]]
 name = "windows-targets"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8e2522491fbfcd58cc84d47aeb2958948c4b8982e9a2d8a2a35bbaed431390e7"
+checksum = "8e5180c00cd44c9b1c88adb3693291f1cd93605ded80c250a75d472756b4d071"
 dependencies = [
- "windows_aarch64_gnullvm",
- "windows_aarch64_msvc",
- "windows_i686_gnu",
- "windows_i686_msvc",
- "windows_x86_64_gnu",
- "windows_x86_64_gnullvm",
- "windows_x86_64_msvc",
+ "windows_aarch64_gnullvm 0.42.2",
+ "windows_aarch64_msvc 0.42.2",
+ "windows_i686_gnu 0.42.2",
+ "windows_i686_msvc 0.42.2",
+ "windows_x86_64_gnu 0.42.2",
+ "windows_x86_64_gnullvm 0.42.2",
+ "windows_x86_64_msvc 0.42.2",
+]
+
+[[package]]
+name = "windows-targets"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7b1eb6f0cd7c80c79759c929114ef071b87354ce476d9d94271031c0497adfd5"
+dependencies = [
+ "windows_aarch64_gnullvm 0.48.0",
+ "windows_aarch64_msvc 0.48.0",
+ "windows_i686_gnu 0.48.0",
+ "windows_i686_msvc 0.48.0",
+ "windows_x86_64_gnu 0.48.0",
+ "windows_x86_64_gnullvm 0.48.0",
+ "windows_x86_64_msvc 0.48.0",
 ]
 
 [[package]]
 name = "windows_aarch64_gnullvm"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8c9864e83243fdec7fc9c5444389dcbbfd258f745e7853198f365e3c4968a608"
+checksum = "597a5118570b68bc08d8d59125332c54f1ba9d9adeedeef5b99b02ba2b0698f8"
+
+[[package]]
+name = "windows_aarch64_gnullvm"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "91ae572e1b79dba883e0d315474df7305d12f569b400fcf90581b06062f7e1bc"
 
 [[package]]
 name = "windows_aarch64_msvc"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4c8b1b673ffc16c47a9ff48570a9d85e25d265735c503681332589af6253c6c7"
+checksum = "e08e8864a60f06ef0d0ff4ba04124db8b0fb3be5776a5cd47641e942e58c4d43"
+
+[[package]]
+name = "windows_aarch64_msvc"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b2ef27e0d7bdfcfc7b868b317c1d32c641a6fe4629c171b8928c7b08d98d7cf3"
 
 [[package]]
 name = "windows_i686_gnu"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "de3887528ad530ba7bdbb1faa8275ec7a1155a45ffa57c37993960277145d640"
+checksum = "c61d927d8da41da96a81f029489353e68739737d3beca43145c8afec9a31a84f"
+
+[[package]]
+name = "windows_i686_gnu"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "622a1962a7db830d6fd0a69683c80a18fda201879f0f447f065a3b7467daa241"
 
 [[package]]
 name = "windows_i686_msvc"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bf4d1122317eddd6ff351aa852118a2418ad4214e6613a50e0191f7004372605"
+checksum = "44d840b6ec649f480a41c8d80f9c65108b92d89345dd94027bfe06ac444d1060"
+
+[[package]]
+name = "windows_i686_msvc"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4542c6e364ce21bf45d69fdd2a8e455fa38d316158cfd43b3ac1c5b1b19f8e00"
 
 [[package]]
 name = "windows_x86_64_gnu"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c1040f221285e17ebccbc2591ffdc2d44ee1f9186324dd3e84e99ac68d699c45"
+checksum = "8de912b8b8feb55c064867cf047dda097f92d51efad5b491dfb98f6bbb70cb36"
+
+[[package]]
+name = "windows_x86_64_gnu"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ca2b8a661f7628cbd23440e50b05d705db3686f894fc9580820623656af974b1"
 
 [[package]]
 name = "windows_x86_64_gnullvm"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "628bfdf232daa22b0d64fdb62b09fcc36bb01f05a3939e20ab73aaf9470d0463"
+checksum = "26d41b46a36d453748aedef1486d5c7a85db22e56aff34643984ea85514e94a3"
+
+[[package]]
+name = "windows_x86_64_gnullvm"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7896dbc1f41e08872e9d5e8f8baa8fdd2677f29468c4e156210174edc7f7b953"
 
 [[package]]
 name = "windows_x86_64_msvc"
-version = "0.42.1"
+version = "0.42.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "447660ad36a13288b1db4d4248e857b510e8c3a225c822ba4fb748c0aafecffd"
+checksum = "9aec5da331524158c6d1a4ac0ab1541149c0b9505fde06423b02f5ef0106b9f0"
+
+[[package]]
+name = "windows_x86_64_msvc"
+version = "0.48.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1a515f5799fe4961cb532f983ce2b23082366b898e52ffbce459c86f67c8378a"

test_runner/pg_clients/rust/tokio-postgres/Cargo.toml

@@ -9,8 +9,8 @@ publish = false
 [dependencies]
 native-tls = "0.2.11"
 postgres-native-tls = "0.5.0"
-tokio = { version = "1.26", features=["rt", "macros"] }
-tokio-postgres = "0.7.7"
+tokio = { version = "1.28", features=["rt", "macros"] }
+tokio-postgres = "0.7.8"
 
 
 # This is not part of the main 'neon' workspace

test_runner/pg_clients/rust/tokio-postgres/Dockerfile

@@ -1,4 +1,4 @@
-FROM rust:1.67
+FROM rust:1.69
 WORKDIR /source
 
 COPY . .