Compare commits

..

31 Commits

Author SHA1 Message Date
Sasha Krassovsky
b16e4f689f Merge pull request #7869 from neondatabase/rc/2024-05-23
Metrics hotfix release
2024-05-23 14:05:30 -07:00
Sasha Krassovsky
dbff725a0c Remove apostrophe (#7868)
## Problem

## Summary of changes

## Checklist before requesting a review

- [ ] I have performed a self-review of my code.
- [ ] If it is a core feature, I have added thorough tests.
- [ ] Do we need to implement analytics? if so did you add the relevant
metrics to the dashboard?
- [ ] If this PR requires public announcement, mark it with
/release-notes label and add several sentences in this section.

## Checklist before merging

- [ ] Do not forget to reformat commit message to not include the above
checklist
2024-05-23 13:47:16 -07:00
Andreas Scherbaum
7fa4628434 Merge pull request #7837 from neondatabase/rc/2024-05-22
Compute-Only Release 2024-05-22
2024-05-22 19:34:39 +02:00
Arthur Petukhovsky
fc538a38b9 Merge pull request #7807 from neondatabase/rc/2024-05-20
Release 2024-05-20
2024-05-20 12:16:00 +01:00
Vlad Lazar
c2e7cb324f Merge pull request #7735 from neondatabase/vlad/release-2024-05-13
Handmade Release 2024-05-13
2024-05-13 16:27:38 +01:00
Vlad Lazar
101043122e Revert protocol version upgrade (#7727)
## Problem

"John pointed out that the switch to protocol version 2 made
test_gc_aggressive test flaky:
https://github.com/neondatabase/neon/issues/7692.
I tracked it down, and that is indeed an issue. Conditions for hitting
the issue:
The problem occurs in the primary
GC horizon is set to a very low value, e.g. 0.
If the primary is actively writing WAL, and GC runs in the pageserver at
the same time that the primary sends a GetPage request, it's possible
that the GC advances the GC horizon past the GetPage request's LSN. I'm
working on a fix here: https://github.com/neondatabase/neon/pull/7708."
- Heikki

## Summary of changes
Use protocol version 1 as default.
2024-05-13 14:17:36 +01:00
Christian Schwarz
c4d7d59825 Merge pull request #7615 from neondatabase/rc/2024-05-06
Release 2024-05-06
2024-05-07 09:41:02 +02:00
Arpad Müller
0de1e1d664 Merge pull request #7530 from neondatabase/rc/2024-04-29
Release 2024-04-29
2024-04-29 15:09:58 +02:00
Joonas Koivunen
271598b77f Merge pull request #7447 from neondatabase/rc/2024-04-22
Release 2024-04-22
2024-04-22 16:10:03 +03:00
John Spray
459bc479dc pageserver: fix unlogged relations with sharding (#7454)
## Problem

- #7451 

INIT_FORKNUM blocks must be stored on shard 0 to enable including them
in basebackup.

This issue can be missed in simple tests because creating an unlogged
table isn't sufficient -- to repro I had to create an _index_ on an
unlogged table (then restart the endpoint).

Closes: #7451 

## Summary of changes

- Add a reproducer for the issue.
- Tweak the condition for `key_is_shard0` to include anything that isn't
a normal relation block _and_ any normal relation block whose forknum is
INIT_FORKNUM.
- To enable existing databases to recover from the issue, add a special
case that omits relations if they were stored on the wrong INITFORK.
This enables postgres to start and the user to drop the table and
recreate it.
2024-04-22 11:55:24 +00:00
Christian Schwarz
c213373a59 Merge pull request #7378 from neondatabase/rc/2024-04-15
Release 2024-04-15
2024-04-15 15:48:14 +03:00
Em Sharnoff
e0addc100d Merge pull request #7356 from neondatabase/rc/2024-04-11-#7348
Release 2024-04-11 (cherry-pick #7348 only)

See here for more: https://neondb.slack.com/archives/C04DGM6SMTM/p1712776981582679
2024-04-11 09:46:34 -07:00
Em Sharnoff
0519138b04 compute_ctl: Auto-set dynamic_shared_memory_type (#7348)
Part of neondatabase/cloud#12047.

The basic idea is that for our VMs, we want to enable swap and disable
Linux memory overcommit. Alongside these, we should set postgres'
dynamic_shared_memory_type to mmap, but we want to avoid setting it to
mmap if swap is not enabled.

Implementing this in the control plane would be fiddly, but it's
relatively straightforward to add to compute_ctl.
2024-04-10 13:13:08 -07:00
Vlad Lazar
5da39b469c Merge pull request #7338 from neondatabase/rc/2024-04-08
Release 2024-04-08
2024-04-08 13:10:24 +01:00
Arseny Sher
82027e22dd Merge pull request #7284 from neondatabase/rc/2024-04-01
Release 2024-04-01
2024-04-02 18:15:28 +03:00
Alex Chi Z
c431e2f1c5 Merge pull request #7263 from neondatabase/rc/2024-03-27
Release 2024-03-27 - compute only release
2024-03-27 14:52:38 -04:00
John Spray
4e5724d9c3 Merge pull request #7248 from neondatabase/rc/2024-03-26
Release 2024-03-26
2024-03-26 15:17:00 +00:00
John Spray
0d3e499059 Merge pull request #7219 from neondatabase/rc/2024-03-25
Release 2024-03-25
2024-03-25 12:28:09 +00:00
Arpad Müller
7b860b837c Merge pull request #7154 from neondatabase/rc/2024-03-18
Release 2024-03-18
2024-03-19 12:07:14 +01:00
Christian Schwarz
41fc96e20f fixup(#7160 / tokio_epoll_uring_ext): double-panic caused by info! in thread-local's drop() (#7164)
Manual testing of the changes in #7160 revealed that, if the
thread-local destructor ever runs (it apparently doesn't in our test
suite runs, otherwise #7160 would not have auto-merged), we can
encounter an `abort()` due to a double-panic in the tracing code.

This github comment here contains the stack trace:
https://github.com/neondatabase/neon/pull/7160#issuecomment-2003778176

This PR reverts #7160 and uses a atomic counter to identify the
thread-local in log messages, instead of the memory address of the
thread local, which may be re-used.
2024-03-18 16:28:17 +01:00
Christian Schwarz
fb2b1ce57b fixup(#7141 / tokio_epoll_uring_ext): high frequency log message
The PR #7141 added log message

```
ThreadLocalState is being dropped and id might be re-used in the future
```

which was supposed to be emitted when the thread-local is destroyed.
Instead, it was emitted on _each_ call to `thread_local_system()`,
ie.., on each tokio-epoll-uring operation.
2024-03-18 13:01:17 +01:00
Joonas Koivunen
464717451b build: make procfs linux only dependency (#7156)
the dependency refuses to build on macos so builds on `main` are broken
right now, including the `release` PR.
2024-03-18 09:32:49 +00:00
Joonas Koivunen
c6ed86d3d0 Merge pull request #7081 from neondatabase/rc/2024-03-11
Release 2024-03-11
2024-03-11 14:41:39 +02:00
Roman Zaynetdinov
f0a9017008 Export db size, deadlocks and changed row metrics (#7050)
## Problem

We want to report metrics for the oldest user database.
2024-03-11 11:55:06 +00:00
Christian Schwarz
bb7949ba00 Merge pull request #6993 from neondatabase/rc/2024-03-04
Release 2024-03-04
2024-03-04 13:08:44 +01:00
Arthur Petukhovsky
1df0f69664 Merge pull request #6973 from neondatabase/rc/2024-02-29-manual
Release 2024-02-29
2024-02-29 17:26:33 +00:00
Vlad Lazar
970066a914 libs: fix expired token in auth decode test (#6963)
The test token expired earlier today (1709200879). I regenerated the
token, but without an expiration date this time.
2024-02-29 17:23:25 +00:00
Arthur Petukhovsky
1ebd3897c0 Merge pull request #6956 from neondatabase/rc/2024-02-28
Release 2024-02-28
2024-02-29 16:39:52 +00:00
Arthur Petukhovsky
6460beffcd Merge pull request #6901 from neondatabase/rc/2024-02-26
Release 2024-02-26
2024-02-26 17:08:19 +00:00
John Spray
6f7f8958db pageserver: only write out legacy tenant config if no generation (#6891)
## Problem

Previously we always wrote out both legacy and modern tenant config
files. The legacy write enabled rollbacks, but we are long past the
point where that is needed.

We still need the legacy format for situations where someone is running
tenants without generations (that will be yanked as well eventually),
but we can avoid writing it out at all if we do have a generation number
set. We implicitly also avoid writing the legacy config if our mode is
Secondary (secondary mode is newer than generations).

## Summary of changes

- Make writing legacy tenant config conditional on there being no
generation number set.
2024-02-26 10:25:25 +00:00
Christian Schwarz
936a00e077 pageserver: remove two obsolete/unused per-timeline metrics (#6893)
over-compensating the addition of a new per-timeline metric in
https://github.com/neondatabase/neon/pull/6834

part of https://github.com/neondatabase/neon/issues/6737
2024-02-26 09:16:24 +00:00
351 changed files with 9291 additions and 19495 deletions

View File

@@ -8,7 +8,6 @@
!scripts/combine_control_files.py
!scripts/ninstall.sh
!vm-cgconfig.conf
!docker-compose/run-tests.sh
# Directories
!.cargo/
@@ -21,7 +20,7 @@
!patches/
!pgxn/
!proxy/
!storage_scrubber/
!s3_scrubber/
!safekeeper/
!storage_broker/
!storage_controller/

View File

@@ -26,7 +26,7 @@ runs:
TARGET: ${{ inputs.path }}
ARCHIVE: /tmp/downloads/${{ inputs.name }}.tar.zst
SKIP_IF_DOES_NOT_EXIST: ${{ inputs.skip-if-does-not-exist }}
PREFIX: artifacts/${{ inputs.prefix || format('{0}/{1}/{2}', github.event.pull_request.head.sha || github.sha, github.run_id, github.run_attempt) }}
PREFIX: artifacts/${{ inputs.prefix || format('{0}/{1}', github.run_id, github.run_attempt) }}
run: |
BUCKET=neon-github-public-dev
FILENAME=$(basename $ARCHIVE)

View File

@@ -3,13 +3,13 @@ description: 'Create Branch using API'
inputs:
api_key:
description: 'Neon API key'
desctiption: 'Neon API key'
required: true
project_id:
description: 'ID of the Project to create Branch in'
desctiption: 'ID of the Project to create Branch in'
required: true
api_host:
description: 'Neon API host'
desctiption: 'Neon API host'
default: console-stage.neon.build
outputs:
dsn:

View File

@@ -3,16 +3,16 @@ description: 'Delete Branch using API'
inputs:
api_key:
description: 'Neon API key'
desctiption: 'Neon API key'
required: true
project_id:
description: 'ID of the Project which should be deleted'
desctiption: 'ID of the Project which should be deleted'
required: true
branch_id:
description: 'ID of the branch to delete'
desctiption: 'ID of the branch to delete'
required: true
api_host:
description: 'Neon API host'
desctiption: 'Neon API host'
default: console-stage.neon.build
runs:

View File

@@ -3,22 +3,22 @@ description: 'Create Neon Project using API'
inputs:
api_key:
description: 'Neon API key'
desctiption: 'Neon API key'
required: true
region_id:
description: 'Region ID, if not set the project will be created in the default region'
desctiption: 'Region ID, if not set the project will be created in the default region'
default: aws-us-east-2
postgres_version:
description: 'Postgres version; default is 15'
default: '15'
desctiption: 'Postgres version; default is 15'
default: 15
api_host:
description: 'Neon API host'
desctiption: 'Neon API host'
default: console-stage.neon.build
provisioner:
description: 'k8s-pod or k8s-neonvm'
desctiption: 'k8s-pod or k8s-neonvm'
default: 'k8s-pod'
compute_units:
description: '[Min, Max] compute units; Min and Max are used for k8s-neonvm with autoscaling, for k8s-pod values Min and Max should be equal'
desctiption: '[Min, Max] compute units; Min and Max are used for k8s-neonvm with autoscaling, for k8s-pod values Min and Max should be equal'
default: '[1, 1]'
outputs:

View File

@@ -3,13 +3,13 @@ description: 'Delete Neon Project using API'
inputs:
api_key:
description: 'Neon API key'
desctiption: 'Neon API key'
required: true
project_id:
description: 'ID of the Project to delete'
desctiption: 'ID of the Project to delete'
required: true
api_host:
description: 'Neon API host'
desctiption: 'Neon API host'
default: console-stage.neon.build
runs:

View File

@@ -183,7 +183,8 @@ runs:
# Run the tests.
#
# --alluredir saves test results in Allure format (in a specified directory)
# The junit.xml file allows CI tools to display more fine-grained test information
# in its "Tests" tab in the results page.
# --verbose prints name of each test (helpful when there are
# multiple tests in one file)
# -rA prints summary in the end
@@ -192,6 +193,7 @@ runs:
#
mkdir -p $TEST_OUTPUT/allure/results
"${cov_prefix[@]}" ./scripts/pytest \
--junitxml=$TEST_OUTPUT/junit.xml \
--alluredir=$TEST_OUTPUT/allure/results \
--tb=short \
--verbose \

View File

@@ -8,7 +8,7 @@ inputs:
description: "A directory or file to upload"
required: true
prefix:
description: "S3 prefix. Default is '${GITHUB_SHA}/${GITHUB_RUN_ID}/${GITHUB_RUN_ATTEMPT}'"
description: "S3 prefix. Default is '${GITHUB_RUN_ID}/${GITHUB_RUN_ATTEMPT}'"
required: false
runs:
@@ -45,7 +45,7 @@ runs:
env:
SOURCE: ${{ inputs.path }}
ARCHIVE: /tmp/uploads/${{ inputs.name }}.tar.zst
PREFIX: artifacts/${{ inputs.prefix || format('{0}/{1}/{2}', github.event.pull_request.head.sha || github.sha, github.run_id , github.run_attempt) }}
PREFIX: artifacts/${{ inputs.prefix || format('{0}/{1}', github.run_id, github.run_attempt) }}
run: |
BUCKET=neon-github-public-dev
FILENAME=$(basename $ARCHIVE)

View File

@@ -24,7 +24,7 @@ jobs:
actionlint:
needs: [ check-permissions ]
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: reviewdog/action-actionlint@v1
@@ -36,16 +36,3 @@ jobs:
fail_on_error: true
filter_mode: nofilter
level: error
- name: Disallow 'ubuntu-latest' runners
run: |
PAT='^\s*runs-on:.*-latest'
if grep -ERq $PAT .github/workflows; then
grep -ERl $PAT .github/workflows |\
while read -r f
do
l=$(grep -nE $PAT .github/workflows/release.yml | awk -F: '{print $1}' | head -1)
echo "::error file=$f,line=$l::Please use 'ubuntu-22.04' instead of 'ubuntu-latest'"
done
exit 1
fi

View File

@@ -44,7 +44,7 @@ jobs:
contains(fromJSON('["opened", "synchronize", "reopened", "closed"]'), github.event.action) &&
contains(github.event.pull_request.labels.*.name, 'approved-for-ci-run')
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- run: gh pr --repo "${GITHUB_REPOSITORY}" edit "${PR_NUMBER}" --remove-label "approved-for-ci-run"
@@ -60,7 +60,7 @@ jobs:
github.event.action == 'labeled' &&
contains(github.event.pull_request.labels.*.name, 'approved-for-ci-run')
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- run: gh pr --repo "${GITHUB_REPOSITORY}" edit "${PR_NUMBER}" --remove-label "approved-for-ci-run"
@@ -69,41 +69,15 @@ jobs:
with:
ref: main
token: ${{ secrets.CI_ACCESS_TOKEN }}
- name: Look for existing PR
id: get-pr
env:
GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
run: |
ALREADY_CREATED="$(gh pr --repo ${GITHUB_REPOSITORY} list --head ${BRANCH} --base main --json number --jq '.[].number')"
echo "ALREADY_CREATED=${ALREADY_CREATED}" >> ${GITHUB_OUTPUT}
- name: Get changed labels
id: get-labels
if: steps.get-pr.outputs.ALREADY_CREATED != ''
env:
ALREADY_CREATED: ${{ steps.get-pr.outputs.ALREADY_CREATED }}
GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
run: |
LABELS_TO_REMOVE=$(comm -23 <(gh pr --repo ${GITHUB_REPOSITORY} view ${ALREADY_CREATED} --json labels --jq '.labels.[].name'| ( grep -E '^run' || true ) | sort) \
<(gh pr --repo ${GITHUB_REPOSITORY} view ${PR_NUMBER} --json labels --jq '.labels.[].name' | ( grep -E '^run' || true ) | sort ) |\
( grep -v run-e2e-tests-in-draft || true ) | paste -sd , -)
LABELS_TO_ADD=$(comm -13 <(gh pr --repo ${GITHUB_REPOSITORY} view ${ALREADY_CREATED} --json labels --jq '.labels.[].name'| ( grep -E '^run' || true ) |sort) \
<(gh pr --repo ${GITHUB_REPOSITORY} view ${PR_NUMBER} --json labels --jq '.labels.[].name' | ( grep -E '^run' || true ) | sort ) |\
paste -sd , -)
echo "LABELS_TO_ADD=${LABELS_TO_ADD}" >> ${GITHUB_OUTPUT}
echo "LABELS_TO_REMOVE=${LABELS_TO_REMOVE}" >> ${GITHUB_OUTPUT}
- run: gh pr checkout "${PR_NUMBER}"
- run: git checkout -b "${BRANCH}"
- run: git push --force origin "${BRANCH}"
if: steps.get-pr.outputs.ALREADY_CREATED == ''
- name: Create a Pull Request for CI run (if required)
if: steps.get-pr.outputs.ALREADY_CREATED == ''
env:
env:
GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
run: |
cat << EOF > body.md
@@ -114,33 +88,16 @@ jobs:
Feel free to review/comment/discuss the original PR #${PR_NUMBER}.
EOF
LABELS=$( (gh pr --repo "${GITHUB_REPOSITORY}" view ${PR_NUMBER} --json labels --jq '.labels.[].name'; echo run-e2e-tests-in-draft )| \
grep -E '^run' | paste -sd , -)
gh pr --repo "${GITHUB_REPOSITORY}" create --title "CI run for PR #${PR_NUMBER}" \
ALREADY_CREATED="$(gh pr --repo ${GITHUB_REPOSITORY} list --head ${BRANCH} --base main --json number --jq '.[].number')"
if [ -z "${ALREADY_CREATED}" ]; then
gh pr --repo "${GITHUB_REPOSITORY}" create --title "CI run for PR #${PR_NUMBER}" \
--body-file "body.md" \
--head "${BRANCH}" \
--base "main" \
--label ${LABELS} \
--label "run-e2e-tests-in-draft" \
--draft
- name: Modify the existing pull request (if required)
if: steps.get-pr.outputs.ALREADY_CREATED != ''
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
LABELS_TO_ADD: ${{ steps.get-labels.outputs.LABELS_TO_ADD }}
LABELS_TO_REMOVE: ${{ steps.get-labels.outputs.LABELS_TO_REMOVE }}
ALREADY_CREATED: ${{ steps.get-pr.outputs.ALREADY_CREATED }}
run: |
ADD_CMD=
REMOVE_CMD=
[ -z "${LABELS_TO_ADD}" ] || ADD_CMD="--add-label ${LABELS_TO_ADD}"
[ -z "${LABELS_TO_REMOVE}" ] || REMOVE_CMD="--remove-label ${LABELS_TO_REMOVE}"
if [ -n "${ADD_CMD}" ] || [ -n "${REMOVE_CMD}" ]; then
gh pr --repo "${GITHUB_REPOSITORY}" edit ${ALREADY_CREATED} ${ADD_CMD} ${REMOVE_CMD}
fi
- run: git push --force origin "${BRANCH}"
if: steps.get-pr.outputs.ALREADY_CREATED != ''
cleanup:
# Close PRs and delete branchs if the original PR is closed.
@@ -152,7 +109,7 @@ jobs:
github.event.action == 'closed' &&
github.event.pull_request.head.repo.full_name != github.repository
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- name: Close PR and delete `ci-run/pr-${{ env.PR_NUMBER }}` branch

View File

@@ -38,11 +38,6 @@ on:
description: 'AWS-RDS and AWS-AURORA normally only run on Saturday. Set this to true to run them on every workflow_dispatch'
required: false
default: false
run_only_pgvector_tests:
type: boolean
description: 'Run pgvector tests but no other tests. If not set, all tests including pgvector tests will be run'
required: false
default: false
defaults:
run:
@@ -55,7 +50,6 @@ concurrency:
jobs:
bench:
if: ${{ github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null }}
env:
TEST_PG_BENCH_DURATIONS_MATRIX: "300"
TEST_PG_BENCH_SCALES_MATRIX: "10,100"
@@ -99,7 +93,7 @@ jobs:
# Set --sparse-ordering option of pytest-order plugin
# to ensure tests are running in order of appears in the file.
# It's important for test_perf_pgbench.py::test_pgbench_remote_* tests
extra_params: -m remote_cluster --sparse-ordering --timeout 5400 --ignore test_runner/performance/test_perf_olap.py --ignore test_runner/performance/test_perf_pgvector_queries.py
extra_params: -m remote_cluster --sparse-ordering --timeout 5400 --ignore test_runner/performance/test_perf_olap.py
env:
BENCHMARK_CONNSTR: ${{ steps.create-neon-project.outputs.dsn }}
VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -126,7 +120,6 @@ jobs:
SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
generate-matrices:
if: ${{ github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null }}
# Create matrices for the benchmarking jobs, so we run benchmarks on rds only once a week (on Saturday)
#
# Available platforms:
@@ -137,7 +130,7 @@ jobs:
# - rds-postgres: RDS Postgres db.m5.large instance (2 vCPU, 8 GiB) with gp3 EBS storage
env:
RUN_AWS_RDS_AND_AURORA: ${{ github.event.inputs.run_AWS_RDS_AND_AURORA || 'false' }}
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
outputs:
pgbench-compare-matrix: ${{ steps.pgbench-compare-matrix.outputs.matrix }}
olap-compare-matrix: ${{ steps.olap-compare-matrix.outputs.matrix }}
@@ -204,7 +197,6 @@ jobs:
echo "matrix=$(echo "$matrix" | jq --compact-output '.')" >> $GITHUB_OUTPUT
pgbench-compare:
if: ${{ github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null }}
needs: [ generate-matrices ]
strategy:
@@ -351,92 +343,6 @@ jobs:
env:
SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
pgbench-pgvector:
env:
TEST_PG_BENCH_DURATIONS_MATRIX: "15m"
TEST_PG_BENCH_SCALES_MATRIX: "1"
POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
DEFAULT_PG_VERSION: 16
TEST_OUTPUT: /tmp/test_output
BUILD_TYPE: remote
SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
PLATFORM: "neon-captest-pgvector"
runs-on: [ self-hosted, us-east-2, x64 ]
container:
image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned
options: --init
steps:
- uses: actions/checkout@v4
- name: Download Neon artifact
uses: ./.github/actions/download
with:
name: neon-${{ runner.os }}-release-artifact
path: /tmp/neon/
prefix: latest
- name: Add Postgres binaries to PATH
run: |
${POSTGRES_DISTRIB_DIR}/v${DEFAULT_PG_VERSION}/bin/pgbench --version
echo "${POSTGRES_DISTRIB_DIR}/v${DEFAULT_PG_VERSION}/bin" >> $GITHUB_PATH
- name: Set up Connection String
id: set-up-connstr
run: |
CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR }}
echo "connstr=${CONNSTR}" >> $GITHUB_OUTPUT
QUERIES=("SELECT version()")
QUERIES+=("SHOW neon.tenant_id")
QUERIES+=("SHOW neon.timeline_id")
for q in "${QUERIES[@]}"; do
psql ${CONNSTR} -c "${q}"
done
- name: Benchmark pgvector hnsw indexing
uses: ./.github/actions/run-python-test-set
with:
build_type: ${{ env.BUILD_TYPE }}
test_selection: performance/test_perf_olap.py
run_in_parallel: false
save_perf_report: ${{ env.SAVE_PERF_REPORT }}
extra_params: -m remote_cluster --timeout 21600 -k test_pgvector_indexing
env:
VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
- name: Benchmark pgvector queries
uses: ./.github/actions/run-python-test-set
with:
build_type: ${{ env.BUILD_TYPE }}
test_selection: performance/test_perf_pgvector_queries.py
run_in_parallel: false
save_perf_report: ${{ env.SAVE_PERF_REPORT }}
extra_params: -m remote_cluster --timeout 21600
env:
BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
- name: Create Allure report
if: ${{ !cancelled() }}
uses: ./.github/actions/allure-report-generate
- name: Post to a Slack channel
if: ${{ github.event.schedule && failure() }}
uses: slackapi/slack-github-action@v1
with:
channel-id: "C033QLM5P7D" # dev-staging-stream
slack-message: "Periodic perf testing neon-captest-pgvector: ${{ job.status }}\n${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}"
env:
SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
clickbench-compare:
# ClichBench DB for rds-aurora and rds-Postgres deployed to the same clusters
# we use for performance testing in pgbench-compare.
@@ -445,7 +351,7 @@ jobs:
#
# *_CLICKBENCH_CONNSTR: Genuine ClickBench DB with ~100M rows
# *_CLICKBENCH_10M_CONNSTR: DB with the first 10M rows of ClickBench DB
if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
if: ${{ !cancelled() }}
needs: [ generate-matrices, pgbench-compare ]
strategy:
@@ -549,7 +455,7 @@ jobs:
# We might change it after https://github.com/neondatabase/neon/issues/2900.
#
# *_TPCH_S10_CONNSTR: DB generated with scale factor 10 (~10 GB)
if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
if: ${{ !cancelled() }}
needs: [ generate-matrices, clickbench-compare ]
strategy:
@@ -651,7 +557,7 @@ jobs:
SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
user-examples-compare:
if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
if: ${{ !cancelled() }}
needs: [ generate-matrices, tpch-compare ]
strategy:

View File

@@ -30,6 +30,7 @@ jobs:
check-image:
uses: ./.github/workflows/check-build-tools-image.yml
# This job uses older version of GitHub Actions because it's run on gen2 runners, which don't support node 20 (for newer versions)
build-image:
needs: [ check-image ]
if: needs.check-image.outputs.found == 'false'
@@ -54,7 +55,7 @@ jobs:
exit 1
fi
- uses: actions/checkout@v4
- uses: actions/checkout@v3
# Use custom DOCKER_CONFIG directory to avoid conflicts with default settings
# The default value is ~/.docker
@@ -78,7 +79,7 @@ jobs:
pull: true
file: Dockerfile.build-tools
cache-from: type=registry,ref=neondatabase/build-tools:cache-${{ matrix.arch }}
cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/build-tools:cache-{0},mode=max', matrix.arch) || '' }}
cache-to: type=registry,ref=neondatabase/build-tools:cache-${{ matrix.arch }},mode=max
tags: neondatabase/build-tools:${{ inputs.image-tag }}-${{ matrix.arch }}
- name: Remove custom docker config directory
@@ -87,7 +88,7 @@ jobs:
merge-images:
needs: [ build-image ]
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
env:
IMAGE_TAG: ${{ inputs.image-tag }}

View File

@@ -35,7 +35,7 @@ jobs:
cancel-previous-e2e-tests:
needs: [ check-permissions ]
if: github.event_name == 'pull_request'
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- name: Cancel previous e2e-tests runs for this PR
@@ -299,21 +299,21 @@ jobs:
uses: actions/cache@v4
with:
path: pg_install/v14
key: v1-${{ runner.os }}-${{ matrix.build_type }}-pg-${{ steps.pg_v14_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
key: v1-${{ runner.os }}-${{ matrix.build_type }}-pg-${{ steps.pg_v14_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
- name: Cache postgres v15 build
id: cache_pg_15
uses: actions/cache@v4
with:
path: pg_install/v15
key: v1-${{ runner.os }}-${{ matrix.build_type }}-pg-${{ steps.pg_v15_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
key: v1-${{ runner.os }}-${{ matrix.build_type }}-pg-${{ steps.pg_v15_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
- name: Cache postgres v16 build
id: cache_pg_16
uses: actions/cache@v4
with:
path: pg_install/v16
key: v1-${{ runner.os }}-${{ matrix.build_type }}-pg-${{ steps.pg_v16_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
key: v1-${{ runner.os }}-${{ matrix.build_type }}-pg-${{ steps.pg_v16_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
- name: Build postgres v14
if: steps.cache_pg_14.outputs.cache-hit != 'true'
@@ -337,8 +337,34 @@ jobs:
run: |
${cov_prefix} mold -run cargo build $CARGO_FLAGS $CARGO_FEATURES --bins --tests
# Do install *before* running rust tests because they might recompile the
# binaries with different features/flags.
- name: Run rust tests
env:
NEXTEST_RETRIES: 3
run: |
#nextest does not yet support running doctests
cargo test --doc $CARGO_FLAGS $CARGO_FEATURES
for io_engine in std-fs tokio-epoll-uring ; do
NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES
done
# Run separate tests for real S3
export ENABLE_REAL_S3_REMOTE_STORAGE=nonempty
export REMOTE_STORAGE_S3_BUCKET=neon-github-ci-tests
export REMOTE_STORAGE_S3_REGION=eu-central-1
# Avoid `$CARGO_FEATURES` since there's no `testing` feature in the e2e tests now
${cov_prefix} cargo nextest run $CARGO_FLAGS -E 'package(remote_storage)' -E 'test(test_real_s3)'
# Run separate tests for real Azure Blob Storage
# XXX: replace region with `eu-central-1`-like region
export ENABLE_REAL_AZURE_REMOTE_STORAGE=y
export AZURE_STORAGE_ACCOUNT="${{ secrets.AZURE_STORAGE_ACCOUNT_DEV }}"
export AZURE_STORAGE_ACCESS_KEY="${{ secrets.AZURE_STORAGE_ACCESS_KEY_DEV }}"
export REMOTE_STORAGE_AZURE_CONTAINER="${{ vars.REMOTE_STORAGE_AZURE_CONTAINER }}"
export REMOTE_STORAGE_AZURE_REGION="${{ vars.REMOTE_STORAGE_AZURE_REGION }}"
# Avoid `$CARGO_FEATURES` since there's no `testing` feature in the e2e tests now
${cov_prefix} cargo nextest run $CARGO_FLAGS -E 'package(remote_storage)' -E 'test(test_real_azure)'
- name: Install rust binaries
run: |
# Install target binaries
@@ -379,32 +405,6 @@ jobs:
done
fi
- name: Run rust tests
env:
NEXTEST_RETRIES: 3
run: |
#nextest does not yet support running doctests
cargo test --doc $CARGO_FLAGS $CARGO_FEATURES
for io_engine in std-fs tokio-epoll-uring ; do
NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES
done
# Run separate tests for real S3
export ENABLE_REAL_S3_REMOTE_STORAGE=nonempty
export REMOTE_STORAGE_S3_BUCKET=neon-github-ci-tests
export REMOTE_STORAGE_S3_REGION=eu-central-1
${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E 'package(remote_storage)' -E 'test(test_real_s3)'
# Run separate tests for real Azure Blob Storage
# XXX: replace region with `eu-central-1`-like region
export ENABLE_REAL_AZURE_REMOTE_STORAGE=y
export AZURE_STORAGE_ACCOUNT="${{ secrets.AZURE_STORAGE_ACCOUNT_DEV }}"
export AZURE_STORAGE_ACCESS_KEY="${{ secrets.AZURE_STORAGE_ACCESS_KEY_DEV }}"
export REMOTE_STORAGE_AZURE_CONTAINER="${{ vars.REMOTE_STORAGE_AZURE_CONTAINER }}"
export REMOTE_STORAGE_AZURE_REGION="${{ vars.REMOTE_STORAGE_AZURE_REGION }}"
${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E 'package(remote_storage)' -E 'test(test_real_azure)'
- name: Install postgres binaries
run: cp -a pg_install /tmp/neon/pg_install
@@ -548,8 +548,8 @@ jobs:
report-benchmarks-failures:
needs: [ benchmarks, create-test-report ]
if: github.ref_name == 'main' && failure() && needs.benchmarks.result == 'failure'
runs-on: ubuntu-22.04
if: github.ref_name == 'main' && needs.benchmarks.result == 'failure'
runs-on: ubuntu-latest
steps:
- uses: slackapi/slack-github-action@v1
@@ -723,13 +723,9 @@ jobs:
uses: ./.github/workflows/trigger-e2e-tests.yml
secrets: inherit
neon-image-arch:
neon-image:
needs: [ check-permissions, build-build-tools-image, tag ]
strategy:
matrix:
arch: [ x64, arm64 ]
runs-on: ${{ fromJson(format('["self-hosted", "gen3", "{0}"]', matrix.arch == 'arm64' && 'large-arm64' || 'large')) }}
runs-on: [ self-hosted, gen3, large ]
steps:
- name: Checkout
@@ -751,6 +747,12 @@ jobs:
username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
- uses: docker/login-action@v3
with:
registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
password: ${{ secrets.AWS_SECRET_KEY_DEV }}
- uses: docker/build-push-action@v5
with:
context: .
@@ -762,52 +764,25 @@ jobs:
push: true
pull: true
file: Dockerfile
cache-from: type=registry,ref=neondatabase/neon:cache-${{ matrix.arch }}
cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/neon:cache-{0},mode=max', matrix.arch) || '' }}
cache-from: type=registry,ref=neondatabase/neon:cache
cache-to: type=registry,ref=neondatabase/neon:cache,mode=max
tags: |
neondatabase/neon:${{ needs.tag.outputs.build-tag }}-${{ matrix.arch }}
369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}}
neondatabase/neon:${{needs.tag.outputs.build-tag}}
- name: Remove custom docker config directory
if: always()
run: |
rm -rf .docker-custom
neon-image:
needs: [ neon-image-arch, tag ]
runs-on: ubuntu-22.04
steps:
- uses: docker/login-action@v3
with:
username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
- name: Create multi-arch image
run: |
docker buildx imagetools create -t neondatabase/neon:${{ needs.tag.outputs.build-tag }} \
neondatabase/neon:${{ needs.tag.outputs.build-tag }}-x64 \
neondatabase/neon:${{ needs.tag.outputs.build-tag }}-arm64
- uses: docker/login-action@v3
with:
registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
password: ${{ secrets.AWS_SECRET_KEY_DEV }}
- name: Push multi-arch image to ECR
run: |
docker buildx imagetools create -t 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{ needs.tag.outputs.build-tag }} \
neondatabase/neon:${{ needs.tag.outputs.build-tag }}
compute-node-image-arch:
compute-node-image:
needs: [ check-permissions, build-build-tools-image, tag ]
runs-on: [ self-hosted, gen3, large ]
strategy:
fail-fast: false
matrix:
version: [ v14, v15, v16 ]
arch: [ x64, arm64 ]
runs-on: ${{ fromJson(format('["self-hosted", "gen3", "{0}"]', matrix.arch == 'arm64' && 'large-arm64' || 'large')) }}
steps:
- name: Checkout
@@ -854,34 +829,15 @@ jobs:
push: true
pull: true
file: Dockerfile.compute-node
cache-from: type=registry,ref=neondatabase/compute-node-${{ matrix.version }}:cache-${{ matrix.arch }}
cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/compute-node-{0}:cache-{1},mode=max', matrix.version, matrix.arch) || '' }}
cache-from: type=registry,ref=neondatabase/compute-node-${{ matrix.version }}:cache
cache-to: type=registry,ref=neondatabase/compute-node-${{ matrix.version }}:cache,mode=max
tags: |
neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}-${{ matrix.arch }}
- name: Build neon extensions test image
if: matrix.version == 'v16'
uses: docker/build-push-action@v5
with:
context: .
build-args: |
GIT_VERSION=${{ github.event.pull_request.head.sha || github.sha }}
PG_VERSION=${{ matrix.version }}
BUILD_TAG=${{ needs.tag.outputs.build-tag }}
TAG=${{ needs.build-build-tools-image.outputs.image-tag }}
provenance: false
push: true
pull: true
file: Dockerfile.compute-node
target: neon-pg-ext-test
cache-from: type=registry,ref=neondatabase/neon-test-extensions-${{ matrix.version }}:cache-${{ matrix.arch }}
cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/neon-test-extensions-{0}:cache-{1},mode=max', matrix.version, matrix.arch) || '' }}
tags: |
neondatabase/neon-test-extensions-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}-${{ matrix.arch }}
369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
neondatabase/compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
- name: Build compute-tools image
# compute-tools are Postgres independent, so build it only once
if: matrix.version == 'v16'
if: ${{ matrix.version == 'v16' }}
uses: docker/build-push-action@v5
with:
target: compute-tools-image
@@ -895,64 +851,14 @@ jobs:
pull: true
file: Dockerfile.compute-node
tags: |
neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}-${{ matrix.arch }}
369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{ needs.tag.outputs.build-tag }}
neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}
- name: Remove custom docker config directory
if: always()
run: |
rm -rf .docker-custom
compute-node-image:
needs: [ compute-node-image-arch, tag ]
runs-on: ubuntu-22.04
strategy:
matrix:
version: [ v14, v15, v16 ]
steps:
- uses: docker/login-action@v3
with:
username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
- name: Create multi-arch compute-node image
run: |
docker buildx imagetools create -t neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }} \
neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}-x64 \
neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}-arm64
- name: Create multi-arch neon-test-extensions image
if: matrix.version == 'v16'
run: |
docker buildx imagetools create -t neondatabase/neon-test-extensions-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }} \
neondatabase/neon-test-extensions-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}-x64 \
neondatabase/neon-test-extensions-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}-arm64
- name: Create multi-arch compute-tools image
if: matrix.version == 'v16'
run: |
docker buildx imagetools create -t neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }} \
neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}-x64 \
neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}-arm64
- uses: docker/login-action@v3
with:
registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
password: ${{ secrets.AWS_SECRET_KEY_DEV }}
- name: Push multi-arch compute-node-${{ matrix.version }} image to ECR
run: |
docker buildx imagetools create -t 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }} \
neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}
- name: Push multi-arch compute-tools image to ECR
if: matrix.version == 'v16'
run: |
docker buildx imagetools create -t 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{ needs.tag.outputs.build-tag }} \
neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}
vm-compute-node-image:
needs: [ check-permissions, tag, compute-node-image ]
runs-on: [ self-hosted, gen3, large ]
@@ -960,12 +866,15 @@ jobs:
fail-fast: false
matrix:
version: [ v14, v15, v16 ]
defaults:
run:
shell: sh -eu {0}
env:
VM_BUILDER_VERSION: v0.29.3
VM_BUILDER_VERSION: v0.28.1
steps:
- name: Checkout
uses: actions/checkout@v4
uses: actions/checkout@v1
with:
fetch-depth: 0
@@ -974,48 +883,26 @@ jobs:
curl -fL https://github.com/neondatabase/autoscaling/releases/download/$VM_BUILDER_VERSION/vm-builder -o vm-builder
chmod +x vm-builder
# Use custom DOCKER_CONFIG directory to avoid conflicts with default settings
# The default value is ~/.docker
- name: Set custom docker config directory
run: |
mkdir -p .docker-custom
echo DOCKER_CONFIG=$(pwd)/.docker-custom >> $GITHUB_ENV
- uses: docker/login-action@v3
with:
username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
# Note: we need a separate pull step here because otherwise vm-builder will try to pull, and
# it won't have the proper authentication (written at v0.6.0)
- name: Pulling compute-node image
run: |
docker pull neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}
docker pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
- name: Build vm image
run: |
./vm-builder \
-spec=vm-image-spec.yaml \
-src=neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }} \
-dst=neondatabase/vm-compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}
-src=369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}} \
-dst=369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
- name: Pushing vm-compute-node image
run: |
docker push neondatabase/vm-compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}
- name: Remove custom docker config directory
if: always()
run: |
rm -rf .docker-custom
docker push 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}
test-images:
needs: [ check-permissions, tag, neon-image, compute-node-image ]
strategy:
fail-fast: false
matrix:
arch: [ x64, arm64 ]
runs-on: ${{ fromJson(format('["self-hosted", "gen3", "{0}"]', matrix.arch == 'arm64' && 'small-arm64' || 'small')) }}
runs-on: [ self-hosted, gen3, small ]
steps:
- name: Checkout
@@ -1023,18 +910,6 @@ jobs:
with:
fetch-depth: 0
# Use custom DOCKER_CONFIG directory to avoid conflicts with default settings
# The default value is ~/.docker
- name: Set custom docker config directory
run: |
mkdir -p .docker-custom
echo DOCKER_CONFIG=$(pwd)/.docker-custom >> $GITHUB_ENV
- uses: docker/login-action@v3
with:
username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
# `neondatabase/neon` contains multiple binaries, all of them use the same input for the version into the same version formatting library.
# Pick pageserver as currently the only binary with extra "version" features printed in the string to verify.
# Regular pageserver version string looks like
@@ -1045,7 +920,7 @@ jobs:
- name: Verify image versions
shell: bash # ensure no set -e for better error messages
run: |
pageserver_version=$(docker run --rm neondatabase/neon:${{ needs.tag.outputs.build-tag }} "/bin/sh" "-c" "/usr/local/bin/pageserver --version")
pageserver_version=$(docker run --rm 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}} "/bin/sh" "-c" "/usr/local/bin/pageserver --version")
echo "Pageserver version string: $pageserver_version"
@@ -1059,7 +934,7 @@ jobs:
exit 1
fi
- name: Verify docker-compose example and test extensions
- name: Verify docker-compose example
timeout-minutes: 20
run: env TAG=${{needs.tag.outputs.build-tag}} ./docker-compose/docker_compose_test.sh
@@ -1069,78 +944,84 @@ jobs:
docker compose -f ./docker-compose/docker-compose.yml logs || 0
docker compose -f ./docker-compose/docker-compose.yml down
- name: Remove custom docker config directory
if: always()
run: |
rm -rf .docker-custom
promote-images:
needs: [ check-permissions, tag, test-images, vm-compute-node-image ]
runs-on: ubuntu-22.04
env:
VERSIONS: v14 v15 v16
runs-on: [ self-hosted, gen3, small ]
container: golang:1.19-bullseye
# Don't add if-condition here.
# The job should always be run because we have dependant other jobs that shouldn't be skipped
steps:
- uses: docker/login-action@v3
with:
username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
- name: Login to dev ECR
uses: docker/login-action@v3
with:
registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
password: ${{ secrets.AWS_SECRET_KEY_DEV }}
- name: Copy vm-compute-node images to ECR
- name: Install Crane & ECR helper
run: |
for version in ${VERSIONS}; do
docker buildx imagetools create -t 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-${version}:${{ needs.tag.outputs.build-tag }} \
neondatabase/vm-compute-node-${version}:${{ needs.tag.outputs.build-tag }}
done
go install github.com/google/go-containerregistry/cmd/crane@31786c6cbb82d6ec4fb8eb79cd9387905130534e # v0.11.0
go install github.com/awslabs/amazon-ecr-credential-helper/ecr-login/cli/docker-credential-ecr-login@69c85dc22db6511932bbf119e1a0cc5c90c69a7f # v0.6.0
- name: Configure ECR login
run: |
mkdir /github/home/.docker/
echo "{\"credsStore\":\"ecr-login\"}" > /github/home/.docker/config.json
- name: Copy vm-compute-node images to Docker Hub
run: |
crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v14:${{needs.tag.outputs.build-tag}} vm-compute-node-v14
crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v15:${{needs.tag.outputs.build-tag}} vm-compute-node-v15
crane pull 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v16:${{needs.tag.outputs.build-tag}} vm-compute-node-v16
- name: Add latest tag to images
if: github.ref_name == 'main'
if: github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy'
run: |
for repo in neondatabase 369495373322.dkr.ecr.eu-central-1.amazonaws.com; do
docker buildx imagetools create -t $repo/neon:latest \
$repo/neon:${{ needs.tag.outputs.build-tag }}
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v14:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v15:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v16:${{needs.tag.outputs.build-tag}} latest
crane tag 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v16:${{needs.tag.outputs.build-tag}} latest
docker buildx imagetools create -t $repo/compute-tools:latest \
$repo/compute-tools:${{ needs.tag.outputs.build-tag }}
for version in ${VERSIONS}; do
docker buildx imagetools create -t $repo/compute-node-${version}:latest \
$repo/compute-node-${version}:${{ needs.tag.outputs.build-tag }}
docker buildx imagetools create -t $repo/vm-compute-node-${version}:latest \
$repo/vm-compute-node-${version}:${{ needs.tag.outputs.build-tag }}
done
done
docker buildx imagetools create -t neondatabase/neon-test-extensions-v16:latest \
neondatabase/neon-test-extensions-v16:${{ needs.tag.outputs.build-tag }}
- name: Login to prod ECR
uses: docker/login-action@v3
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy'
with:
registry: 093970136003.dkr.ecr.eu-central-1.amazonaws.com
username: ${{ secrets.PROD_GHA_RUNNER_LIMITED_AWS_ACCESS_KEY_ID }}
password: ${{ secrets.PROD_GHA_RUNNER_LIMITED_AWS_SECRET_ACCESS_KEY }}
- name: Copy all images to prod ECR
if: github.ref_name == 'release'|| github.ref_name == 'release-proxy'
- name: Push images to production ECR
if: github.ref_name == 'main' || github.ref_name == 'release'|| github.ref_name == 'release-proxy'
run: |
for image in neon compute-tools {vm-,}compute-node-{v14,v15,v16}; do
docker buildx imagetools create -t 093970136003.dkr.ecr.eu-central-1.amazonaws.com/${image}:${{ needs.tag.outputs.build-tag }} \
369495373322.dkr.ecr.eu-central-1.amazonaws.com/${image}:${{ needs.tag.outputs.build-tag }}
done
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/neon:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/neon:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-tools:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v14:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v14:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v14:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v15:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v15:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v15:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v16:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/compute-node-v16:latest
crane copy 369495373322.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v16:${{needs.tag.outputs.build-tag}} 093970136003.dkr.ecr.eu-central-1.amazonaws.com/vm-compute-node-v16:latest
- name: Configure Docker Hub login
run: |
# ECR Credential Helper & Docker Hub don't work together in config, hence reset
echo "" > /github/home/.docker/config.json
crane auth login -u ${{ secrets.NEON_DOCKERHUB_USERNAME }} -p ${{ secrets.NEON_DOCKERHUB_PASSWORD }} index.docker.io
- name: Push vm-compute-node to Docker Hub
run: |
crane push vm-compute-node-v14 neondatabase/vm-compute-node-v14:${{needs.tag.outputs.build-tag}}
crane push vm-compute-node-v15 neondatabase/vm-compute-node-v15:${{needs.tag.outputs.build-tag}}
crane push vm-compute-node-v16 neondatabase/vm-compute-node-v16:${{needs.tag.outputs.build-tag}}
- name: Push latest tags to Docker Hub
if: github.ref_name == 'main' || github.ref_name == 'release'|| github.ref_name == 'release-proxy'
run: |
crane tag neondatabase/neon:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/compute-tools:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/compute-node-v14:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/vm-compute-node-v14:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/compute-node-v15:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/vm-compute-node-v15:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/compute-node-v16:${{needs.tag.outputs.build-tag}} latest
crane tag neondatabase/vm-compute-node-v16:${{needs.tag.outputs.build-tag}} latest
- name: Cleanup ECR folder
run: rm -rf ~/.ecr
trigger-custom-extensions-build-and-wait:
needs: [ check-permissions, tag ]
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- name: Set PR's status to pending and request a remote CI test
run: |
@@ -1245,7 +1126,6 @@ jobs:
run: |
if [[ "$GITHUB_REF_NAME" == "main" ]]; then
gh workflow --repo neondatabase/aws run deploy-dev.yml --ref main -f branch=main -f dockerTag=${{needs.tag.outputs.build-tag}} -f deployPreprodRegion=false
gh workflow --repo neondatabase/azure run deploy.yml -f dockerTag=${{needs.tag.outputs.build-tag}}
elif [[ "$GITHUB_REF_NAME" == "release" ]]; then
gh workflow --repo neondatabase/aws run deploy-dev.yml --ref main \
-f deployPgSniRouter=false \

View File

@@ -19,23 +19,32 @@ permissions: {}
jobs:
check-image:
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
outputs:
tag: ${{ steps.get-build-tools-tag.outputs.image-tag }}
found: ${{ steps.check-image.outputs.found }}
steps:
- uses: actions/checkout@v4
- name: Get build-tools image tag for the current commit
id: get-build-tools-tag
env:
IMAGE_TAG: |
${{ hashFiles('Dockerfile.build-tools',
'.github/workflows/check-build-tools-image.yml',
'.github/workflows/build-build-tools-image.yml') }}
# Usually, for COMMIT_SHA, we use `github.event.pull_request.head.sha || github.sha`, but here, even for PRs,
# we want to use `github.sha` i.e. point to a phantom merge commit to determine the image tag correctly.
COMMIT_SHA: ${{ github.sha }}
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
echo "image-tag=${IMAGE_TAG}" | tee -a $GITHUB_OUTPUT
LAST_BUILD_TOOLS_SHA=$(
gh api \
-H "Accept: application/vnd.github+json" \
-H "X-GitHub-Api-Version: 2022-11-28" \
--method GET \
--field path=Dockerfile.build-tools \
--field sha=${COMMIT_SHA} \
--field per_page=1 \
--jq ".[0].sha" \
"/repos/${GITHUB_REPOSITORY}/commits"
)
echo "image-tag=${LAST_BUILD_TOOLS_SHA}" | tee -a $GITHUB_OUTPUT
- name: Check if such tag found in the registry
id: check-image

View File

@@ -16,7 +16,7 @@ permissions: {}
jobs:
check-permissions:
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- name: Disallow CI runs on PRs from forks
if: |

View File

@@ -9,7 +9,7 @@ on:
jobs:
cleanup:
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- name: Cleanup
run: |

View File

@@ -20,7 +20,7 @@ concurrency:
jobs:
test-postgres-client-libs:
# TODO: switch to gen2 runner, requires docker
runs-on: ubuntu-22.04
runs-on: [ ubuntu-latest ]
env:
DEFAULT_PG_VERSION: 14

View File

@@ -26,7 +26,7 @@ permissions: {}
jobs:
tag-image:
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
env:
FROM_TAG: ${{ inputs.from-tag }}

View File

@@ -19,7 +19,7 @@ on:
jobs:
notify:
runs-on: ubuntu-22.04
runs-on: [ ubuntu-latest ]
steps:
- uses: neondatabase/dev-actions/release-pr-notify@main

View File

@@ -26,7 +26,7 @@ defaults:
jobs:
create-storage-release-branch:
if: ${{ github.event.schedule == '0 6 * * MON' || format('{0}', inputs.create-storage-release-branch) == 'true' }}
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
permissions:
contents: write # for `git push`
@@ -52,22 +52,20 @@ jobs:
env:
GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
run: |
TITLE="Storage & Compute release ${RELEASE_DATE}"
cat << EOF > body.md
## ${TITLE}
## Release ${RELEASE_DATE}
**Please merge this Pull Request using 'Create a merge commit' button**
EOF
gh pr create --title "${TITLE}" \
gh pr create --title "Release ${RELEASE_DATE}" \
--body-file "body.md" \
--head "${RELEASE_BRANCH}" \
--base "release"
create-proxy-release-branch:
if: ${{ github.event.schedule == '0 6 * * THU' || format('{0}', inputs.create-proxy-release-branch) == 'true' }}
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
permissions:
contents: write # for `git push`
@@ -93,15 +91,13 @@ jobs:
env:
GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
run: |
TITLE="Proxy release ${RELEASE_DATE}"
cat << EOF > body.md
## ${TITLE}
## Proxy release ${RELEASE_DATE}
**Please merge this Pull Request using 'Create a merge commit' button**
EOF
gh pr create --title "${TITLE}" \
gh pr create --title "Proxy release ${RELEASE_DATE}" \
--body-file "body.md" \
--head "${RELEASE_BRANCH}" \
--base "release-proxy"

View File

@@ -19,7 +19,7 @@ env:
jobs:
cancel-previous-e2e-tests:
if: github.event_name == 'pull_request'
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
steps:
- name: Cancel previous e2e-tests runs for this PR
@@ -31,7 +31,7 @@ jobs:
--field concurrency_group="${{ env.E2E_CONCURRENCY_GROUP }}"
tag:
runs-on: ubuntu-22.04
runs-on: [ ubuntu-latest ]
outputs:
build-tag: ${{ steps.build-tag.outputs.tag }}
@@ -62,7 +62,7 @@ jobs:
trigger-e2e-tests:
needs: [ tag ]
runs-on: ubuntu-22.04
runs-on: ubuntu-latest
env:
TAG: ${{ needs.tag.outputs.build-tag }}
steps:

View File

@@ -1,5 +1,4 @@
# * `-A unknown_lints` do not warn about unknown lint suppressions
# that people with newer toolchains might use
# * `-D warnings` - fail on any warnings (`cargo` returns non-zero exit status)
# * `-D clippy::todo` - don't let `todo!()` slip into `main`
export CLIPPY_COMMON_ARGS="--locked --workspace --all-targets -- -A unknown_lints -D warnings -D clippy::todo"
export CLIPPY_COMMON_ARGS="--locked --workspace --all-targets -- -A unknown_lints -D warnings"

387
Cargo.lock generated
View File

@@ -776,6 +776,7 @@ dependencies = [
"pin-project",
"serde",
"time",
"tz-rs",
"url",
"uuid",
]
@@ -1014,9 +1015,6 @@ name = "camino"
version = "1.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "c59e92b5a388f549b863a7bea62612c09f24c8393560709a54558a9abdfb3b9c"
dependencies = [
"serde",
]
[[package]]
name = "camino-tempfile"
@@ -1246,7 +1244,7 @@ dependencies = [
"tokio-postgres",
"tokio-stream",
"tokio-util",
"toml_edit 0.19.10",
"toml_edit",
"tracing",
"tracing-opentelemetry",
"tracing-subscriber",
@@ -1293,6 +1291,12 @@ dependencies = [
"tiny-keccak",
]
[[package]]
name = "const_fn"
version = "0.4.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fbdcdcb6d86f71c5e97409ad45898af11cbc995b4ee8112d59095a28d376c935"
[[package]]
name = "const_format"
version = "0.2.30"
@@ -1362,8 +1366,8 @@ dependencies = [
"tokio",
"tokio-postgres",
"tokio-util",
"toml 0.7.4",
"toml_edit 0.19.10",
"toml",
"toml_edit",
"tracing",
"url",
"utils",
@@ -1467,21 +1471,26 @@ dependencies = [
[[package]]
name = "crossbeam-deque"
version = "0.8.5"
version = "0.8.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "613f8cc01fe9cf1a3eb3d7f488fd2fa8388403e97039e2f73692932e291a770d"
checksum = "ce6fd6f855243022dcecf8702fef0c297d4338e226845fe067f6341ad9fa0cef"
dependencies = [
"cfg-if",
"crossbeam-epoch",
"crossbeam-utils",
]
[[package]]
name = "crossbeam-epoch"
version = "0.9.18"
version = "0.9.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "5b82ac4a3c2ca9c3460964f020e1402edd5753411d7737aa39c3714ad1b5420e"
checksum = "46bd5f3f85273295a9d14aedfb86f6aadbff6d8f5295c4a9edb08e819dcf5695"
dependencies = [
"autocfg",
"cfg-if",
"crossbeam-utils",
"memoffset 0.8.0",
"scopeguard",
]
[[package]]
@@ -1669,9 +1678,9 @@ dependencies = [
[[package]]
name = "diesel"
version = "2.2.1"
version = "2.1.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "62d6dcd069e7b5fe49a302411f759d4cf1cf2c27fe798ef46fb8baefc053dd2b"
checksum = "62c6fcf842f17f8c78ecf7c81d75c5ce84436b41ee07e03f490fbb5f5a8731d8"
dependencies = [
"bitflags 2.4.1",
"byteorder",
@@ -1684,12 +1693,11 @@ dependencies = [
[[package]]
name = "diesel_derives"
version = "2.2.1"
version = "2.1.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "59de76a222c2b8059f789cbe07afbfd8deb8c31dd0bc2a21f85e256c1def8259"
checksum = "ef8337737574f55a468005a83499da720f20c65586241ffea339db9ecdfd2b44"
dependencies = [
"diesel_table_macro_syntax",
"dsl_auto_type",
"proc-macro2",
"quote",
"syn 2.0.52",
@@ -1697,9 +1705,9 @@ dependencies = [
[[package]]
name = "diesel_migrations"
version = "2.2.0"
version = "2.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8a73ce704bad4231f001bff3314d91dce4aba0770cee8b233991859abc15c1f6"
checksum = "6036b3f0120c5961381b570ee20a02432d7e2d27ea60de9578799cf9156914ac"
dependencies = [
"diesel",
"migrations_internals",
@@ -1708,9 +1716,9 @@ dependencies = [
[[package]]
name = "diesel_table_macro_syntax"
version = "0.2.0"
version = "0.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "209c735641a413bc68c4923a9d6ad4bcb3ca306b794edaa7eb0b3228a99ffb25"
checksum = "fc5557efc453706fed5e4fa85006fe9817c224c3f480a34c7e5959fd700921c5"
dependencies = [
"syn 2.0.52",
]
@@ -1746,20 +1754,6 @@ dependencies = [
"const-random",
]
[[package]]
name = "dsl_auto_type"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0892a17df262a24294c382f0d5997571006e7a4348b4327557c4ff1cd4a8bccc"
dependencies = [
"darling",
"either",
"heck 0.5.0",
"proc-macro2",
"quote",
"syn 2.0.52",
]
[[package]]
name = "dyn-clone"
version = "1.0.14"
@@ -1987,6 +1981,21 @@ version = "1.0.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
[[package]]
name = "foreign-types"
version = "0.3.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f6f339eb8adc052cd2ca78910fda869aefa38d22d5cb648e6485e4d3fc06f3b1"
dependencies = [
"foreign-types-shared",
]
[[package]]
name = "foreign-types-shared"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "00b0228411908ca8685dba7fc2cdd70ec9990a6e753e89b6ac91a84c40fbaf4b"
[[package]]
name = "form_urlencoded"
version = "1.1.0"
@@ -2616,6 +2625,19 @@ dependencies = [
"tokio-io-timeout",
]
[[package]]
name = "hyper-tls"
version = "0.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d6183ddfa99b85da61a140bea0efc93fdf56ceaa041b37d553518030827f9905"
dependencies = [
"bytes",
"hyper 0.14.26",
"native-tls",
"tokio",
"tokio-native-tls",
]
[[package]]
name = "hyper-util"
version = "0.1.3"
@@ -2933,12 +2955,6 @@ version = "0.4.13"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "01cda141df6706de531b6c46c3a33ecca755538219bd484262fa09410c13539c"
[[package]]
name = "linux-raw-sys"
version = "0.6.4"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f0b5399f6804fbab912acbd8878ed3532d506b7c951b8f9f164ef90fef39e3f4"
[[package]]
name = "lock_api"
version = "0.4.10"
@@ -3099,19 +3115,19 @@ dependencies = [
[[package]]
name = "migrations_internals"
version = "2.2.0"
version = "2.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "fd01039851e82f8799046eabbb354056283fb265c8ec0996af940f4e85a380ff"
checksum = "0f23f71580015254b020e856feac3df5878c2c7a8812297edd6c0a485ac9dada"
dependencies = [
"serde",
"toml 0.8.14",
"toml",
]
[[package]]
name = "migrations_macros"
version = "2.2.0"
version = "2.1.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ffb161cc72176cb37aa47f1fc520d3ef02263d67d661f44f05d05a079e1237fd"
checksum = "cce3325ac70e67bbab5bd837a31cae01f1a6db64e0e744a33cb03a543469ef08"
dependencies = [
"migrations_internals",
"proc-macro2",
@@ -3157,6 +3173,24 @@ version = "0.8.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e5ce46fe64a9d73be07dcbe690a38ce1b293be448fd8ce1e6c1b8062c9f72c6a"
[[package]]
name = "native-tls"
version = "0.2.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "07226173c32f2926027b63cce4bcd8076c3552846cbe7925f3aaffeac0a3b92e"
dependencies = [
"lazy_static",
"libc",
"log",
"openssl",
"openssl-probe",
"openssl-sys",
"schannel",
"security-framework",
"security-framework-sys",
"tempfile",
]
[[package]]
name = "nix"
version = "0.25.1"
@@ -3327,6 +3361,15 @@ dependencies = [
"libc",
]
[[package]]
name = "num_threads"
version = "0.1.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2819ce041d2ee131036f4fc9d6ae7ae125a3a40e97ba64d04fe799ad9dabbb44"
dependencies = [
"libc",
]
[[package]]
name = "oauth2"
version = "4.4.2"
@@ -3376,12 +3419,50 @@ version = "11.1.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "0ab1bc2a289d34bd04a330323ac98a1b4bc82c9d9fcb1e66b63caa84da26b575"
[[package]]
name = "openssl"
version = "0.10.60"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "79a4c6c3a2b158f7f8f2a2fc5a969fa3a068df6fc9dbb4a43845436e3af7c800"
dependencies = [
"bitflags 2.4.1",
"cfg-if",
"foreign-types",
"libc",
"once_cell",
"openssl-macros",
"openssl-sys",
]
[[package]]
name = "openssl-macros"
version = "0.1.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "a948666b637a0f465e8564c73e89d4dde00d72d4d473cc972f390fc3dcee7d9c"
dependencies = [
"proc-macro2",
"quote",
"syn 2.0.52",
]
[[package]]
name = "openssl-probe"
version = "0.1.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "ff011a302c396a5197692431fc1948019154afc178baf7d8e37367442a4601cf"
[[package]]
name = "openssl-sys"
version = "0.9.96"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3812c071ba60da8b5677cc12bcb1d42989a65553772897a7e0355545a819838f"
dependencies = [
"cc",
"libc",
"pkg-config",
"vcpkg",
]
[[package]]
name = "opentelemetry"
version = "0.20.0"
@@ -3588,10 +3669,9 @@ dependencies = [
"serde",
"serde_json",
"svg_fmt",
"thiserror",
"tokio",
"tokio-util",
"toml_edit 0.19.10",
"toml_edit",
"utils",
"workspace_hack",
]
@@ -3674,7 +3754,7 @@ dependencies = [
"tokio-stream",
"tokio-tar",
"tokio-util",
"toml_edit 0.19.10",
"toml_edit",
"tracing",
"twox-hash",
"url",
@@ -3881,9 +3961,9 @@ checksum = "de3145af08024dea9fa9914f381a17b8fc6034dfb00f3a84013f7ff43f29ed4c"
[[package]]
name = "pbkdf2"
version = "0.12.2"
version = "0.12.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f8ed6a7761f76e3b9f92dfb0a60a6a6477c61024b775147ff0973a02653abaf2"
checksum = "f0ca0b5a68607598bf3bad68f32227a8164f6254833f84eafaac409cd6746c31"
dependencies = [
"digest",
"hmac",
@@ -4030,6 +4110,17 @@ dependencies = [
"tokio-postgres",
]
[[package]]
name = "postgres-native-tls"
version = "0.5.0"
source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#20031d7a9ee1addeae6e0968e3899ae6bf01cee2"
dependencies = [
"native-tls",
"tokio",
"tokio-native-tls",
"tokio-postgres",
]
[[package]]
name = "postgres-protocol"
version = "0.6.4"
@@ -4138,7 +4229,6 @@ version = "0.1.0"
dependencies = [
"byteorder",
"bytes",
"itertools",
"pin-project-lite",
"postgres-protocol",
"rand 0.8.5",
@@ -4296,7 +4386,6 @@ dependencies = [
name = "proxy"
version = "0.1.0"
dependencies = [
"ahash",
"anyhow",
"async-compression",
"async-trait",
@@ -4313,7 +4402,6 @@ dependencies = [
"chrono",
"clap",
"consumption_metrics",
"crossbeam-deque",
"dashmap",
"env_logger",
"fallible-iterator",
@@ -4328,7 +4416,6 @@ dependencies = [
"http 1.1.0",
"http-body-util",
"humantime",
"humantime-serde",
"hyper 0.14.26",
"hyper 1.2.0",
"hyper-util",
@@ -4339,6 +4426,7 @@ dependencies = [
"md5",
"measured",
"metrics",
"native-tls",
"once_cell",
"opentelemetry",
"parking_lot 0.12.1",
@@ -4346,6 +4434,7 @@ dependencies = [
"parquet_derive",
"pbkdf2",
"pin-project-lite",
"postgres-native-tls",
"postgres-protocol",
"postgres_backend",
"pq_proto",
@@ -4364,7 +4453,6 @@ dependencies = [
"rstest",
"rustc-hash",
"rustls 0.22.4",
"rustls-native-certs 0.7.0",
"rustls-pemfile 2.1.1",
"scopeguard",
"serde",
@@ -4394,6 +4482,7 @@ dependencies = [
"utils",
"uuid",
"walkdir",
"webpki-roots 0.25.2",
"workspace_hack",
"x509-parser",
]
@@ -4665,7 +4754,6 @@ dependencies = [
"futures-util",
"http-types",
"humantime",
"humantime-serde",
"hyper 0.14.26",
"itertools",
"metrics",
@@ -4680,7 +4768,7 @@ dependencies = [
"tokio",
"tokio-stream",
"tokio-util",
"toml_edit 0.19.10",
"toml_edit",
"tracing",
"utils",
"workspace_hack",
@@ -4701,21 +4789,20 @@ dependencies = [
"http 0.2.9",
"http-body 0.4.5",
"hyper 0.14.26",
"hyper-rustls 0.24.0",
"hyper-tls",
"ipnet",
"js-sys",
"log",
"mime",
"native-tls",
"once_cell",
"percent-encoding",
"pin-project-lite",
"rustls 0.21.11",
"rustls-pemfile 1.0.2",
"serde",
"serde_json",
"serde_urlencoded",
"tokio",
"tokio-rustls 0.24.0",
"tokio-native-tls",
"tokio-util",
"tower-service",
"url",
@@ -4723,7 +4810,6 @@ dependencies = [
"wasm-bindgen-futures",
"wasm-streams 0.3.0",
"web-sys",
"webpki-roots 0.25.2",
"winreg 0.50.0",
]
@@ -5128,6 +5214,51 @@ version = "1.0.13"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f91339c0467de62360649f8d3e185ca8de4224ff281f66000de5eb2a77a79041"
[[package]]
name = "s3_scrubber"
version = "0.1.0"
dependencies = [
"anyhow",
"async-stream",
"aws-config",
"aws-sdk-s3",
"aws-smithy-async",
"bincode",
"bytes",
"camino",
"chrono",
"clap",
"crc32c",
"either",
"futures",
"futures-util",
"hex",
"histogram",
"itertools",
"native-tls",
"pageserver",
"pageserver_api",
"postgres-native-tls",
"postgres_ffi",
"rand 0.8.5",
"remote_storage",
"reqwest 0.12.4",
"serde",
"serde_json",
"serde_with",
"thiserror",
"tokio",
"tokio-postgres",
"tokio-rustls 0.25.0",
"tokio-stream",
"tokio-util",
"tracing",
"tracing-appender",
"tracing-subscriber",
"utils",
"workspace_hack",
]
[[package]]
name = "safekeeper"
version = "0.1.0"
@@ -5177,9 +5308,8 @@ dependencies = [
"tokio-io-timeout",
"tokio-postgres",
"tokio-stream",
"tokio-tar",
"tokio-util",
"toml_edit 0.19.10",
"toml_edit",
"tracing",
"tracing-subscriber",
"url",
@@ -5458,9 +5588,9 @@ dependencies = [
[[package]]
name = "serde_spanned"
version = "0.6.6"
version = "0.6.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "79e674e01f999af37c49f70a6ede167a8a60b2503e56c5599532a65baa5969a0"
checksum = "93107647184f6027e3b7dcb2e11034cf95ffa1e3a682c67951963ac69c1c007d"
dependencies = [
"serde",
]
@@ -5773,7 +5903,6 @@ dependencies = [
"r2d2",
"reqwest 0.12.4",
"routerify",
"scopeguard",
"serde",
"serde_json",
"strum",
@@ -5786,54 +5915,6 @@ dependencies = [
"workspace_hack",
]
[[package]]
name = "storage_scrubber"
version = "0.1.0"
dependencies = [
"anyhow",
"async-stream",
"aws-config",
"aws-sdk-s3",
"aws-smithy-async",
"bincode",
"bytes",
"camino",
"chrono",
"clap",
"crc32c",
"either",
"futures",
"futures-util",
"hex",
"histogram",
"humantime",
"itertools",
"once_cell",
"pageserver",
"pageserver_api",
"postgres_ffi",
"rand 0.8.5",
"remote_storage",
"reqwest 0.12.4",
"rustls 0.22.4",
"rustls-native-certs 0.7.0",
"serde",
"serde_json",
"serde_with",
"thiserror",
"tokio",
"tokio-postgres",
"tokio-postgres-rustls",
"tokio-rustls 0.25.0",
"tokio-stream",
"tokio-util",
"tracing",
"tracing-appender",
"tracing-subscriber",
"utils",
"workspace_hack",
]
[[package]]
name = "storcon_cli"
version = "0.1.0"
@@ -5841,8 +5922,6 @@ dependencies = [
"anyhow",
"clap",
"comfy-table",
"futures",
"humantime",
"hyper 0.14.26",
"pageserver_api",
"pageserver_client",
@@ -6113,6 +6192,8 @@ checksum = "8f3403384eaacbca9923fa06940178ac13e4edb725486d70e8e15881d0c836cc"
dependencies = [
"itoa",
"js-sys",
"libc",
"num_threads",
"serde",
"time-core",
"time-macros",
@@ -6188,7 +6269,7 @@ dependencies = [
[[package]]
name = "tokio-epoll-uring"
version = "0.1.0"
source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#08ccfa94ff5507727bf4d8d006666b5b192e04c6"
source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#342ddd197a060a8354e8f11f4d12994419fff939"
dependencies = [
"futures",
"nix 0.26.4",
@@ -6222,6 +6303,16 @@ dependencies = [
"syn 2.0.52",
]
[[package]]
name = "tokio-native-tls"
version = "0.3.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "bbae76ab933c85776efabc971569dd6119c580d8f5d448769dec1764bf796ef2"
dependencies = [
"native-tls",
"tokio",
]
[[package]]
name = "tokio-postgres"
version = "0.7.7"
@@ -6345,26 +6436,14 @@ dependencies = [
"serde",
"serde_spanned",
"toml_datetime",
"toml_edit 0.19.10",
]
[[package]]
name = "toml"
version = "0.8.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "6f49eb2ab21d2f26bd6db7bf383edc527a7ebaee412d17af4d40fdccd442f335"
dependencies = [
"serde",
"serde_spanned",
"toml_datetime",
"toml_edit 0.22.14",
"toml_edit",
]
[[package]]
name = "toml_datetime"
version = "0.6.6"
version = "0.6.2"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "4badfd56924ae69bcc9039335b2e017639ce3f9b001c393c1b2d1ef846ce2cbf"
checksum = "5a76a9312f5ba4c2dec6b9161fdf25d87ad8a09256ccea5a556fef03c706a10f"
dependencies = [
"serde",
]
@@ -6379,20 +6458,7 @@ dependencies = [
"serde",
"serde_spanned",
"toml_datetime",
"winnow 0.4.6",
]
[[package]]
name = "toml_edit"
version = "0.22.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "f21c7aaf97f1bd9ca9d4f9e73b0a6c74bd5afef56f2bc931943a6e1c37e04e38"
dependencies = [
"indexmap 2.0.1",
"serde",
"serde_spanned",
"toml_datetime",
"winnow 0.6.13",
"winnow",
]
[[package]]
@@ -6653,6 +6719,15 @@ version = "1.16.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "497961ef93d974e23eb6f433eb5fe1b7930b659f06d12dec6fc44a8f554c0bba"
[[package]]
name = "tz-rs"
version = "0.6.14"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "33851b15c848fad2cf4b105c6bb66eb9512b6f6c44a4b13f57c53c73c707e2b4"
dependencies = [
"const_fn",
]
[[package]]
name = "uname"
version = "0.1.1"
@@ -6725,12 +6800,11 @@ dependencies = [
[[package]]
name = "uring-common"
version = "0.1.0"
source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#08ccfa94ff5507727bf4d8d006666b5b192e04c6"
source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#342ddd197a060a8354e8f11f4d12994419fff939"
dependencies = [
"bytes",
"io-uring",
"libc",
"linux-raw-sys 0.6.4",
]
[[package]]
@@ -7375,15 +7449,6 @@ dependencies = [
"memchr",
]
[[package]]
name = "winnow"
version = "0.6.13"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "59b5e5f6c299a3c7890b876a2a587f3115162487e704907d9b6cd29473052ba1"
dependencies = [
"memchr",
]
[[package]]
name = "winreg"
version = "0.50.0"
@@ -7408,7 +7473,6 @@ dependencies = [
name = "workspace_hack"
version = "0.1.0"
dependencies = [
"ahash",
"anyhow",
"aws-config",
"aws-runtime",
@@ -7420,7 +7484,6 @@ dependencies = [
"base64 0.21.1",
"base64ct",
"bytes",
"camino",
"cc",
"chrono",
"clap",
@@ -7473,7 +7536,7 @@ dependencies = [
"tokio-rustls 0.24.0",
"tokio-util",
"toml_datetime",
"toml_edit 0.19.10",
"toml_edit",
"tonic",
"tower",
"tracing",
@@ -7568,9 +7631,9 @@ dependencies = [
[[package]]
name = "zeroize"
version = "1.7.0"
version = "1.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "525b4ec142c6b68a2d10f01f7bbf6755599ca3f81ea53b8431b7dd348f5fdb2d"
checksum = "2a0956f1ba7c7909bfb66c2e9e4124ab6f6482560f6628b5aaeba39207c9aad9"
dependencies = [
"zeroize_derive",
]

View File

@@ -13,7 +13,7 @@ members = [
"safekeeper",
"storage_broker",
"storage_controller",
"storage_scrubber",
"s3_scrubber",
"workspace_hack",
"trace",
"libs/compute_api",
@@ -41,15 +41,14 @@ license = "Apache-2.0"
## All dependency versions, used in the project
[workspace.dependencies]
ahash = "0.8"
anyhow = { version = "1.0", features = ["backtrace"] }
arc-swap = "1.6"
async-compression = { version = "0.4.0", features = ["tokio", "gzip", "zstd"] }
atomic-take = "1.1.0"
azure_core = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls", "hmac_rust"] }
azure_identity = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls"] }
azure_storage = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls"] }
azure_storage_blobs = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls"] }
azure_core = "0.19"
azure_identity = "0.19"
azure_storage = "0.19"
azure_storage_blobs = "0.19"
flate2 = "1.0.26"
async-stream = "0.3"
async-trait = "0.1"
@@ -75,7 +74,6 @@ clap = { version = "4.0", features = ["derive"] }
comfy-table = "6.1"
const_format = "0.2"
crc32c = "0.6"
crossbeam-deque = "0.8.5"
crossbeam-utils = "0.8.5"
dashmap = { version = "5.5.0", features = ["raw-api"] }
either = "1.8"
@@ -114,13 +112,14 @@ md5 = "0.7.0"
measured = { version = "0.0.21", features=["lasso"] }
measured-process = { version = "0.0.21" }
memoffset = "0.8"
native-tls = "0.2"
nix = { version = "0.27", features = ["fs", "process", "socket", "signal", "poll"] }
notify = "6.0.0"
num_cpus = "1.15"
num-traits = "0.2.15"
once_cell = "1.13"
opentelemetry = "0.20.0"
opentelemetry-otlp = { version = "0.13.0", default-features=false, features = ["http-proto", "trace", "http", "reqwest-client"] }
opentelemetry-otlp = { version = "0.13.0", default_features=false, features = ["http-proto", "trace", "http", "reqwest-client"] }
opentelemetry-semantic-conventions = "0.12.0"
parking_lot = "0.12"
parquet = { version = "51.0.0", default-features = false, features = ["zstd"] }
@@ -128,7 +127,7 @@ parquet_derive = "51.0.0"
pbkdf2 = { version = "0.12.1", features = ["simple", "std"] }
pin-project-lite = "0.2"
procfs = "0.14"
prometheus = {version = "0.13", default-features=false, features = ["process"]} # removes protobuf dependency
prometheus = {version = "0.13", default_features=false, features = ["process"]} # removes protobuf dependency
prost = "0.11"
rand = "0.8"
redis = { version = "0.25.2", features = ["tokio-rustls-comp", "keep-alive"] }
@@ -184,13 +183,13 @@ tower-service = "0.3.2"
tracing = "0.1"
tracing-error = "0.2.0"
tracing-opentelemetry = "0.21.0"
tracing-subscriber = { version = "0.3", default-features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json", "ansi"] }
tracing-subscriber = { version = "0.3", default_features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json", "ansi"] }
twox-hash = { version = "1.6.3", default-features = false }
url = "2.2"
urlencoding = "2.1"
uuid = { version = "1.6.1", features = ["v4", "v7", "serde"] }
walkdir = "2.3.2"
rustls-native-certs = "0.7"
webpki-roots = "0.25"
x509-parser = "0.15"
## TODO replace this with tracing
@@ -199,6 +198,7 @@ log = "0.4"
## Libraries from neondatabase/ git forks, ideally with changes to be upstreamed
postgres = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
postgres-native-tls = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
postgres-protocol = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
postgres-types = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
tokio-postgres = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
@@ -239,7 +239,8 @@ tonic-build = "0.9"
[patch.crates-io]
# Needed to get `tokio-postgres-rustls` to depend on our fork.
# This is only needed for proxy's tests.
# TODO: we should probably fork `tokio-postgres-rustls` instead.
tokio-postgres = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
# bug fixes for UUID

View File

@@ -69,6 +69,8 @@ RUN set -e \
&& apt install -y \
libreadline-dev \
libseccomp-dev \
libicu67 \
openssl \
ca-certificates \
&& rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* \
&& useradd -d /data neon \

View File

@@ -73,6 +73,13 @@ RUN curl -fsSL 'https://apt.llvm.org/llvm-snapshot.gpg.key' | apt-key add - \
&& bash -c 'for f in /usr/bin/clang*-${LLVM_VERSION} /usr/bin/llvm*-${LLVM_VERSION}; do ln -s "${f}" "${f%-${LLVM_VERSION}}"; done' \
&& rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
# PostgreSQL 14
RUN curl -fsSL 'https://www.postgresql.org/media/keys/ACCC4CF8.asc' | apt-key add - \
&& echo 'deb http://apt.postgresql.org/pub/repos/apt bullseye-pgdg main' > /etc/apt/sources.list.d/pgdg.list \
&& apt update \
&& apt install -y postgresql-client-14 \
&& rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
# AWS CLI
RUN curl "https://awscli.amazonaws.com/awscli-exe-linux-$(uname -m).zip" -o "awscliv2.zip" \
&& unzip -q awscliv2.zip \
@@ -105,45 +112,6 @@ RUN for package in Capture::Tiny DateTime Devel::Cover Digest::MD5 File::Spec JS
&& make install \
&& rm -rf ../lcov.tar.gz
# Compile and install the static OpenSSL library
ENV OPENSSL_VERSION=1.1.1w
ENV OPENSSL_PREFIX=/usr/local/openssl
RUN wget -O /tmp/openssl-${OPENSSL_VERSION}.tar.gz https://www.openssl.org/source/openssl-${OPENSSL_VERSION}.tar.gz && \
echo "cf3098950cb4d853ad95c0841f1f9c6d3dc102dccfcacd521d93925208b76ac8 /tmp/openssl-${OPENSSL_VERSION}.tar.gz" | sha256sum --check && \
cd /tmp && \
tar xzvf /tmp/openssl-${OPENSSL_VERSION}.tar.gz && \
rm /tmp/openssl-${OPENSSL_VERSION}.tar.gz && \
cd /tmp/openssl-${OPENSSL_VERSION} && \
./config --prefix=${OPENSSL_PREFIX} -static --static no-shared -fPIC && \
make -j "$(nproc)" && \
make install && \
cd /tmp && \
rm -rf /tmp/openssl-${OPENSSL_VERSION}
# Use the same version of libicu as the compute nodes so that
# clusters created using inidb on pageserver can be used by computes.
#
# TODO: at this time, Dockerfile.compute-node uses the debian bullseye libicu
# package, which is 67.1. We're duplicating that knowledge here, and also, technically,
# Debian has a few patches on top of 67.1 that we're not adding here.
ENV ICU_VERSION=67.1
ENV ICU_PREFIX=/usr/local/icu
# Download and build static ICU
RUN wget -O /tmp/libicu-${ICU_VERSION}.tgz https://github.com/unicode-org/icu/releases/download/release-${ICU_VERSION//./-}/icu4c-${ICU_VERSION//./_}-src.tgz && \
echo "94a80cd6f251a53bd2a997f6f1b5ac6653fe791dfab66e1eb0227740fb86d5dc /tmp/libicu-${ICU_VERSION}.tgz" | sha256sum --check && \
mkdir /tmp/icu && \
pushd /tmp/icu && \
tar -xzf /tmp/libicu-${ICU_VERSION}.tgz && \
pushd icu/source && \
./configure --prefix=${ICU_PREFIX} --enable-static --enable-shared=no CXXFLAGS="-fPIC" CFLAGS="-fPIC" && \
make -j "$(nproc)" && \
make install && \
popd && \
rm -rf icu && \
rm -f /tmp/libicu-${ICU_VERSION}.tgz && \
popd
# Switch to nonroot user
USER nonroot:nonroot
WORKDIR /home/nonroot
@@ -173,7 +141,7 @@ WORKDIR /home/nonroot
# Rust
# Please keep the version of llvm (installed above) in sync with rust llvm (`rustc --version --verbose | grep LLVM`)
ENV RUSTC_VERSION=1.79.0
ENV RUSTC_VERSION=1.78.0
ENV RUSTUP_HOME="/home/nonroot/.rustup"
ENV PATH="/home/nonroot/.cargo/bin:${PATH}"
RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux-gnu/rustup-init && whoami && \
@@ -202,6 +170,3 @@ RUN whoami \
&& rustup --version --verbose \
&& rustc --version --verbose \
&& clang --version
# Set following flag to check in Makefile if its running in Docker
RUN touch /home/nonroot/.docker_build

View File

@@ -89,7 +89,7 @@ RUN apt update && \
# SFCGAL > 1.3 requires CGAL > 5.2, Bullseye's libcgal-dev is 5.2
RUN wget https://gitlab.com/Oslandia/SFCGAL/-/archive/v1.3.10/SFCGAL-v1.3.10.tar.gz -O SFCGAL.tar.gz && \
echo "4e39b3b2adada6254a7bdba6d297bb28e1a9835a9f879b74f37e2dab70203232 SFCGAL.tar.gz" | sha256sum --check && \
mkdir sfcgal-src && cd sfcgal-src && tar xzf ../SFCGAL.tar.gz --strip-components=1 -C . && \
mkdir sfcgal-src && cd sfcgal-src && tar xvzf ../SFCGAL.tar.gz --strip-components=1 -C . && \
cmake -DCMAKE_BUILD_TYPE=Release . && make -j $(getconf _NPROCESSORS_ONLN) && \
DESTDIR=/sfcgal make install -j $(getconf _NPROCESSORS_ONLN) && \
make clean && cp -R /sfcgal/* /
@@ -98,7 +98,7 @@ ENV PATH "/usr/local/pgsql/bin:$PATH"
RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.3.tar.gz -O postgis.tar.gz && \
echo "74eb356e3f85f14233791013360881b6748f78081cc688ff9d6f0f673a762d13 postgis.tar.gz" | sha256sum --check && \
mkdir postgis-src && cd postgis-src && tar xzf ../postgis.tar.gz --strip-components=1 -C . && \
mkdir postgis-src && cd postgis-src && tar xvzf ../postgis.tar.gz --strip-components=1 -C . && \
find /usr/local/pgsql -type f | sed 's|^/usr/local/pgsql/||' > /before.txt &&\
./autogen.sh && \
./configure --with-sfcgal=/usr/local/bin/sfcgal-config && \
@@ -124,7 +124,7 @@ RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.3.tar.gz -O postg
RUN wget https://github.com/pgRouting/pgrouting/archive/v3.4.2.tar.gz -O pgrouting.tar.gz && \
echo "cac297c07d34460887c4f3b522b35c470138760fe358e351ad1db4edb6ee306e pgrouting.tar.gz" | sha256sum --check && \
mkdir pgrouting-src && cd pgrouting-src && tar xzf ../pgrouting.tar.gz --strip-components=1 -C . && \
mkdir pgrouting-src && cd pgrouting-src && tar xvzf ../pgrouting.tar.gz --strip-components=1 -C . && \
mkdir build && cd build && \
cmake -DCMAKE_BUILD_TYPE=Release .. && \
make -j $(getconf _NPROCESSORS_ONLN) && \
@@ -149,7 +149,7 @@ RUN apt update && \
RUN wget https://github.com/plv8/plv8/archive/refs/tags/v3.1.10.tar.gz -O plv8.tar.gz && \
echo "7096c3290928561f0d4901b7a52794295dc47f6303102fae3f8e42dd575ad97d plv8.tar.gz" | sha256sum --check && \
mkdir plv8-src && cd plv8-src && tar xzf ../plv8.tar.gz --strip-components=1 -C . && \
mkdir plv8-src && cd plv8-src && tar xvzf ../plv8.tar.gz --strip-components=1 -C . && \
# generate and copy upgrade scripts
mkdir -p upgrade && ./generate_upgrade.sh 3.1.10 && \
cp upgrade/* /usr/local/pgsql/share/extension/ && \
@@ -194,7 +194,7 @@ RUN case "$(uname -m)" in \
RUN wget https://github.com/uber/h3/archive/refs/tags/v4.1.0.tar.gz -O h3.tar.gz && \
echo "ec99f1f5974846bde64f4513cf8d2ea1b8d172d2218ab41803bf6a63532272bc h3.tar.gz" | sha256sum --check && \
mkdir h3-src && cd h3-src && tar xzf ../h3.tar.gz --strip-components=1 -C . && \
mkdir h3-src && cd h3-src && tar xvzf ../h3.tar.gz --strip-components=1 -C . && \
mkdir build && cd build && \
cmake .. -DCMAKE_BUILD_TYPE=Release && \
make -j $(getconf _NPROCESSORS_ONLN) && \
@@ -204,7 +204,7 @@ RUN wget https://github.com/uber/h3/archive/refs/tags/v4.1.0.tar.gz -O h3.tar.gz
RUN wget https://github.com/zachasme/h3-pg/archive/refs/tags/v4.1.3.tar.gz -O h3-pg.tar.gz && \
echo "5c17f09a820859ffe949f847bebf1be98511fb8f1bd86f94932512c00479e324 h3-pg.tar.gz" | sha256sum --check && \
mkdir h3-pg-src && cd h3-pg-src && tar xzf ../h3-pg.tar.gz --strip-components=1 -C . && \
mkdir h3-pg-src && cd h3-pg-src && tar xvzf ../h3-pg.tar.gz --strip-components=1 -C . && \
export PATH="/usr/local/pgsql/bin:$PATH" && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
@@ -222,7 +222,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/df7cb/postgresql-unit/archive/refs/tags/7.7.tar.gz -O postgresql-unit.tar.gz && \
echo "411d05beeb97e5a4abf17572bfcfbb5a68d98d1018918feff995f6ee3bb03e79 postgresql-unit.tar.gz" | sha256sum --check && \
mkdir postgresql-unit-src && cd postgresql-unit-src && tar xzf ../postgresql-unit.tar.gz --strip-components=1 -C . && \
mkdir postgresql-unit-src && cd postgresql-unit-src && tar xvzf ../postgresql-unit.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
# unit extension's "create extension" script relies on absolute install path to fill some reference tables.
@@ -243,15 +243,12 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
COPY patches/pgvector.patch /pgvector.patch
# By default, pgvector Makefile uses `-march=native`. We don't want that,
# because we build the images on different machines than where we run them.
# Pass OPTFLAGS="" to remove it.
RUN wget https://github.com/pgvector/pgvector/archive/refs/tags/v0.7.2.tar.gz -O pgvector.tar.gz && \
echo "617fba855c9bcb41a2a9bc78a78567fd2e147c72afd5bf9d37b31b9591632b30 pgvector.tar.gz" | sha256sum --check && \
mkdir pgvector-src && cd pgvector-src && tar xzf ../pgvector.tar.gz --strip-components=1 -C . && \
RUN wget https://github.com/pgvector/pgvector/archive/refs/tags/v0.7.0.tar.gz -O pgvector.tar.gz && \
echo "1b5503a35c265408b6eb282621c5e1e75f7801afc04eecb950796cfee2e3d1d8 pgvector.tar.gz" | sha256sum --check && \
mkdir pgvector-src && cd pgvector-src && tar xvzf ../pgvector.tar.gz --strip-components=1 -C . && \
patch -p1 < /pgvector.patch && \
make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/vector.control
#########################################################################################
@@ -266,7 +263,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
# 9742dab1b2f297ad3811120db7b21451bca2d3c9 made on 13/11/2021
RUN wget https://github.com/michelp/pgjwt/archive/9742dab1b2f297ad3811120db7b21451bca2d3c9.tar.gz -O pgjwt.tar.gz && \
echo "cfdefb15007286f67d3d45510f04a6a7a495004be5b3aecb12cda667e774203f pgjwt.tar.gz" | sha256sum --check && \
mkdir pgjwt-src && cd pgjwt-src && tar xzf ../pgjwt.tar.gz --strip-components=1 -C . && \
mkdir pgjwt-src && cd pgjwt-src && tar xvzf ../pgjwt.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pgjwt.control
@@ -281,7 +278,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/HypoPG/hypopg/archive/refs/tags/1.4.0.tar.gz -O hypopg.tar.gz && \
echo "0821011743083226fc9b813c1f2ef5897a91901b57b6bea85a78e466187c6819 hypopg.tar.gz" | sha256sum --check && \
mkdir hypopg-src && cd hypopg-src && tar xzf ../hypopg.tar.gz --strip-components=1 -C . && \
mkdir hypopg-src && cd hypopg-src && tar xvzf ../hypopg.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/hypopg.control
@@ -297,7 +294,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/iCyberon/pg_hashids/archive/refs/tags/v1.2.1.tar.gz -O pg_hashids.tar.gz && \
echo "74576b992d9277c92196dd8d816baa2cc2d8046fe102f3dcd7f3c3febed6822a pg_hashids.tar.gz" | sha256sum --check && \
mkdir pg_hashids-src && cd pg_hashids-src && tar xzf ../pg_hashids.tar.gz --strip-components=1 -C . && \
mkdir pg_hashids-src && cd pg_hashids-src && tar xvzf ../pg_hashids.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pg_hashids.control
@@ -313,7 +310,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/postgrespro/rum/archive/refs/tags/1.3.13.tar.gz -O rum.tar.gz && \
echo "6ab370532c965568df6210bd844ac6ba649f53055e48243525b0b7e5c4d69a7d rum.tar.gz" | sha256sum --check && \
mkdir rum-src && cd rum-src && tar xzf ../rum.tar.gz --strip-components=1 -C . && \
mkdir rum-src && cd rum-src && tar xvzf ../rum.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/rum.control
@@ -329,7 +326,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/theory/pgtap/archive/refs/tags/v1.2.0.tar.gz -O pgtap.tar.gz && \
echo "9c7c3de67ea41638e14f06da5da57bac6f5bd03fea05c165a0ec862205a5c052 pgtap.tar.gz" | sha256sum --check && \
mkdir pgtap-src && cd pgtap-src && tar xzf ../pgtap.tar.gz --strip-components=1 -C . && \
mkdir pgtap-src && cd pgtap-src && tar xvzf ../pgtap.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pgtap.control
@@ -345,7 +342,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/RhodiumToad/ip4r/archive/refs/tags/2.4.2.tar.gz -O ip4r.tar.gz && \
echo "0f7b1f159974f49a47842a8ab6751aecca1ed1142b6d5e38d81b064b2ead1b4b ip4r.tar.gz" | sha256sum --check && \
mkdir ip4r-src && cd ip4r-src && tar xzf ../ip4r.tar.gz --strip-components=1 -C . && \
mkdir ip4r-src && cd ip4r-src && tar xvzf ../ip4r.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/ip4r.control
@@ -361,7 +358,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/dimitri/prefix/archive/refs/tags/v1.2.10.tar.gz -O prefix.tar.gz && \
echo "4342f251432a5f6fb05b8597139d3ccde8dcf87e8ca1498e7ee931ca057a8575 prefix.tar.gz" | sha256sum --check && \
mkdir prefix-src && cd prefix-src && tar xzf ../prefix.tar.gz --strip-components=1 -C . && \
mkdir prefix-src && cd prefix-src && tar xvzf ../prefix.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/prefix.control
@@ -377,7 +374,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/citusdata/postgresql-hll/archive/refs/tags/v2.18.tar.gz -O hll.tar.gz && \
echo "e2f55a6f4c4ab95ee4f1b4a2b73280258c5136b161fe9d059559556079694f0e hll.tar.gz" | sha256sum --check && \
mkdir hll-src && cd hll-src && tar xzf ../hll.tar.gz --strip-components=1 -C . && \
mkdir hll-src && cd hll-src && tar xvzf ../hll.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/hll.control
@@ -393,7 +390,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
RUN wget https://github.com/okbob/plpgsql_check/archive/refs/tags/v2.5.3.tar.gz -O plpgsql_check.tar.gz && \
echo "6631ec3e7fb3769eaaf56e3dfedb829aa761abf163d13dba354b4c218508e1c0 plpgsql_check.tar.gz" | sha256sum --check && \
mkdir plpgsql_check-src && cd plpgsql_check-src && tar xzf ../plpgsql_check.tar.gz --strip-components=1 -C . && \
mkdir plpgsql_check-src && cd plpgsql_check-src && tar xvzf ../plpgsql_check.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/plpgsql_check.control
@@ -424,7 +421,7 @@ RUN case "${PG_VERSION}" in \
apt-get install -y cmake && \
wget https://github.com/timescale/timescaledb/archive/refs/tags/${TIMESCALEDB_VERSION}.tar.gz -O timescaledb.tar.gz && \
echo "${TIMESCALEDB_CHECKSUM} timescaledb.tar.gz" | sha256sum --check && \
mkdir timescaledb-src && cd timescaledb-src && tar xzf ../timescaledb.tar.gz --strip-components=1 -C . && \
mkdir timescaledb-src && cd timescaledb-src && tar xvzf ../timescaledb.tar.gz --strip-components=1 -C . && \
./bootstrap -DSEND_TELEMETRY_DEFAULT:BOOL=OFF -DUSE_TELEMETRY:BOOL=OFF -DAPACHE_ONLY:BOOL=ON -DCMAKE_BUILD_TYPE=Release && \
cd build && \
make -j $(getconf _NPROCESSORS_ONLN) && \
@@ -462,11 +459,36 @@ RUN case "${PG_VERSION}" in \
esac && \
wget https://github.com/ossc-db/pg_hint_plan/archive/refs/tags/REL${PG_HINT_PLAN_VERSION}.tar.gz -O pg_hint_plan.tar.gz && \
echo "${PG_HINT_PLAN_CHECKSUM} pg_hint_plan.tar.gz" | sha256sum --check && \
mkdir pg_hint_plan-src && cd pg_hint_plan-src && tar xzf ../pg_hint_plan.tar.gz --strip-components=1 -C . && \
mkdir pg_hint_plan-src && cd pg_hint_plan-src && tar xvzf ../pg_hint_plan.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
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 . && \
find /usr/local/pgsql -type f | sed 's|^/usr/local/pgsql/||' > /before.txt &&\
mkdir build && cd build && \
cmake -DCMAKE_BUILD_TYPE=Release .. && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/kq_imcx.control && \
find /usr/local/pgsql -type f | sed 's|^/usr/local/pgsql/||' > /after.txt &&\
mkdir -p /extensions/kq_imcx && cp /usr/local/pgsql/share/extension/kq_imcx.control /extensions/kq_imcx && \
sort -o /before.txt /before.txt && sort -o /after.txt /after.txt && \
comm -13 /before.txt /after.txt | tar --directory=/usr/local/pgsql --zstd -cf /extensions/kq_imcx.tar.zst -T -
#########################################################################################
#
@@ -480,7 +502,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/citusdata/pg_cron/archive/refs/tags/v1.6.0.tar.gz -O pg_cron.tar.gz && \
echo "383a627867d730222c272bfd25cd5e151c578d73f696d32910c7db8c665cc7db pg_cron.tar.gz" | sha256sum --check && \
mkdir pg_cron-src && cd pg_cron-src && tar xzf ../pg_cron.tar.gz --strip-components=1 -C . && \
mkdir pg_cron-src && cd pg_cron-src && tar xvzf ../pg_cron.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pg_cron.control
@@ -506,7 +528,7 @@ RUN apt-get update && \
ENV PATH "/usr/local/pgsql/bin/:/usr/local/pgsql/:$PATH"
RUN wget https://github.com/rdkit/rdkit/archive/refs/tags/Release_2023_03_3.tar.gz -O rdkit.tar.gz && \
echo "bdbf9a2e6988526bfeb8c56ce3cdfe2998d60ac289078e2215374288185e8c8d rdkit.tar.gz" | sha256sum --check && \
mkdir rdkit-src && cd rdkit-src && tar xzf ../rdkit.tar.gz --strip-components=1 -C . && \
mkdir rdkit-src && cd rdkit-src && tar xvzf ../rdkit.tar.gz --strip-components=1 -C . && \
cmake \
-D RDK_BUILD_CAIRO_SUPPORT=OFF \
-D RDK_BUILD_INCHI_SUPPORT=ON \
@@ -546,7 +568,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/fboulnois/pg_uuidv7/archive/refs/tags/v1.0.1.tar.gz -O pg_uuidv7.tar.gz && \
echo "0d0759ab01b7fb23851ecffb0bce27822e1868a4a5819bfd276101c716637a7a pg_uuidv7.tar.gz" | sha256sum --check && \
mkdir pg_uuidv7-src && cd pg_uuidv7-src && tar xzf ../pg_uuidv7.tar.gz --strip-components=1 -C . && \
mkdir pg_uuidv7-src && cd pg_uuidv7-src && tar xvzf ../pg_uuidv7.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pg_uuidv7.control
@@ -563,7 +585,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/ChenHuajun/pg_roaringbitmap/archive/refs/tags/v0.5.4.tar.gz -O pg_roaringbitmap.tar.gz && \
echo "b75201efcb1c2d1b014ec4ae6a22769cc7a224e6e406a587f5784a37b6b5a2aa pg_roaringbitmap.tar.gz" | sha256sum --check && \
mkdir pg_roaringbitmap-src && cd pg_roaringbitmap-src && tar xzf ../pg_roaringbitmap.tar.gz --strip-components=1 -C . && \
mkdir pg_roaringbitmap-src && cd pg_roaringbitmap-src && tar xvzf ../pg_roaringbitmap.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/roaringbitmap.control
@@ -580,7 +602,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/theory/pg-semver/archive/refs/tags/v0.32.1.tar.gz -O pg_semver.tar.gz && \
echo "fbdaf7512026d62eec03fad8687c15ed509b6ba395bff140acd63d2e4fbe25d7 pg_semver.tar.gz" | sha256sum --check && \
mkdir pg_semver-src && cd pg_semver-src && tar xzf ../pg_semver.tar.gz --strip-components=1 -C . && \
mkdir pg_semver-src && cd pg_semver-src && tar xvzf ../pg_semver.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/semver.control
@@ -606,7 +628,7 @@ RUN case "${PG_VERSION}" in \
esac && \
wget https://github.com/neondatabase/pg_embedding/archive/refs/tags/${PG_EMBEDDING_VERSION}.tar.gz -O pg_embedding.tar.gz && \
echo "${PG_EMBEDDING_CHECKSUM} pg_embedding.tar.gz" | sha256sum --check && \
mkdir pg_embedding-src && cd pg_embedding-src && tar xzf ../pg_embedding.tar.gz --strip-components=1 -C . && \
mkdir pg_embedding-src && cd pg_embedding-src && tar xvzf ../pg_embedding.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install
@@ -622,7 +644,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/neondatabase/postgresql_anonymizer/archive/refs/tags/neon_1.1.1.tar.gz -O pg_anon.tar.gz && \
echo "321ea8d5c1648880aafde850a2c576e4a9e7b9933a34ce272efc839328999fa9 pg_anon.tar.gz" | sha256sum --check && \
mkdir pg_anon-src && cd pg_anon-src && tar xzf ../pg_anon.tar.gz --strip-components=1 -C . && \
mkdir pg_anon-src && cd pg_anon-src && tar xvzf ../pg_anon.tar.gz --strip-components=1 -C . && \
find /usr/local/pgsql -type f | sed 's|^/usr/local/pgsql/||' > /before.txt &&\
make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/anon.control && \
@@ -671,7 +693,7 @@ ARG PG_VERSION
RUN wget https://github.com/supabase/pg_jsonschema/archive/refs/tags/v0.2.0.tar.gz -O pg_jsonschema.tar.gz && \
echo "9118fc508a6e231e7a39acaa6f066fcd79af17a5db757b47d2eefbe14f7794f0 pg_jsonschema.tar.gz" | sha256sum --check && \
mkdir pg_jsonschema-src && cd pg_jsonschema-src && tar xzf ../pg_jsonschema.tar.gz --strip-components=1 -C . && \
mkdir pg_jsonschema-src && cd pg_jsonschema-src && tar xvzf ../pg_jsonschema.tar.gz --strip-components=1 -C . && \
sed -i 's/pgrx = "0.10.2"/pgrx = { version = "0.10.2", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
cargo pgrx install --release && \
echo "trusted = true" >> /usr/local/pgsql/share/extension/pg_jsonschema.control
@@ -688,7 +710,7 @@ ARG PG_VERSION
RUN wget https://github.com/supabase/pg_graphql/archive/refs/tags/v1.4.0.tar.gz -O pg_graphql.tar.gz && \
echo "bd8dc7230282b3efa9ae5baf053a54151ed0e66881c7c53750e2d0c765776edc pg_graphql.tar.gz" | sha256sum --check && \
mkdir pg_graphql-src && cd pg_graphql-src && tar xzf ../pg_graphql.tar.gz --strip-components=1 -C . && \
mkdir pg_graphql-src && cd pg_graphql-src && tar xvzf ../pg_graphql.tar.gz --strip-components=1 -C . && \
sed -i 's/pgrx = "=0.10.2"/pgrx = { version = "0.10.2", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
cargo pgrx install --release && \
# it's needed to enable extension because it uses untrusted C language
@@ -708,7 +730,7 @@ ARG PG_VERSION
# 26806147b17b60763039c6a6878884c41a262318 made on 26/09/2023
RUN wget https://github.com/kelvich/pg_tiktoken/archive/26806147b17b60763039c6a6878884c41a262318.tar.gz -O pg_tiktoken.tar.gz && \
echo "e64e55aaa38c259512d3e27c572da22c4637418cf124caba904cd50944e5004e pg_tiktoken.tar.gz" | sha256sum --check && \
mkdir pg_tiktoken-src && cd pg_tiktoken-src && tar xzf ../pg_tiktoken.tar.gz --strip-components=1 -C . && \
mkdir pg_tiktoken-src && cd pg_tiktoken-src && tar xvzf ../pg_tiktoken.tar.gz --strip-components=1 -C . && \
cargo pgrx install --release && \
echo "trusted = true" >> /usr/local/pgsql/share/extension/pg_tiktoken.control
@@ -724,7 +746,7 @@ ARG PG_VERSION
RUN wget https://github.com/pksunkara/pgx_ulid/archive/refs/tags/v0.1.3.tar.gz -O pgx_ulid.tar.gz && \
echo "ee5db82945d2d9f2d15597a80cf32de9dca67b897f605beb830561705f12683c pgx_ulid.tar.gz" | sha256sum --check && \
mkdir pgx_ulid-src && cd pgx_ulid-src && tar xzf ../pgx_ulid.tar.gz --strip-components=1 -C . && \
mkdir pgx_ulid-src && cd pgx_ulid-src && tar xvzf ../pgx_ulid.tar.gz --strip-components=1 -C . && \
echo "******************* Apply a patch for Postgres 16 support; delete in the next release ******************" && \
wget https://github.com/pksunkara/pgx_ulid/commit/f84954cf63fc8c80d964ac970d9eceed3c791196.patch && \
patch -p1 < f84954cf63fc8c80d964ac970d9eceed3c791196.patch && \
@@ -746,7 +768,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/eulerto/wal2json/archive/refs/tags/wal2json_2_5.tar.gz && \
echo "b516653575541cf221b99cf3f8be9b6821f6dbcfc125675c85f35090f824f00e wal2json_2_5.tar.gz" | sha256sum --check && \
mkdir wal2json-src && cd wal2json-src && tar xzf ../wal2json_2_5.tar.gz --strip-components=1 -C . && \
mkdir wal2json-src && cd wal2json-src && tar xvzf ../wal2json_2_5.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install
@@ -762,7 +784,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/sraoss/pg_ivm/archive/refs/tags/v1.7.tar.gz -O pg_ivm.tar.gz && \
echo "ebfde04f99203c7be4b0e873f91104090e2e83e5429c32ac242d00f334224d5e pg_ivm.tar.gz" | sha256sum --check && \
mkdir pg_ivm-src && cd pg_ivm-src && tar xzf ../pg_ivm.tar.gz --strip-components=1 -C . && \
mkdir pg_ivm-src && cd pg_ivm-src && tar xvzf ../pg_ivm.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pg_ivm.control
@@ -779,7 +801,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
ENV PATH "/usr/local/pgsql/bin/:$PATH"
RUN wget https://github.com/pgpartman/pg_partman/archive/refs/tags/v5.0.1.tar.gz -O pg_partman.tar.gz && \
echo "75b541733a9659a6c90dbd40fccb904a630a32880a6e3044d0c4c5f4c8a65525 pg_partman.tar.gz" | sha256sum --check && \
mkdir pg_partman-src && cd pg_partman-src && tar xzf ../pg_partman.tar.gz --strip-components=1 -C . && \
mkdir pg_partman-src && cd pg_partman-src && tar xvzf ../pg_partman.tar.gz --strip-components=1 -C . && \
make -j $(getconf _NPROCESSORS_ONLN) && \
make -j $(getconf _NPROCESSORS_ONLN) install && \
echo 'trusted = true' >> /usr/local/pgsql/share/extension/pg_partman.control
@@ -815,6 +837,7 @@ 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 --from=pg-cron-pg-build /usr/local/pgsql/ /usr/local/pgsql/
COPY --from=pg-pgx-ulid-build /usr/local/pgsql/ /usr/local/pgsql/
COPY --from=rdkit-pg-build /usr/local/pgsql/ /usr/local/pgsql/
@@ -902,68 +925,6 @@ RUN rm -r /usr/local/pgsql/include
# if they were to be used by other libraries.
RUN rm /usr/local/pgsql/lib/lib*.a
#########################################################################################
#
# Layer neon-pg-ext-test
#
#########################################################################################
FROM neon-pg-ext-build AS neon-pg-ext-test
ARG PG_VERSION
RUN mkdir /ext-src
#COPY --from=postgis-build /postgis.tar.gz /ext-src/
#COPY --from=postgis-build /sfcgal/* /usr
COPY --from=plv8-build /plv8.tar.gz /ext-src/
COPY --from=h3-pg-build /h3-pg.tar.gz /ext-src/
COPY --from=unit-pg-build /postgresql-unit.tar.gz /ext-src/
COPY --from=vector-pg-build /pgvector.tar.gz /ext-src/
COPY --from=vector-pg-build /pgvector.patch /ext-src/
COPY --from=pgjwt-pg-build /pgjwt.tar.gz /ext-src
#COPY --from=pg-jsonschema-pg-build /home/nonroot/pg_jsonschema.tar.gz /ext-src
#COPY --from=pg-graphql-pg-build /home/nonroot/pg_graphql.tar.gz /ext-src
#COPY --from=pg-tiktoken-pg-build /home/nonroot/pg_tiktoken.tar.gz /ext-src
COPY --from=hypopg-pg-build /hypopg.tar.gz /ext-src
COPY --from=pg-hashids-pg-build /pg_hashids.tar.gz /ext-src
#COPY --from=rum-pg-build /rum.tar.gz /ext-src
#COPY --from=pgtap-pg-build /pgtap.tar.gz /ext-src
COPY --from=ip4r-pg-build /ip4r.tar.gz /ext-src
COPY --from=prefix-pg-build /prefix.tar.gz /ext-src
COPY --from=hll-pg-build /hll.tar.gz /ext-src
COPY --from=plpgsql-check-pg-build /plpgsql_check.tar.gz /ext-src
#COPY --from=timescaledb-pg-build /timescaledb.tar.gz /ext-src
COPY --from=pg-hint-plan-pg-build /pg_hint_plan.tar.gz /ext-src
COPY patches/pg_hintplan.patch /ext-src
COPY --from=pg-cron-pg-build /pg_cron.tar.gz /ext-src
COPY patches/pg_cron.patch /ext-src
#COPY --from=pg-pgx-ulid-build /home/nonroot/pgx_ulid.tar.gz /ext-src
COPY --from=rdkit-pg-build /rdkit.tar.gz /ext-src
COPY --from=pg-uuidv7-pg-build /pg_uuidv7.tar.gz /ext-src
COPY --from=pg-roaringbitmap-pg-build /pg_roaringbitmap.tar.gz /ext-src
COPY --from=pg-semver-pg-build /pg_semver.tar.gz /ext-src
#COPY --from=pg-embedding-pg-build /home/nonroot/pg_embedding-src/ /ext-src
#COPY --from=wal2json-pg-build /wal2json_2_5.tar.gz /ext-src
COPY --from=pg-anon-pg-build /pg_anon.tar.gz /ext-src
COPY patches/pg_anon.patch /ext-src
COPY --from=pg-ivm-build /pg_ivm.tar.gz /ext-src
COPY --from=pg-partman-build /pg_partman.tar.gz /ext-src
RUN cd /ext-src/ && for f in *.tar.gz; \
do echo $f; dname=$(echo $f | sed 's/\.tar.*//')-src; \
rm -rf $dname; mkdir $dname; tar xzf $f --strip-components=1 -C $dname \
|| exit 1; rm -f $f; done
RUN cd /ext-src/pgvector-src && patch -p1 <../pgvector.patch
# cmake is required for the h3 test
RUN apt-get update && apt-get install -y cmake
RUN patch -p1 < /ext-src/pg_hintplan.patch
COPY --chmod=755 docker-compose/run-tests.sh /run-tests.sh
RUN patch -p1 </ext-src/pg_anon.patch
RUN patch -p1 </ext-src/pg_cron.patch
ENV PATH=/usr/local/pgsql/bin:$PATH
ENV PGHOST=compute
ENV PGPORT=55433
ENV PGUSER=cloud_admin
ENV PGDATABASE=postgres
#########################################################################################
#
# Final layer

View File

@@ -3,9 +3,6 @@ ROOT_PROJECT_DIR := $(dir $(abspath $(lastword $(MAKEFILE_LIST))))
# Where to install Postgres, default is ./pg_install, maybe useful for package managers
POSTGRES_INSTALL_DIR ?= $(ROOT_PROJECT_DIR)/pg_install/
OPENSSL_PREFIX_DIR := /usr/local/openssl
ICU_PREFIX_DIR := /usr/local/icu
#
# We differentiate between release / debug build types using the BUILD_TYPE
# environment variable.
@@ -23,16 +20,6 @@ else
$(error Bad build type '$(BUILD_TYPE)', see Makefile for options)
endif
ifeq ($(shell test -e /home/nonroot/.docker_build && echo -n yes),yes)
# Exclude static build openssl, icu for local build (MacOS, Linux)
# Only keep for build type release and debug
PG_CFLAGS += -I$(OPENSSL_PREFIX_DIR)/include
PG_CONFIGURE_OPTS += --with-icu
PG_CONFIGURE_OPTS += ICU_CFLAGS='-I/$(ICU_PREFIX_DIR)/include -DU_STATIC_IMPLEMENTATION'
PG_CONFIGURE_OPTS += ICU_LIBS='-L$(ICU_PREFIX_DIR)/lib -L$(ICU_PREFIX_DIR)/lib64 -licui18n -licuuc -licudata -lstdc++ -Wl,-Bdynamic -lm'
PG_CONFIGURE_OPTS += LDFLAGS='-L$(OPENSSL_PREFIX_DIR)/lib -L$(OPENSSL_PREFIX_DIR)/lib64 -L$(ICU_PREFIX_DIR)/lib -L$(ICU_PREFIX_DIR)/lib64 -Wl,-Bstatic -lssl -lcrypto -Wl,-Bdynamic -lrt -lm -ldl -lpthread'
endif
UNAME_S := $(shell uname -s)
ifeq ($(UNAME_S),Linux)
# Seccomp BPF is only available for Linux
@@ -41,7 +28,7 @@ else ifeq ($(UNAME_S),Darwin)
ifndef DISABLE_HOMEBREW
# macOS with brew-installed openssl requires explicit paths
# It can be configured with OPENSSL_PREFIX variable
OPENSSL_PREFIX := $(shell brew --prefix openssl@3)
OPENSSL_PREFIX ?= $(shell brew --prefix openssl@3)
PG_CONFIGURE_OPTS += --with-includes=$(OPENSSL_PREFIX)/include --with-libraries=$(OPENSSL_PREFIX)/lib
PG_CONFIGURE_OPTS += PKG_CONFIG_PATH=$(shell brew --prefix icu4c)/lib/pkgconfig
# macOS already has bison and flex in the system, but they are old and result in postgres-v14 target failure
@@ -137,8 +124,6 @@ postgres-%: postgres-configure-% \
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/pageinspect install
+@echo "Compiling amcheck $*"
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/amcheck install
+@echo "Compiling test_decoding $*"
$(MAKE) -C $(POSTGRES_INSTALL_DIR)/build/$*/contrib/test_decoding install
.PHONY: postgres-clean-%
postgres-clean-%:

View File

@@ -735,7 +735,7 @@ fn cli() -> clap::Command {
Arg::new("filecache-connstr")
.long("filecache-connstr")
.default_value(
"host=localhost port=5432 dbname=postgres user=cloud_admin sslmode=disable application_name=vm-monitor",
"host=localhost port=5432 dbname=postgres user=cloud_admin sslmode=disable",
)
.value_name("FILECACHE_CONNSTR"),
)

View File

@@ -918,39 +918,38 @@ impl ComputeNode {
// temporarily reset max_cluster_size in config
// to avoid the possibility of hitting the limit, while we are reconfiguring:
// creating new extensions, roles, etc...
config::with_compute_ctl_tmp_override(pgdata_path, "neon.max_cluster_size=-1", || {
self.pg_reload_conf()?;
config::compute_ctl_temp_override_create(pgdata_path, "neon.max_cluster_size=-1")?;
self.pg_reload_conf()?;
let mut client = Client::connect(self.connstr.as_str(), NoTls)?;
let mut client = Client::connect(self.connstr.as_str(), NoTls)?;
// Proceed with post-startup configuration. Note, that order of operations is important.
// Disable DDL forwarding because control plane already knows about these roles/databases.
if spec.mode == ComputeMode::Primary {
client.simple_query("SET neon.forward_ddl = false")?;
cleanup_instance(&mut client)?;
handle_roles(&spec, &mut client)?;
handle_databases(&spec, &mut client)?;
handle_role_deletions(&spec, self.connstr.as_str(), &mut client)?;
handle_grants(
&spec,
&mut client,
self.connstr.as_str(),
self.has_feature(ComputeFeature::AnonExtension),
)?;
handle_extensions(&spec, &mut client)?;
handle_extension_neon(&mut client)?;
// We can skip handle_migrations here because a new migration can only appear
// if we have a new version of the compute_ctl binary, which can only happen
// if compute got restarted, in which case we'll end up inside of apply_config
// instead of reconfigure.
}
// Proceed with post-startup configuration. Note, that order of operations is important.
// Disable DDL forwarding because control plane already knows about these roles/databases.
if spec.mode == ComputeMode::Primary {
client.simple_query("SET neon.forward_ddl = false")?;
cleanup_instance(&mut client)?;
handle_roles(&spec, &mut client)?;
handle_databases(&spec, &mut client)?;
handle_role_deletions(&spec, self.connstr.as_str(), &mut client)?;
handle_grants(
&spec,
&mut client,
self.connstr.as_str(),
self.has_feature(ComputeFeature::AnonExtension),
)?;
handle_extensions(&spec, &mut client)?;
handle_extension_neon(&mut client)?;
// We can skip handle_migrations here because a new migration can only appear
// if we have a new version of the compute_ctl binary, which can only happen
// if compute got restarted, in which case we'll end up inside of apply_config
// instead of reconfigure.
}
// 'Close' connection
drop(client);
Ok(())
})?;
// 'Close' connection
drop(client);
// reset max_cluster_size in config back to original value and reload config
config::compute_ctl_temp_override_remove(pgdata_path)?;
self.pg_reload_conf()?;
let unknown_op = "unknown".to_string();
@@ -1041,17 +1040,12 @@ impl ComputeNode {
// temporarily reset max_cluster_size in config
// to avoid the possibility of hitting the limit, while we are applying config:
// creating new extensions, roles, etc...
config::with_compute_ctl_tmp_override(
pgdata_path,
"neon.max_cluster_size=-1",
|| {
self.pg_reload_conf()?;
config::compute_ctl_temp_override_create(pgdata_path, "neon.max_cluster_size=-1")?;
self.pg_reload_conf()?;
self.apply_config(&compute_state)?;
self.apply_config(&compute_state)?;
Ok(())
},
)?;
config::compute_ctl_temp_override_remove(pgdata_path)?;
self.pg_reload_conf()?;
}
self.post_apply_config()?;

View File

@@ -83,6 +83,12 @@ pub fn write_postgres_conf(
ComputeMode::Replica => {
// hot_standby is 'on' by default, but let's be explicit
writeln!(file, "hot_standby=on")?;
// Inform the replica about the primary state
// Default is 'false'
if let Some(primary_is_running) = spec.primary_is_running {
writeln!(file, "neon.primary_is_running={}", primary_is_running)?;
}
}
}
@@ -125,17 +131,18 @@ pub fn write_postgres_conf(
Ok(())
}
pub fn with_compute_ctl_tmp_override<F>(pgdata_path: &Path, options: &str, exec: F) -> Result<()>
where
F: FnOnce() -> Result<()>,
{
/// create file compute_ctl_temp_override.conf in pgdata_dir
/// add provided options to this file
pub fn compute_ctl_temp_override_create(pgdata_path: &Path, options: &str) -> Result<()> {
let path = pgdata_path.join("compute_ctl_temp_override.conf");
let mut file = File::create(path)?;
write!(file, "{}", options)?;
let res = exec();
file.set_len(0)?;
res
Ok(())
}
/// remove file compute_ctl_temp_override.conf in pgdata_dir
pub fn compute_ctl_temp_override_remove(pgdata_path: &Path) -> Result<()> {
let path = pgdata_path.join("compute_ctl_temp_override.conf");
std::fs::remove_file(path)?;
Ok(())
}

View File

@@ -17,7 +17,7 @@ use hyper::header::CONTENT_TYPE;
use hyper::service::{make_service_fn, service_fn};
use hyper::{Body, Method, Request, Response, Server, StatusCode};
use tokio::task;
use tracing::{debug, error, info, warn};
use tracing::{error, info, warn};
use tracing_utils::http::OtelName;
use utils::http::request::must_get_query_param;
@@ -48,7 +48,7 @@ async fn routes(req: Request<Body>, compute: &Arc<ComputeNode>) -> Response<Body
match (req.method(), req.uri().path()) {
// Serialized compute state.
(&Method::GET, "/status") => {
debug!("serving /status GET request");
info!("serving /status GET request");
let state = compute.state.lock().unwrap();
let status_response = status_response_from_state(&state);
Response::new(Body::from(serde_json::to_string(&status_response).unwrap()))

View File

@@ -1 +0,0 @@
ALTER ROLE neon_superuser BYPASSRLS;

View File

@@ -1,18 +0,0 @@
DO $$
DECLARE
role_name text;
BEGIN
FOR role_name IN SELECT rolname FROM pg_roles WHERE pg_has_role(rolname, 'neon_superuser', 'member')
LOOP
RAISE NOTICE 'EXECUTING ALTER ROLE % INHERIT', quote_ident(role_name);
EXECUTE 'ALTER ROLE ' || quote_ident(role_name) || ' INHERIT';
END LOOP;
FOR role_name IN SELECT rolname FROM pg_roles
WHERE
NOT pg_has_role(rolname, 'neon_superuser', 'member') AND NOT starts_with(rolname, 'pg_')
LOOP
RAISE NOTICE 'EXECUTING ALTER ROLE % NOBYPASSRLS', quote_ident(role_name);
EXECUTE 'ALTER ROLE ' || quote_ident(role_name) || ' NOBYPASSRLS';
END LOOP;
END $$;

View File

@@ -1,6 +0,0 @@
DO $$
BEGIN
IF (SELECT setting::numeric >= 160000 FROM pg_settings WHERE name = 'server_version_num') THEN
EXECUTE 'GRANT pg_create_subscription TO neon_superuser';
END IF;
END $$;

View File

@@ -1 +0,0 @@
GRANT pg_monitor TO neon_superuser WITH ADMIN OPTION;

View File

@@ -1,4 +0,0 @@
-- SKIP: Deemed insufficient for allowing relations created by extensions to be
-- interacted with by neon_superuser without permission issues.
ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT ALL ON TABLES TO neon_superuser;

View File

@@ -1,4 +0,0 @@
-- SKIP: Deemed insufficient for allowing relations created by extensions to be
-- interacted with by neon_superuser without permission issues.
ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT ALL ON SEQUENCES TO neon_superuser;

View File

@@ -1,3 +0,0 @@
-- SKIP: Moved inline to the handle_grants() functions.
ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT ALL ON TABLES TO neon_superuser WITH GRANT OPTION;

View File

@@ -1,3 +0,0 @@
-- SKIP: Moved inline to the handle_grants() functions.
ALTER DEFAULT PRIVILEGES IN SCHEMA public GRANT ALL ON SEQUENCES TO neon_superuser WITH GRANT OPTION;

View File

@@ -1,13 +0,0 @@
-- SKIP: The original goal of this migration was to prevent creating
-- subscriptions, but this migration was insufficient.
DO $$
DECLARE
role_name TEXT;
BEGIN
FOR role_name IN SELECT rolname FROM pg_roles WHERE rolreplication IS TRUE
LOOP
RAISE NOTICE 'EXECUTING ALTER ROLE % NOREPLICATION', quote_ident(role_name);
EXECUTE 'ALTER ROLE ' || quote_ident(role_name) || ' NOREPLICATION';
END LOOP;
END $$;

View File

@@ -774,21 +774,44 @@ pub fn handle_migrations(client: &mut Client) -> Result<()> {
// !BE SURE TO ONLY ADD MIGRATIONS TO THE END OF THIS ARRAY. IF YOU DO NOT, VERY VERY BAD THINGS MAY HAPPEN!
// !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
// Add new migrations in numerical order.
let migrations = [
include_str!("./migrations/0000-neon_superuser_bypass_rls.sql"),
include_str!("./migrations/0001-alter_roles.sql"),
include_str!("./migrations/0002-grant_pg_create_subscription_to_neon_superuser.sql"),
include_str!("./migrations/0003-grant_pg_monitor_to_neon_superuser.sql"),
include_str!("./migrations/0004-grant_all_on_tables_to_neon_superuser.sql"),
include_str!("./migrations/0005-grant_all_on_sequences_to_neon_superuser.sql"),
include_str!(
"./migrations/0006-grant_all_on_tables_to_neon_superuser_with_grant_option.sql"
),
include_str!(
"./migrations/0007-grant_all_on_sequences_to_neon_superuser_with_grant_option.sql"
),
include_str!("./migrations/0008-revoke_replication_for_previously_allowed_roles.sql"),
"ALTER ROLE neon_superuser BYPASSRLS",
r#"
DO $$
DECLARE
role_name text;
BEGIN
FOR role_name IN SELECT rolname FROM pg_roles WHERE pg_has_role(rolname, 'neon_superuser', 'member')
LOOP
RAISE NOTICE 'EXECUTING ALTER ROLE % INHERIT', quote_ident(role_name);
EXECUTE 'ALTER ROLE ' || quote_ident(role_name) || ' INHERIT';
END LOOP;
FOR role_name IN SELECT rolname FROM pg_roles
WHERE
NOT pg_has_role(rolname, 'neon_superuser', 'member') AND NOT starts_with(rolname, 'pg_')
LOOP
RAISE NOTICE 'EXECUTING ALTER ROLE % NOBYPASSRLS', quote_ident(role_name);
EXECUTE 'ALTER ROLE ' || quote_ident(role_name) || ' NOBYPASSRLS';
END LOOP;
END $$;
"#,
r#"
DO $$
BEGIN
IF (SELECT setting::numeric >= 160000 FROM pg_settings WHERE name = 'server_version_num') THEN
EXECUTE 'GRANT pg_create_subscription TO neon_superuser';
END IF;
END
$$;"#,
"GRANT pg_monitor TO neon_superuser WITH ADMIN OPTION",
// Don't remove: these are some SQLs that we originally applied in migrations but turned out to execute somewhere else.
"",
"",
"",
"",
"",
// Add new migrations below.
];
let mut func = || {
@@ -824,13 +847,10 @@ pub fn handle_migrations(client: &mut Client) -> Result<()> {
while current_migration < migrations.len() {
let migration = &migrations[current_migration];
if migration.starts_with("-- SKIP") {
info!("Skipping migration id={}", current_migration);
if migration.is_empty() {
info!("Skip migration id={}", current_migration);
} else {
info!(
"Running migration id={}:\n{}\n",
current_migration, migration
);
info!("Running migration:\n{}\n", migration);
client.simple_query(migration).with_context(|| {
format!("handle_migrations current_migration={}", current_migration)
})?;

View File

@@ -36,11 +36,11 @@ use utils::pid_file::{self, PidFileRead};
// it's waiting. If the process hasn't started/stopped after 5 seconds,
// it prints a notice that it's taking long, but keeps waiting.
//
const STOP_RETRY_TIMEOUT: Duration = Duration::from_secs(10);
const STOP_RETRIES: u128 = STOP_RETRY_TIMEOUT.as_millis() / RETRY_INTERVAL.as_millis();
const RETRY_INTERVAL: Duration = Duration::from_millis(100);
const DOT_EVERY_RETRIES: u128 = 10;
const NOTICE_AFTER_RETRIES: u128 = 50;
const RETRY_UNTIL_SECS: u64 = 10;
const RETRIES: u64 = (RETRY_UNTIL_SECS * 1000) / RETRY_INTERVAL_MILLIS;
const RETRY_INTERVAL_MILLIS: u64 = 100;
const DOT_EVERY_RETRIES: u64 = 10;
const NOTICE_AFTER_RETRIES: u64 = 50;
/// Argument to `start_process`, to indicate whether it should create pidfile or if the process creates
/// it itself.
@@ -52,7 +52,6 @@ pub enum InitialPidFile {
}
/// Start a background child process using the parameters given.
#[allow(clippy::too_many_arguments)]
pub async fn start_process<F, Fut, AI, A, EI>(
process_name: &str,
datadir: &Path,
@@ -60,7 +59,6 @@ pub async fn start_process<F, Fut, AI, A, EI>(
args: AI,
envs: EI,
initial_pid_file: InitialPidFile,
retry_timeout: &Duration,
process_status_check: F,
) -> anyhow::Result<()>
where
@@ -71,10 +69,6 @@ where
// Not generic AsRef<OsStr>, otherwise empty `envs` prevents type inference
EI: IntoIterator<Item = (String, String)>,
{
let retries: u128 = retry_timeout.as_millis() / RETRY_INTERVAL.as_millis();
if !datadir.metadata().context("stat datadir")?.is_dir() {
anyhow::bail!("`datadir` must be a directory when calling this function: {datadir:?}");
}
let log_path = datadir.join(format!("{process_name}.log"));
let process_log_file = fs::OpenOptions::new()
.create(true)
@@ -91,13 +85,7 @@ where
let background_command = command
.stdout(process_log_file)
.stderr(same_file_for_stderr)
.args(args)
// spawn all child processes in their datadir, useful for all kinds of things,
// not least cleaning up child processes e.g. after an unclean exit from the test suite:
// ```
// lsof -d cwd -a +D Users/cs/src/neon/test_output
// ```
.current_dir(datadir);
.args(args);
let filled_cmd = fill_env_vars_prefixed_neon(fill_remote_storage_secrets_vars(
fill_rust_env_vars(background_command),
@@ -133,7 +121,7 @@ where
.unwrap();
});
for retries in 0..retries {
for retries in 0..RETRIES {
match process_started(pid, pid_file_to_check, &process_status_check).await {
Ok(true) => {
println!("\n{process_name} started and passed status check, pid: {pid}");
@@ -151,7 +139,7 @@ where
print!(".");
io::stdout().flush().unwrap();
}
thread::sleep(RETRY_INTERVAL);
thread::sleep(Duration::from_millis(RETRY_INTERVAL_MILLIS));
}
Err(e) => {
println!("error starting process {process_name:?}: {e:#}");
@@ -160,10 +148,9 @@ where
}
}
println!();
anyhow::bail!(format!(
"{} did not start+pass status checks within {:?} seconds",
process_name, retry_timeout
));
anyhow::bail!(
"{process_name} did not start+pass status checks within {RETRY_UNTIL_SECS} seconds"
);
}
/// Stops the process, using the pid file given. Returns Ok also if the process is already not running.
@@ -219,7 +206,7 @@ pub fn stop_process(
}
pub fn wait_until_stopped(process_name: &str, pid: Pid) -> anyhow::Result<()> {
for retries in 0..STOP_RETRIES {
for retries in 0..RETRIES {
match process_has_stopped(pid) {
Ok(true) => {
println!("\n{process_name} stopped");
@@ -235,7 +222,7 @@ pub fn wait_until_stopped(process_name: &str, pid: Pid) -> anyhow::Result<()> {
print!(".");
io::stdout().flush().unwrap();
}
thread::sleep(RETRY_INTERVAL);
thread::sleep(Duration::from_millis(RETRY_INTERVAL_MILLIS));
}
Err(e) => {
println!("{process_name} with pid {pid} failed to stop: {e:#}");
@@ -244,10 +231,7 @@ pub fn wait_until_stopped(process_name: &str, pid: Pid) -> anyhow::Result<()> {
}
}
println!();
anyhow::bail!(format!(
"{} with pid {} did not stop in {:?} seconds",
process_name, pid, STOP_RETRY_TIMEOUT
));
anyhow::bail!("{process_name} with pid {pid} did not stop in {RETRY_UNTIL_SECS} seconds");
}
fn fill_rust_env_vars(cmd: &mut Command) -> &mut Command {

View File

@@ -36,7 +36,6 @@ use std::collections::{BTreeSet, HashMap};
use std::path::PathBuf;
use std::process::exit;
use std::str::FromStr;
use std::time::Duration;
use storage_broker::DEFAULT_LISTEN_ADDR as DEFAULT_BROKER_ADDR;
use url::Host;
use utils::{
@@ -88,8 +87,7 @@ fn main() -> Result<()> {
handle_init(sub_args).map(Some)
} else {
// all other commands need an existing config
let mut env =
LocalEnv::load_config(&local_env::base_path()).context("Error loading config")?;
let mut env = LocalEnv::load_config().context("Error loading config")?;
let original_env = env.clone();
let rt = tokio::runtime::Builder::new_current_thread()
@@ -100,7 +98,7 @@ fn main() -> Result<()> {
let subcommand_result = match sub_name {
"tenant" => rt.block_on(handle_tenant(sub_args, &mut env)),
"timeline" => rt.block_on(handle_timeline(sub_args, &mut env)),
"start" => rt.block_on(handle_start_all(&env, get_start_timeout(sub_args))),
"start" => rt.block_on(handle_start_all(&env)),
"stop" => rt.block_on(handle_stop_all(sub_args, &env)),
"pageserver" => rt.block_on(handle_pageserver(sub_args, &env)),
"storage_controller" => rt.block_on(handle_storage_controller(sub_args, &env)),
@@ -366,8 +364,7 @@ fn handle_init(init_match: &ArgMatches) -> anyhow::Result<LocalEnv> {
LocalEnv::init(init_conf, force)
.context("materialize initial neon_local environment on disk")?;
Ok(LocalEnv::load_config(&local_env::base_path())
.expect("freshly written config should be loadable"))
Ok(LocalEnv::load_config().expect("freshly written config should be loadable"))
}
/// The default pageserver is the one where CLI tenant/timeline operations are sent by default.
@@ -600,9 +597,13 @@ async fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::Local
Some(("import", import_match)) => {
let tenant_id = get_tenant_id(import_match, env)?;
let timeline_id = parse_timeline_id(import_match)?.expect("No timeline id provided");
let branch_name = import_match
.get_one::<String>("branch-name")
.ok_or_else(|| anyhow!("No branch name provided"))?;
let name = import_match
.get_one::<String>("node-name")
.ok_or_else(|| anyhow!("No node name provided"))?;
let update_catalog = import_match
.get_one::<bool>("update-catalog")
.cloned()
.unwrap_or_default();
// Parse base inputs
let base_tarfile = import_match
@@ -629,11 +630,24 @@ async fn handle_timeline(timeline_match: &ArgMatches, env: &mut local_env::Local
.copied()
.context("Failed to parse postgres version from the argument string")?;
let mut cplane = ComputeControlPlane::load(env.clone())?;
println!("Importing timeline into pageserver ...");
pageserver
.timeline_import(tenant_id, timeline_id, base, pg_wal, pg_version)
.await?;
env.register_branch_mapping(branch_name.to_string(), tenant_id, timeline_id)?;
env.register_branch_mapping(name.to_string(), tenant_id, timeline_id)?;
println!("Creating endpoint for imported timeline ...");
cplane.new_endpoint(
name,
tenant_id,
timeline_id,
None,
None,
pg_version,
ComputeMode::Primary,
!update_catalog,
)?;
println!("Done");
}
Some(("branch", branch_match)) => {
@@ -1032,20 +1046,10 @@ fn get_pageserver(env: &local_env::LocalEnv, args: &ArgMatches) -> Result<PageSe
))
}
fn get_start_timeout(args: &ArgMatches) -> &Duration {
let humantime_duration = args
.get_one::<humantime::Duration>("start-timeout")
.expect("invalid value for start-timeout");
humantime_duration.as_ref()
}
async fn handle_pageserver(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
match sub_match.subcommand() {
Some(("start", subcommand_args)) => {
if let Err(e) = get_pageserver(env, subcommand_args)?
.start(get_start_timeout(subcommand_args))
.await
{
if let Err(e) = get_pageserver(env, subcommand_args)?.start().await {
eprintln!("pageserver start failed: {e}");
exit(1);
}
@@ -1071,7 +1075,7 @@ async fn handle_pageserver(sub_match: &ArgMatches, env: &local_env::LocalEnv) ->
exit(1);
}
if let Err(e) = pageserver.start(get_start_timeout(sub_match)).await {
if let Err(e) = pageserver.start().await {
eprintln!("pageserver start failed: {e}");
exit(1);
}
@@ -1099,8 +1103,8 @@ async fn handle_storage_controller(
) -> Result<()> {
let svc = StorageController::from_env(env);
match sub_match.subcommand() {
Some(("start", start_match)) => {
if let Err(e) = svc.start(get_start_timeout(start_match)).await {
Some(("start", _start_match)) => {
if let Err(e) = svc.start().await {
eprintln!("start failed: {e}");
exit(1);
}
@@ -1159,10 +1163,7 @@ async fn handle_safekeeper(sub_match: &ArgMatches, env: &local_env::LocalEnv) ->
"start" => {
let extra_opts = safekeeper_extra_opts(sub_args);
if let Err(e) = safekeeper
.start(extra_opts, get_start_timeout(sub_args))
.await
{
if let Err(e) = safekeeper.start(extra_opts).await {
eprintln!("safekeeper start failed: {}", e);
exit(1);
}
@@ -1188,10 +1189,7 @@ async fn handle_safekeeper(sub_match: &ArgMatches, env: &local_env::LocalEnv) ->
}
let extra_opts = safekeeper_extra_opts(sub_args);
if let Err(e) = safekeeper
.start(extra_opts, get_start_timeout(sub_args))
.await
{
if let Err(e) = safekeeper.start(extra_opts).await {
eprintln!("safekeeper start failed: {}", e);
exit(1);
}
@@ -1204,18 +1202,15 @@ async fn handle_safekeeper(sub_match: &ArgMatches, env: &local_env::LocalEnv) ->
Ok(())
}
async fn handle_start_all(
env: &local_env::LocalEnv,
retry_timeout: &Duration,
) -> anyhow::Result<()> {
async fn handle_start_all(env: &local_env::LocalEnv) -> anyhow::Result<()> {
// Endpoints are not started automatically
broker::start_broker_process(env, retry_timeout).await?;
broker::start_broker_process(env).await?;
// Only start the storage controller if the pageserver is configured to need it
if env.control_plane_api.is_some() {
let storage_controller = StorageController::from_env(env);
if let Err(e) = storage_controller.start(retry_timeout).await {
if let Err(e) = storage_controller.start().await {
eprintln!("storage_controller start failed: {:#}", e);
try_stop_all(env, true).await;
exit(1);
@@ -1224,7 +1219,7 @@ async fn handle_start_all(
for ps_conf in &env.pageservers {
let pageserver = PageServerNode::from_env(env, ps_conf);
if let Err(e) = pageserver.start(retry_timeout).await {
if let Err(e) = pageserver.start().await {
eprintln!("pageserver {} start failed: {:#}", ps_conf.id, e);
try_stop_all(env, true).await;
exit(1);
@@ -1233,7 +1228,7 @@ async fn handle_start_all(
for node in env.safekeepers.iter() {
let safekeeper = SafekeeperNode::from_env(env, node);
if let Err(e) = safekeeper.start(vec![], retry_timeout).await {
if let Err(e) = safekeeper.start(vec![]).await {
eprintln!("safekeeper {} start failed: {:#}", safekeeper.id, e);
try_stop_all(env, false).await;
exit(1);
@@ -1293,15 +1288,6 @@ async fn try_stop_all(env: &local_env::LocalEnv, immediate: bool) {
}
fn cli() -> Command {
let timeout_arg = Arg::new("start-timeout")
.long("start-timeout")
.short('t')
.global(true)
.help("timeout until we fail the command, e.g. 30s")
.value_parser(value_parser!(humantime::Duration))
.default_value("10s")
.required(false);
let branch_name_arg = Arg::new("branch-name")
.long("branch-name")
.help("Name of the branch to be created or used as an alias for other services")
@@ -1470,7 +1456,8 @@ fn cli() -> Command {
.about("Import timeline from basebackup directory")
.arg(tenant_id_arg.clone())
.arg(timeline_id_arg.clone())
.arg(branch_name_arg.clone())
.arg(Arg::new("node-name").long("node-name")
.help("Name to assign to the imported timeline"))
.arg(Arg::new("base-tarfile")
.long("base-tarfile")
.value_parser(value_parser!(PathBuf))
@@ -1486,6 +1473,7 @@ fn cli() -> Command {
.arg(Arg::new("end-lsn").long("end-lsn")
.help("Lsn the basebackup ends at"))
.arg(pg_version_arg.clone())
.arg(update_catalog.clone())
)
).subcommand(
Command::new("tenant")
@@ -1519,7 +1507,6 @@ fn cli() -> Command {
.subcommand(Command::new("status"))
.subcommand(Command::new("start")
.about("Start local pageserver")
.arg(timeout_arg.clone())
)
.subcommand(Command::new("stop")
.about("Stop local pageserver")
@@ -1527,15 +1514,13 @@ fn cli() -> Command {
)
.subcommand(Command::new("restart")
.about("Restart local pageserver")
.arg(timeout_arg.clone())
)
)
.subcommand(
Command::new("storage_controller")
.arg_required_else_help(true)
.about("Manage storage_controller")
.subcommand(Command::new("start").about("Start storage controller")
.arg(timeout_arg.clone()))
.subcommand(Command::new("start").about("Start storage controller"))
.subcommand(Command::new("stop").about("Stop storage controller")
.arg(stop_mode_arg.clone()))
)
@@ -1547,7 +1532,6 @@ fn cli() -> Command {
.about("Start local safekeeper")
.arg(safekeeper_id_arg.clone())
.arg(safekeeper_extra_opt_arg.clone())
.arg(timeout_arg.clone())
)
.subcommand(Command::new("stop")
.about("Stop local safekeeper")
@@ -1559,7 +1543,6 @@ fn cli() -> Command {
.arg(safekeeper_id_arg)
.arg(stop_mode_arg.clone())
.arg(safekeeper_extra_opt_arg)
.arg(timeout_arg.clone())
)
)
.subcommand(
@@ -1594,7 +1577,6 @@ fn cli() -> Command {
.arg(remote_ext_config_args)
.arg(create_test_user)
.arg(allow_multiple.clone())
.arg(timeout_arg.clone())
)
.subcommand(Command::new("reconfigure")
.about("Reconfigure the endpoint")
@@ -1646,7 +1628,6 @@ fn cli() -> Command {
.subcommand(
Command::new("start")
.about("Start page server and safekeepers")
.arg(timeout_arg.clone())
)
.subcommand(
Command::new("stop")

View File

@@ -5,18 +5,13 @@
//! ```text
//! .neon/safekeepers/<safekeeper id>
//! ```
use std::time::Duration;
use anyhow::Context;
use camino::Utf8PathBuf;
use crate::{background_process, local_env};
pub async fn start_broker_process(
env: &local_env::LocalEnv,
retry_timeout: &Duration,
) -> anyhow::Result<()> {
pub async fn start_broker_process(env: &local_env::LocalEnv) -> anyhow::Result<()> {
let broker = &env.broker;
let listen_addr = &broker.listen_addr;
@@ -32,7 +27,6 @@ pub async fn start_broker_process(
args,
[],
background_process::InitialPidFile::Create(storage_broker_pid_file_path(env)),
retry_timeout,
|| async {
let url = broker.client_url();
let status_url = url.join("status").with_context(|| {

View File

@@ -592,6 +592,7 @@ impl Endpoint {
remote_extensions,
pgbouncer_settings: None,
shard_stripe_size: Some(shard_stripe_size),
primary_is_running: None,
};
let spec_path = self.endpoint_path().join("spec.json");
std::fs::write(spec_path, serde_json::to_string_pretty(&spec)?)?;

View File

@@ -42,8 +42,8 @@ pub struct LocalEnv {
// compute endpoints).
//
// This is not stored in the config file. Rather, this is the path where the
// config file itself is. It is read from the NEON_REPO_DIR env variable which
// must be an absolute path. If the env var is not set, $PWD/.neon is used.
// config file itself is. It is read from the NEON_REPO_DIR env variable or
// '.neon' if not given.
pub base_data_dir: PathBuf,
// Path to postgres distribution. It's expected that "bin", "include",
@@ -431,7 +431,9 @@ impl LocalEnv {
}
/// Construct `Self` from on-disk state.
pub fn load_config(repopath: &Path) -> anyhow::Result<Self> {
pub fn load_config() -> anyhow::Result<Self> {
let repopath = base_path();
if !repopath.exists() {
bail!(
"Neon config is not found in {}. You need to run 'neon_local init' first",
@@ -459,7 +461,7 @@ impl LocalEnv {
branch_name_mappings,
} = on_disk_config;
LocalEnv {
base_data_dir: repopath.to_owned(),
base_data_dir: repopath.clone(),
pg_distrib_dir,
neon_distrib_dir,
default_tenant_id,
@@ -480,7 +482,7 @@ impl LocalEnv {
"we ensure this during deserialization"
);
env.pageservers = {
let iter = std::fs::read_dir(repopath).context("open dir")?;
let iter = std::fs::read_dir(&repopath).context("open dir")?;
let mut pageservers = Vec::new();
for res in iter {
let dentry = res?;
@@ -717,25 +719,10 @@ impl LocalEnv {
}
pub fn base_path() -> PathBuf {
let path = match std::env::var_os("NEON_REPO_DIR") {
Some(val) => {
let path = PathBuf::from(val);
if !path.is_absolute() {
// repeat the env var in the error because our default is always absolute
panic!("NEON_REPO_DIR must be an absolute path, got {path:?}");
}
path
}
None => {
let pwd = std::env::current_dir()
// technically this can fail but it's quite unlikeley
.expect("determine current directory");
let pwd_abs = pwd.canonicalize().expect("canonicalize current directory");
pwd_abs.join(".neon")
}
};
assert!(path.is_absolute());
path
match std::env::var_os("NEON_REPO_DIR") {
Some(val) => PathBuf::from(val),
None => PathBuf::from(".neon"),
}
}
/// Generate a public/private key pair for JWT authentication

View File

@@ -158,8 +158,8 @@ impl PageServerNode {
.expect("non-Unicode path")
}
pub async fn start(&self, retry_timeout: &Duration) -> anyhow::Result<()> {
self.start_node(retry_timeout).await
pub async fn start(&self) -> anyhow::Result<()> {
self.start_node().await
}
fn pageserver_init(&self, conf: NeonLocalInitPageserverConf) -> anyhow::Result<()> {
@@ -214,15 +214,14 @@ impl PageServerNode {
Ok(())
}
async fn start_node(&self, retry_timeout: &Duration) -> anyhow::Result<()> {
async fn start_node(&self) -> anyhow::Result<()> {
// TODO: using a thread here because start_process() is not async but we need to call check_status()
let datadir = self.repo_path();
print!(
"Starting pageserver node {} at '{}' in {:?}, retrying for {:?}",
"Starting pageserver node {} at '{}' in {:?}",
self.conf.id,
self.pg_connection_config.raw_address(),
datadir,
retry_timeout
datadir
);
io::stdout().flush().context("flush stdout")?;
@@ -240,7 +239,6 @@ impl PageServerNode {
args,
self.pageserver_env_variables()?,
background_process::InitialPidFile::Expect(self.pid_file()),
retry_timeout,
|| async {
let st = self.check_status().await;
match st {
@@ -385,10 +383,6 @@ impl PageServerNode {
.map(|x| x.parse::<AuxFilePolicy>())
.transpose()
.context("Failed to parse 'switch_aux_file_policy'")?,
lsn_lease_length: settings.remove("lsn_lease_length").map(|x| x.to_string()),
lsn_lease_length_for_ts: settings
.remove("lsn_lease_length_for_ts")
.map(|x| x.to_string()),
};
if !settings.is_empty() {
bail!("Unrecognized tenant settings: {settings:?}")
@@ -512,10 +506,6 @@ impl PageServerNode {
.map(|x| x.parse::<AuxFilePolicy>())
.transpose()
.context("Failed to parse 'switch_aux_file_policy'")?,
lsn_lease_length: settings.remove("lsn_lease_length").map(|x| x.to_string()),
lsn_lease_length_for_ts: settings
.remove("lsn_lease_length_for_ts")
.map(|x| x.to_string()),
}
};

View File

@@ -7,7 +7,6 @@
//! ```
use std::io::Write;
use std::path::PathBuf;
use std::time::Duration;
use std::{io, result};
use anyhow::Context;
@@ -15,7 +14,6 @@ use camino::Utf8PathBuf;
use postgres_connection::PgConnectionConfig;
use reqwest::{IntoUrl, Method};
use thiserror::Error;
use utils::auth::{Claims, Scope};
use utils::{http::error::HttpErrorBody, id::NodeId};
use crate::{
@@ -112,16 +110,11 @@ impl SafekeeperNode {
.expect("non-Unicode path")
}
pub async fn start(
&self,
extra_opts: Vec<String>,
retry_timeout: &Duration,
) -> anyhow::Result<()> {
pub async fn start(&self, extra_opts: Vec<String>) -> anyhow::Result<()> {
print!(
"Starting safekeeper at '{}' in '{}', retrying for {:?}",
"Starting safekeeper at '{}' in '{}'",
self.pg_connection_config.raw_address(),
self.datadir_path().display(),
retry_timeout,
self.datadir_path().display()
);
io::stdout().flush().unwrap();
@@ -204,9 +197,8 @@ impl SafekeeperNode {
&datadir,
&self.env.safekeeper_bin(),
&args,
self.safekeeper_env_variables()?,
[],
background_process::InitialPidFile::Expect(self.pid_file()),
retry_timeout,
|| async {
match self.check_status().await {
Ok(()) => Ok(true),
@@ -218,18 +210,6 @@ impl SafekeeperNode {
.await
}
fn safekeeper_env_variables(&self) -> anyhow::Result<Vec<(String, String)>> {
// Generate a token to connect from safekeeper to peers
if self.conf.auth_enabled {
let token = self
.env
.generate_auth_token(&Claims::new(None, Scope::SafekeeperData))?;
Ok(vec![("SAFEKEEPER_AUTH_TOKEN".to_owned(), token)])
} else {
Ok(Vec::new())
}
}
///
/// Stop the server.
///

View File

@@ -18,7 +18,7 @@ use pageserver_client::mgmt_api::ResponseErrorMessageExt;
use postgres_backend::AuthType;
use reqwest::Method;
use serde::{de::DeserializeOwned, Deserialize, Serialize};
use std::{fs, str::FromStr, time::Duration};
use std::{fs, str::FromStr};
use tokio::process::Command;
use tracing::instrument;
use url::Url;
@@ -46,7 +46,6 @@ const STORAGE_CONTROLLER_POSTGRES_VERSION: u32 = 16;
pub struct AttachHookRequest {
pub tenant_shard_id: TenantShardId,
pub node_id: Option<NodeId>,
pub generation_override: Option<i32>,
}
#[derive(Serialize, Deserialize)]
@@ -224,7 +223,7 @@ impl StorageController {
Ok(database_url)
}
pub async fn start(&self, retry_timeout: &Duration) -> anyhow::Result<()> {
pub async fn start(&self) -> anyhow::Result<()> {
// Start a vanilla Postgres process used by the storage controller for persistence.
let pg_data_path = Utf8PathBuf::from_path_buf(self.env.base_data_dir.clone())
.unwrap()
@@ -272,7 +271,6 @@ impl StorageController {
db_start_args,
[],
background_process::InitialPidFile::Create(self.postgres_pid_file()),
retry_timeout,
|| self.pg_isready(&pg_bin_dir),
)
.await?;
@@ -315,19 +313,16 @@ impl StorageController {
args.push(format!("--split-threshold={split_threshold}"))
}
args.push(format!(
"--neon-local-repo-dir={}",
self.env.base_data_dir.display()
));
background_process::start_process(
COMMAND,
&self.env.base_data_dir,
&self.env.storage_controller_bin(),
args,
[],
[(
"NEON_REPO_DIR".to_string(),
self.env.base_data_dir.to_string_lossy().to_string(),
)],
background_process::InitialPidFile::Create(self.pid_file()),
retry_timeout,
|| async {
match self.ready().await {
Ok(_) => Ok(true),
@@ -445,7 +440,6 @@ impl StorageController {
let request = AttachHookRequest {
tenant_shard_id,
node_id: Some(pageserver_id),
generation_override: None,
};
let response = self

View File

@@ -9,8 +9,6 @@ license.workspace = true
anyhow.workspace = true
clap.workspace = true
comfy-table.workspace = true
futures.workspace = true
humantime.workspace = true
hyper.workspace = true
pageserver_api.workspace = true
pageserver_client.workspace = true

View File

@@ -1,5 +1,4 @@
use futures::StreamExt;
use std::{str::FromStr, time::Duration};
use std::{collections::HashMap, str::FromStr, time::Duration};
use clap::{Parser, Subcommand};
use pageserver_api::{
@@ -8,9 +7,8 @@ use pageserver_api::{
TenantDescribeResponse, TenantPolicyRequest,
},
models::{
EvictionPolicy, EvictionPolicyLayerAccessThreshold, LocationConfigSecondary,
ShardParameters, TenantConfig, TenantConfigRequest, TenantCreateRequest,
TenantShardSplitRequest, TenantShardSplitResponse,
LocationConfigSecondary, ShardParameters, TenantConfig, TenantConfigRequest,
TenantCreateRequest, TenantShardSplitRequest, TenantShardSplitResponse,
},
shard::{ShardStripeSize, TenantShardId},
};
@@ -21,7 +19,7 @@ use utils::id::{NodeId, TenantId};
use pageserver_api::controller_api::{
NodeConfigureRequest, NodeRegisterRequest, NodeSchedulingPolicy, PlacementPolicy,
TenantShardMigrateRequest, TenantShardMigrateResponse,
TenantLocateResponse, TenantShardMigrateRequest, TenantShardMigrateResponse,
};
#[derive(Subcommand, Debug)]
@@ -110,6 +108,12 @@ enum Command {
#[arg(long)]
config: String,
},
/// Attempt to balance the locations for a tenant across pageservers. This is a client-side
/// alternative to the storage controller's scheduling optimization behavior.
TenantScatter {
#[arg(long)]
tenant_id: TenantId,
},
/// Print details about a particular tenant, including all its shards' states.
TenantDescribe {
#[arg(long)]
@@ -121,44 +125,6 @@ enum Command {
#[arg(long)]
tenant_id: TenantId,
},
/// Uncleanly drop a tenant from the storage controller: this doesn't delete anything from pageservers. Appropriate
/// if you e.g. used `tenant-warmup` by mistake on a tenant ID that doesn't really exist, or is in some other region.
TenantDrop {
#[arg(long)]
tenant_id: TenantId,
#[arg(long)]
unclean: bool,
},
NodeDrop {
#[arg(long)]
node_id: NodeId,
#[arg(long)]
unclean: bool,
},
TenantSetTimeBasedEviction {
#[arg(long)]
tenant_id: TenantId,
#[arg(long)]
period: humantime::Duration,
#[arg(long)]
threshold: humantime::Duration,
},
// Drain a set of specified pageservers by moving the primary attachments to pageservers
// outside of the specified set.
Drain {
// Set of pageserver node ids to drain.
#[arg(long)]
nodes: Vec<NodeId>,
// Optional: migration concurrency (default is 8)
#[arg(long)]
concurrency: Option<usize>,
// Optional: maximum number of shards to migrate
#[arg(long)]
max_shards: Option<usize>,
// Optional: when set to true, nothing is migrated, but the plan is printed to stdout
#[arg(long)]
dry_run: Option<bool>,
},
}
#[derive(Parser)]
@@ -492,6 +458,88 @@ async fn main() -> anyhow::Result<()> {
})
.await?;
}
Command::TenantScatter { tenant_id } => {
// Find the shards
let locate_response = storcon_client
.dispatch::<(), TenantLocateResponse>(
Method::GET,
format!("control/v1/tenant/{tenant_id}/locate"),
None,
)
.await?;
let shards = locate_response.shards;
let mut node_to_shards: HashMap<NodeId, Vec<TenantShardId>> = HashMap::new();
let shard_count = shards.len();
for s in shards {
let entry = node_to_shards.entry(s.node_id).or_default();
entry.push(s.shard_id);
}
// Load list of available nodes
let nodes_resp = storcon_client
.dispatch::<(), Vec<NodeDescribeResponse>>(
Method::GET,
"control/v1/node".to_string(),
None,
)
.await?;
for node in nodes_resp {
if matches!(node.availability, NodeAvailabilityWrapper::Active) {
node_to_shards.entry(node.id).or_default();
}
}
let max_shard_per_node = shard_count / node_to_shards.len();
loop {
let mut migrate_shard = None;
for shards in node_to_shards.values_mut() {
if shards.len() > max_shard_per_node {
// Pick the emptiest
migrate_shard = Some(shards.pop().unwrap());
}
}
let Some(migrate_shard) = migrate_shard else {
break;
};
// Pick the emptiest node to migrate to
let mut destinations = node_to_shards
.iter()
.map(|(k, v)| (k, v.len()))
.collect::<Vec<_>>();
destinations.sort_by_key(|i| i.1);
let (destination_node, destination_count) = *destinations.first().unwrap();
if destination_count + 1 > max_shard_per_node {
// Even the emptiest destination doesn't have space: we're done
break;
}
let destination_node = *destination_node;
node_to_shards
.get_mut(&destination_node)
.unwrap()
.push(migrate_shard);
println!("Migrate {} -> {} ...", migrate_shard, destination_node);
storcon_client
.dispatch::<TenantShardMigrateRequest, TenantShardMigrateResponse>(
Method::PUT,
format!("control/v1/tenant/{migrate_shard}/migrate"),
Some(TenantShardMigrateRequest {
tenant_shard_id: migrate_shard,
node_id: destination_node,
}),
)
.await?;
println!("Migrate {} -> {} OK", migrate_shard, destination_node);
}
// Spread the shards across the nodes
}
Command::TenantDescribe { tenant_id } => {
let describe_response = storcon_client
.dispatch::<(), TenantDescribeResponse>(
@@ -626,234 +674,6 @@ async fn main() -> anyhow::Result<()> {
}
}
}
Command::TenantDrop { tenant_id, unclean } => {
if !unclean {
anyhow::bail!("This command is not a tenant deletion, and uncleanly drops all controller state for the tenant. If you know what you're doing, add `--unclean` to proceed.")
}
storcon_client
.dispatch::<(), ()>(
Method::POST,
format!("debug/v1/tenant/{tenant_id}/drop"),
None,
)
.await?;
}
Command::NodeDrop { node_id, unclean } => {
if !unclean {
anyhow::bail!("This command is not a clean node decommission, and uncleanly drops all controller state for the node, without checking if any tenants still refer to it. If you know what you're doing, add `--unclean` to proceed.")
}
storcon_client
.dispatch::<(), ()>(Method::POST, format!("debug/v1/node/{node_id}/drop"), None)
.await?;
}
Command::TenantSetTimeBasedEviction {
tenant_id,
period,
threshold,
} => {
vps_client
.tenant_config(&TenantConfigRequest {
tenant_id,
config: TenantConfig {
eviction_policy: Some(EvictionPolicy::LayerAccessThreshold(
EvictionPolicyLayerAccessThreshold {
period: period.into(),
threshold: threshold.into(),
},
)),
..Default::default()
},
})
.await?;
}
Command::Drain {
nodes,
concurrency,
max_shards,
dry_run,
} => {
// Load the list of nodes, split them up into the drained and filled sets,
// and validate that draining is possible.
let node_descs = storcon_client
.dispatch::<(), Vec<NodeDescribeResponse>>(
Method::GET,
"control/v1/node".to_string(),
None,
)
.await?;
let mut node_to_drain_descs = Vec::new();
let mut node_to_fill_descs = Vec::new();
for desc in node_descs {
let to_drain = nodes.iter().any(|id| *id == desc.id);
if to_drain {
node_to_drain_descs.push(desc);
} else {
node_to_fill_descs.push(desc);
}
}
if nodes.len() != node_to_drain_descs.len() {
anyhow::bail!("Drain requested for node which doesn't exist.")
}
node_to_fill_descs.retain(|desc| {
matches!(desc.availability, NodeAvailabilityWrapper::Active)
&& matches!(
desc.scheduling,
NodeSchedulingPolicy::Active | NodeSchedulingPolicy::Filling
)
});
if node_to_fill_descs.is_empty() {
anyhow::bail!("There are no nodes to drain to")
}
// Set the node scheduling policy to draining for the nodes which
// we plan to drain.
for node_desc in node_to_drain_descs.iter() {
let req = NodeConfigureRequest {
node_id: node_desc.id,
availability: None,
scheduling: Some(NodeSchedulingPolicy::Draining),
};
storcon_client
.dispatch::<_, ()>(
Method::PUT,
format!("control/v1/node/{}/config", node_desc.id),
Some(req),
)
.await?;
}
// Perform the drain: move each tenant shard scheduled on a node to
// be drained to a node which is being filled. A simple round robin
// strategy is used to pick the new node.
let tenants = storcon_client
.dispatch::<(), Vec<TenantDescribeResponse>>(
Method::GET,
"control/v1/tenant".to_string(),
None,
)
.await?;
let mut selected_node_idx = 0;
struct DrainMove {
tenant_shard_id: TenantShardId,
from: NodeId,
to: NodeId,
}
let mut moves: Vec<DrainMove> = Vec::new();
let shards = tenants
.into_iter()
.flat_map(|tenant| tenant.shards.into_iter());
for shard in shards {
if let Some(max_shards) = max_shards {
if moves.len() >= max_shards {
println!(
"Stop planning shard moves since the requested maximum was reached"
);
break;
}
}
let should_migrate = {
if let Some(attached_to) = shard.node_attached {
node_to_drain_descs
.iter()
.map(|desc| desc.id)
.any(|id| id == attached_to)
} else {
false
}
};
if !should_migrate {
continue;
}
moves.push(DrainMove {
tenant_shard_id: shard.tenant_shard_id,
from: shard
.node_attached
.expect("We only migrate attached tenant shards"),
to: node_to_fill_descs[selected_node_idx].id,
});
selected_node_idx = (selected_node_idx + 1) % node_to_fill_descs.len();
}
let total_moves = moves.len();
if dry_run == Some(true) {
println!("Dryrun requested. Planned {total_moves} moves:");
for mv in &moves {
println!("{}: {} -> {}", mv.tenant_shard_id, mv.from, mv.to)
}
return Ok(());
}
const DEFAULT_MIGRATE_CONCURRENCY: usize = 8;
let mut stream = futures::stream::iter(moves)
.map(|mv| {
let client = Client::new(cli.api.clone(), cli.jwt.clone());
async move {
client
.dispatch::<TenantShardMigrateRequest, TenantShardMigrateResponse>(
Method::PUT,
format!("control/v1/tenant/{}/migrate", mv.tenant_shard_id),
Some(TenantShardMigrateRequest {
tenant_shard_id: mv.tenant_shard_id,
node_id: mv.to,
}),
)
.await
.map_err(|e| (mv.tenant_shard_id, mv.from, mv.to, e))
}
})
.buffered(concurrency.unwrap_or(DEFAULT_MIGRATE_CONCURRENCY));
let mut success = 0;
let mut failure = 0;
while let Some(res) = stream.next().await {
match res {
Ok(_) => {
success += 1;
}
Err((tenant_shard_id, from, to, error)) => {
failure += 1;
println!(
"Failed to migrate {} from node {} to node {}: {}",
tenant_shard_id, from, to, error
);
}
}
if (success + failure) % 20 == 0 {
println!(
"Processed {}/{} shards: {} succeeded, {} failed",
success + failure,
total_moves,
success,
failure
);
}
}
println!(
"Processed {}/{} shards: {} succeeded, {} failed",
success + failure,
total_moves,
success,
failure
);
}
}
Ok(())

View File

@@ -99,13 +99,6 @@ name = "async-executor"
[[bans.deny]]
name = "smol"
[[bans.deny]]
# We want to use rustls instead of the platform's native tls implementation.
name = "native-tls"
[[bans.deny]]
name = "openssl"
# This section is considered when running `cargo deny check sources`.
# More documentation about the 'sources' section can be found here:
# https://embarkstudios.github.io/cargo-deny/checks/sources/cfg.html

View File

@@ -1,10 +0,0 @@
# Example docker compose configuration
The configuration in this directory is used for testing Neon docker images: it is
not intended for deploying a usable system. To run a development environment where
you can experiment with a minature Neon system, use `cargo neon` rather than container images.
This configuration does not start the storage controller, because the controller
needs a way to reconfigure running computes, and no such thing exists in this setup.

View File

@@ -1,4 +1,4 @@
ARG REPOSITORY=neondatabase
ARG REPOSITORY=369495373322.dkr.ecr.eu-central-1.amazonaws.com
ARG COMPUTE_IMAGE=compute-node-v14
ARG TAG=latest
@@ -8,11 +8,6 @@ USER root
RUN apt-get update && \
apt-get install -y curl \
jq \
python3-pip \
netcat
#Faker is required for the pg_anon test
RUN pip3 install Faker
#This is required for the pg_hintplan test
RUN mkdir -p /ext-src/pg_hint_plan-src && chown postgres /ext-src/pg_hint_plan-src
USER postgres
USER postgres

View File

@@ -23,10 +23,11 @@ echo "Page server is ready."
echo "Create a tenant and timeline"
generate_id tenant_id
PARAMS=(
-X PUT
-sb
-X POST
-H "Content-Type: application/json"
-d "{\"mode\": \"AttachedSingle\", \"generation\": 1, \"tenant_conf\": {}}"
"http://pageserver:9898/v1/tenant/${tenant_id}/location_config"
-d "{\"new_tenant_id\": \"${tenant_id}\"}"
http://pageserver:9898/v1/tenant/
)
result=$(curl "${PARAMS[@]}")
echo $result | jq .

View File

@@ -95,7 +95,7 @@
},
{
"name": "shared_preload_libraries",
"value": "neon,pg_cron,timescaledb,pg_stat_statements",
"value": "neon",
"vartype": "string"
},
{
@@ -127,16 +127,6 @@
"name": "max_replication_flush_lag",
"value": "10GB",
"vartype": "string"
},
{
"name": "cron.database",
"value": "postgres",
"vartype": "string"
},
{
"name": "session_preload_libraries",
"value": "anon",
"vartype": "string"
}
]
},

View File

@@ -1,3 +1,5 @@
version: '3'
services:
minio:
restart: always
@@ -159,12 +161,12 @@ services:
context: ./compute_wrapper/
args:
- REPOSITORY=${REPOSITORY:-neondatabase}
- COMPUTE_IMAGE=compute-node-v${PG_VERSION:-16}
- COMPUTE_IMAGE=compute-node-v${PG_VERSION:-14}
- TAG=${TAG:-latest}
- http_proxy=$http_proxy
- https_proxy=$https_proxy
environment:
- PG_VERSION=${PG_VERSION:-16}
- PG_VERSION=${PG_VERSION:-14}
#- RUST_BACKTRACE=1
# Mount the test files directly, for faster editing cycle.
volumes:
@@ -192,14 +194,3 @@ services:
done"
depends_on:
- compute
neon-test-extensions:
profiles: ["test-extensions"]
image: ${REPOSITORY:-neondatabase}/neon-test-extensions-v${PG_TEST_VERSION:-16}:${TAG:-latest}
entrypoint:
- "/bin/bash"
- "-c"
command:
- sleep 1800
depends_on:
- compute

View File

@@ -7,94 +7,54 @@
# Implicitly accepts `REPOSITORY` and `TAG` env vars that are passed into the compose file
# Their defaults point at DockerHub `neondatabase/neon:latest` image.`,
# to verify custom image builds (e.g pre-published ones).
#
# A test script for postgres extensions
# Currently supports only v16
#
# XXX: Current does not work on M1 macs due to x86_64 Docker images compiled only, and no seccomp support in M1 Docker emulation layer.
set -eux -o pipefail
COMPOSE_FILE='docker-compose.yml'
cd $(dirname $0)
SCRIPT_DIR="$( cd -- "$( dirname -- "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )"
COMPOSE_FILE=$SCRIPT_DIR/docker-compose.yml
COMPUTE_CONTAINER_NAME=docker-compose-compute-1
TEST_CONTAINER_NAME=docker-compose-neon-test-extensions-1
PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -d postgres"
: ${http_proxy:=}
: ${https_proxy:=}
export http_proxy https_proxy
SQL="CREATE TABLE t(key int primary key, value text); insert into t values(1,1); select * from t;"
PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -c '$SQL' postgres"
cleanup() {
echo "show container information"
docker ps
docker compose --profile test-extensions -f $COMPOSE_FILE logs
docker compose -f $COMPOSE_FILE logs
echo "stop containers..."
docker compose --profile test-extensions -f $COMPOSE_FILE down
docker compose -f $COMPOSE_FILE down
}
echo "clean up containers if exists"
cleanup
for pg_version in 14 15 16; do
echo "clean up containers if exists"
cleanup
PG_TEST_VERSION=$(($pg_version < 16 ? 16 : $pg_version))
PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --build -d
echo "start containers (pg_version=$pg_version)."
PG_VERSION=$pg_version docker compose -f $COMPOSE_FILE up --build -d
echo "wait until the compute is ready. timeout after 60s. "
cnt=0
while sleep 3; do
while sleep 1; do
# check timeout
cnt=`expr $cnt + 3`
cnt=`expr $cnt + 1`
if [ $cnt -gt 60 ]; then
echo "timeout before the compute is ready."
cleanup
exit 1
fi
if docker compose --profile test-extensions -f $COMPOSE_FILE logs "compute_is_ready" | grep -q "accepting connections"; then
# check if the compute is ready
set +o pipefail
result=`docker compose -f $COMPOSE_FILE logs "compute_is_ready" | grep "accepting connections" | wc -l`
set -o pipefail
if [ $result -eq 1 ]; then
echo "OK. The compute is ready to connect."
echo "execute simple queries."
docker exec $COMPUTE_CONTAINER_NAME /bin/bash -c "psql $PSQL_OPTION"
cleanup
break
fi
done
if [ $pg_version -ge 16 ]
then
echo Enabling trust connection
docker exec $COMPUTE_CONTAINER_NAME bash -c "sed -i '\$d' /var/db/postgres/compute/pg_hba.conf && echo -e 'host\t all\t all\t all\t trust' >> /var/db/postgres/compute/pg_hba.conf && psql $PSQL_OPTION -c 'select pg_reload_conf()' "
echo Adding postgres role
docker exec $COMPUTE_CONTAINER_NAME psql $PSQL_OPTION -c "CREATE ROLE postgres SUPERUSER LOGIN"
# This is required for the pg_hint_plan test, to prevent flaky log message causing the test to fail
# It cannot be moved to Dockerfile now because the database directory is created after the start of the container
echo Adding dummy config
docker exec $COMPUTE_CONTAINER_NAME touch /var/db/postgres/compute/compute_ctl_temp_override.conf
# This block is required for the pg_anon extension test.
# The test assumes that it is running on the same host with the postgres engine.
# In our case it's not true, that's why we are copying files to the compute node
TMPDIR=$(mktemp -d)
docker cp $TEST_CONTAINER_NAME:/ext-src/pg_anon-src/data $TMPDIR/data
echo -e '1\t too \t many \t tabs' > $TMPDIR/data/bad.csv
docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/tmp/tmp_anon_alternate_data
rm -rf $TMPDIR
TMPDIR=$(mktemp -d)
# The following block does the same for the pg_hintplan test
docker cp $TEST_CONTAINER_NAME:/ext-src/pg_hint_plan-src/data $TMPDIR/data
docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/ext-src/pg_hint_plan-src/
rm -rf $TMPDIR
# We are running tests now
if docker exec -e SKIP=rum-src,timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,pg_graphql-src,kq_imcx-src,wal2json_2_5-src \
$TEST_CONTAINER_NAME /run-tests.sh | tee testout.txt
then
cleanup
else
FAILED=$(tail -1 testout.txt)
for d in $FAILED
do
mkdir $d
docker cp $TEST_CONTAINER_NAME:/ext-src/$d/regression.diffs $d || true
docker cp $TEST_CONTAINER_NAME:/ext-src/$d/regression.out $d || true
cat $d/regression.out $d/regression.diffs || true
done
rm -rf $FAILED
cleanup
exit 1
fi
fi
cleanup
done

View File

@@ -1,15 +0,0 @@
#!/bin/bash
set -x
cd /ext-src
FAILED=
LIST=$((echo ${SKIP} | sed 's/,/\n/g'; ls -d *-src) | sort | uniq -u)
for d in ${LIST}
do
[ -d ${d} ] || continue
psql -c "select 1" >/dev/null || break
make -C ${d} installcheck || FAILED="${d} ${FAILED}"
done
[ -z "${FAILED}" ] && exit 0
echo ${FAILED}
exit 1

View File

@@ -11,28 +11,15 @@ page server. We currently use the same binary for both, with --wal-redo runtime
the WAL redo mode. Some PostgreSQL changes are needed in the compute node, while others are just for
the WAL redo process.
In addition to core PostgreSQL changes, there is a Neon extension in the pgxn/neon directory that
hooks into the smgr interface, and rmgr extension in pgxn/neon_rmgr. The extensions are loaded into
the Postgres processes with shared_preload_libraries. Most of the Neon-specific code is in the
extensions, and for any new features, that is preferred over modifying core PostgreSQL code.
In addition to core PostgreSQL changes, there is a Neon extension in contrib/neon, to hook into the
smgr interface. Once all the core changes have been submitted to upstream or eliminated some other
way, the extension could live outside the postgres repository and build against vanilla PostgreSQL.
Below is a list of all the PostgreSQL source code changes, categorized into changes needed for
compute, and changes needed for the WAL redo process:
# Changes for Compute node
## Prefetching
There are changes in many places to perform prefetching, for example for sequential scans. Neon
doesn't benefit from OS readahead, and the latency to pageservers is quite high compared to local
disk, so prefetching is critical for performance, also for sequential scans.
### How to get rid of the patch
Upcoming "streaming read" work in v17 might simplify this. And async I/O work in v18 will hopefully
do more.
## Add t_cid to heap WAL records
```
@@ -50,11 +37,54 @@ The problem is that the XLOG_HEAP_INSERT record does not include the command id
Bite the bullet and submit the patch to PostgreSQL, to add the t_cid to the WAL records. It makes the WAL records larger, which could make this unpopular in the PostgreSQL community. However, it might simplify some logical decoding code; Andres Freund briefly mentioned in PGCon 2022 discussion on Heikki's Neon presentation that logical decoding currently needs to jump through some hoops to reconstruct the same information.
Update from Heikki (2024-04-17): I tried to write an upstream patch for that, to use the t_cid field for logical decoding, but it was not as straightforward as it first sounded.
### Alternatives
Perhaps we could write an extra WAL record with the t_cid information, when a page is evicted that contains rows that were touched a transaction that's still running. However, that seems very complicated.
## ginfast.c
```
diff --git a/src/backend/access/gin/ginfast.c b/src/backend/access/gin/ginfast.c
index e0d9940946..2d964c02e9 100644
--- a/src/backend/access/gin/ginfast.c
+++ b/src/backend/access/gin/ginfast.c
@@ -285,6 +285,17 @@ ginHeapTupleFastInsert(GinState *ginstate, GinTupleCollector *collector)
memset(&sublist, 0, sizeof(GinMetaPageData));
makeSublist(index, collector->tuples, collector->ntuples, &sublist);
+ if (metadata->head != InvalidBlockNumber)
+ {
+ /*
+ * ZENITH: Get buffer before XLogBeginInsert() to avoid recursive call
+ * of XLogBeginInsert(). Reading a new buffer might evict a dirty page from
+ * the buffer cache, and if that page happens to be an FSM or VM page, zenith_write()
+ * will try to WAL-log an image of the page.
+ */
+ buffer = ReadBuffer(index, metadata->tail);
+ }
+
if (needWal)
XLogBeginInsert();
@@ -316,7 +327,6 @@ ginHeapTupleFastInsert(GinState *ginstate, GinTupleCollector *collector)
data.prevTail = metadata->tail;
data.newRightlink = sublist.head;
- buffer = ReadBuffer(index, metadata->tail);
LockBuffer(buffer, GIN_EXCLUSIVE);
page = BufferGetPage(buffer);
```
The problem is explained in the comment above
### How to get rid of the patch
Can we stop WAL-logging FSM or VM pages? Or delay the WAL logging until we're out of the critical
section or something.
Maybe some bigger rewrite of FSM and VM would help to avoid WAL-logging FSM and VM page images?
## Mark index builds that use buffer manager without logging explicitly
```
@@ -65,8 +95,6 @@ Perhaps we could write an extra WAL record with the t_cid information, when a pa
also some changes in src/backend/storage/smgr/smgr.c
```
pgvector 0.6.0 also needs a similar change, which would be very nice to get rid of too.
When a GIN index is built, for example, it is built by inserting the entries into the index more or
less normally, but without WAL-logging anything. After the index has been built, we iterate through
all pages and write them to the WAL. That doesn't work for Neon, because if a page is not WAL-logged
@@ -81,10 +109,6 @@ an operation: `smgr_start_unlogged_build`, `smgr_finish_unlogged_build_phase_1`
I think it would make sense to be more explicit about that in PostgreSQL too. So extract these
changes to a patch and post to pgsql-hackers.
Perhaps we could deduce that an unlogged index build has started when we see a page being evicted
with zero LSN. How to be sure it's an unlogged index build rather than a bug? Currently we have a
check for that and PANIC if we see page with zero LSN being evicted. And how do we detect when the
index build has finished? See https://github.com/neondatabase/neon/pull/7440 for an attempt at that.
## Track last-written page LSN
@@ -116,6 +140,57 @@ The old method is still available, though.
Wait until v15?
## Cache relation sizes
The Neon extension contains a little cache for smgrnblocks() and smgrexists() calls, to avoid going
to the page server every time. It might be useful to cache those in PostgreSQL, maybe in the
relcache? (I think we do cache nblocks in relcache already, check why that's not good enough for
Neon)
## Use buffer manager when extending VM or FSM
```
src/backend/storage/freespace/freespace.c | 14 +-
src/backend/access/heap/visibilitymap.c | 15 +-
diff --git a/src/backend/access/heap/visibilitymap.c b/src/backend/access/heap/visibilitymap.c
index e198df65d8..addfe93eac 100644
--- a/src/backend/access/heap/visibilitymap.c
+++ b/src/backend/access/heap/visibilitymap.c
@@ -652,10 +652,19 @@ vm_extend(Relation rel, BlockNumber vm_nblocks)
/* Now extend the file */
while (vm_nblocks_now < vm_nblocks)
{
- PageSetChecksumInplace((Page) pg.data, vm_nblocks_now);
+ /*
+ * ZENITH: Initialize VM pages through buffer cache to prevent loading
+ * them from pageserver.
+ */
+ Buffer buffer = ReadBufferExtended(rel, VISIBILITYMAP_FORKNUM, P_NEW,
+ RBM_ZERO_AND_LOCK, NULL);
+ Page page = BufferGetPage(buffer);
+
+ PageInit((Page) page, BLCKSZ, 0);
+ PageSetChecksumInplace(page, vm_nblocks_now);
+ MarkBufferDirty(buffer);
+ UnlockReleaseBuffer(buffer);
- smgrextend(rel->rd_smgr, VISIBILITYMAP_FORKNUM, vm_nblocks_now,
- pg.data, false);
vm_nblocks_now++;
}
```
### Problem we're trying to solve
???
### How to get rid of the patch
Maybe this would be a reasonable change in PostgreSQL too?
## Allow startup without reading checkpoint record
In Neon, the compute node is stateless. So when we are launching compute node, we need to provide
@@ -156,7 +231,7 @@ index 0415df9ccb..9f9db3c8bc 100644
* crash we can lose (skip over) as many values as we pre-logged.
*/
-#define SEQ_LOG_VALS 32
+/* Neon XXX: to ensure sequence order of sequence in Zenith we need to WAL log each sequence update. */
+/* Zenith XXX: to ensure sequence order of sequence in Zenith we need to WAL log each sequence update. */
+/* #define SEQ_LOG_VALS 32 */
+#define SEQ_LOG_VALS 0
```
@@ -175,6 +250,66 @@ would be weird if the sequence moved backwards though, think of PITR.
Or add a GUC for the amount to prefix to PostgreSQL, and force it to 1 in Neon.
## Walproposer
```
src/Makefile | 1 +
src/backend/replication/libpqwalproposer/Makefile | 37 +
src/backend/replication/libpqwalproposer/libpqwalproposer.c | 416 ++++++++++++
src/backend/postmaster/bgworker.c | 4 +
src/backend/postmaster/postmaster.c | 6 +
src/backend/replication/Makefile | 4 +-
src/backend/replication/walproposer.c | 2350 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
src/backend/replication/walproposer_utils.c | 402 +++++++++++
src/backend/replication/walreceiver.c | 7 +
src/backend/replication/walsender.c | 320 ++++++---
src/backend/storage/ipc/ipci.c | 6 +
src/include/replication/walproposer.h | 565 ++++++++++++++++
```
WAL proposer is communicating with safekeeper and ensures WAL durability by quorum writes. It is
currently implemented as patch to standard WAL sender.
### How to get rid of the patch
Refactor into an extension. Submit hooks or APIs into upstream if necessary.
@MMeent did some work on this already: https://github.com/neondatabase/postgres/pull/96
## Ignore unexpected data beyond EOF in bufmgr.c
```
@@ -922,11 +928,14 @@ ReadBuffer_common(SMgrRelation smgr, char relpersistence, ForkNumber forkNum,
*/
bufBlock = isLocalBuf ? LocalBufHdrGetBlock(bufHdr) : BufHdrGetBlock(bufHdr);
if (!PageIsNew((Page) bufBlock))
- ereport(ERROR,
+ {
+ // XXX-ZENITH
+ MemSet((char *) bufBlock, 0, BLCKSZ);
+ ereport(DEBUG1,
(errmsg("unexpected data beyond EOF in block %u of relation %s",
blockNum, relpath(smgr->smgr_rnode, forkNum)),
errhint("This has been seen to occur with buggy kernels; consider updating your system.")));
-
+ }
/*
* We *must* do smgrextend before succeeding, else the page will not
* be reserved by the kernel, and the next P_NEW call will decide to
```
PostgreSQL is a bit sloppy with extending relations. Usually, the relation is extended with zeros
first, then the page is filled, and finally the new page WAL-logged. But if multiple backends extend
a relation at the same time, the pages can be WAL-logged in different order.
I'm not sure what scenario exactly required this change in Neon, though.
### How to get rid of the patch
Submit patches to pgsql-hackers, to tighten up the WAL-logging around relation extension. It's a bit
confusing even in PostgreSQL. Maybe WAL log the intention to extend first, then extend the relation,
and finally WAL-log that the extension succeeded.
## Make smgr interface available to extensions
```
@@ -186,8 +321,6 @@ Or add a GUC for the amount to prefix to PostgreSQL, and force it to 1 in Neon.
Submit to upstream. This could be useful for the Disk Encryption patches too, or for compression.
We have submitted this to upstream, but it's moving at glacial a speed.
https://commitfest.postgresql.org/47/4428/
## Added relpersistence argument to smgropen()
@@ -311,148 +444,6 @@ Ignore it. This is only needed for disaster recovery, so once we've eliminated a
patches, we can just keep it around as a patch or as separate branch in a repo.
## pg_waldump flags to ignore errors
After creating a new project or branch in Neon, the first timeline can begin in the middle of a WAL segment. pg_waldump chokes on that, so we added some flags to make it possible to ignore errors.
### How to get rid of the patch
Like previous one, ignore it.
## Backpressure if pageserver doesn't ingest WAL fast enough
```
@@ -3200,6 +3202,7 @@ ProcessInterrupts(void)
return;
InterruptPending = false;
+retry:
if (ProcDiePending)
{
ProcDiePending = false;
@@ -3447,6 +3450,13 @@ ProcessInterrupts(void)
if (ParallelApplyMessagePending)
HandleParallelApplyMessages();
+
+ /* Call registered callback if any */
+ if (ProcessInterruptsCallback)
+ {
+ if (ProcessInterruptsCallback())
+ goto retry;
+ }
}
```
### How to get rid of the patch
Submit a patch to upstream, for a hook in ProcessInterrupts. Could be useful for other extensions
too.
## SLRU on-demand download
```
src/backend/access/transam/slru.c | 105 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-------------
1 file changed, 92 insertions(+), 13 deletions(-)
```
### Problem we're trying to solve
Previously, SLRU files were included in the basebackup, but the total size of them can be large,
several GB, and downloading them all made the startup time too long.
### Alternatives
FUSE hook or LD_PRELOAD trick to intercept the reads on SLRU files
## WAL-log an all-zeros page as one large hole
- In XLogRecordAssemble()
### Problem we're trying to solve
This change was made in v16. Starting with v16, when PostgreSQL extends a relation, it first extends
it with zeros, and it can extend the relation more than one block at a time. The all-zeros page is WAL-ogged, but it's very wasteful to include 8 kB of zeros in the WAL for that. This hack was made so that we WAL logged a compact record with a whole-page "hole". However, PostgreSQL has assertions that prevent that such WAL records from being replayed, so this breaks compatibility such that unmodified PostreSQL cannot process Neon-generated WAL.
### How to get rid of the patch
Find another compact representation for a full-page image of an all-zeros page. A compressed image perhaps.
## Shut down walproposer after checkpointer
```
+ /* Neon: Also allow walproposer background worker to be treated like a WAL sender, so that it's shut down last */
+ if ((bp->bkend_type == BACKEND_TYPE_NORMAL || bp->bkend_type == BACKEND_TYPE_BGWORKER) &&
```
This changes was needed so that postmaster shuts down the walproposer process only after the shutdown checkpoint record is written. Otherwise, the shutdown record will never make it to the safekeepers.
### How to get rid of the patch
Do a bigger refactoring of the postmaster state machine, such that a background worker can specify
the shutdown ordering by itself. The postmaster state machine has grown pretty complicated, and
would benefit from a refactoring for the sake of readability anyway.
## EXPLAIN changes for prefetch and LFC
### How to get rid of the patch
Konstantin submitted a patch to -hackers already: https://commitfest.postgresql.org/47/4643/. Get that into a committable state.
## On-demand download of extensions
### How to get rid of the patch
FUSE or LD_PRELOAD trickery to intercept reads?
## Publication superuser checks
We have hacked CreatePublication so that also neon_superuser can create them.
### How to get rid of the patch
Create an upstream patch with more fine-grained privileges for publications CREATE/DROP that can be GRANTed to users.
## WAL log replication slots
### How to get rid of the patch
Utilize the upcoming v17 "slot sync worker", or a similar neon-specific background worker process, to periodically WAL-log the slots, or to export them somewhere else.
## WAL-log replication snapshots
### How to get rid of the patch
WAL-log them periodically, from a backgound worker.
## WAL-log relmapper files
Similarly to replications snapshot files, the CID mapping files generated during VACUUM FULL of a catalog table are WAL-logged
### How to get rid of the patch
WAL-log them periodically, from a backgound worker.
## XLogWaitForReplayOf()
??
# Not currently committed but proposed
## Disable ring buffer buffer manager strategies
@@ -481,10 +472,23 @@ hint bits are set. Wal logging hint bits updates requires FPI which significantl
Add special WAL record for setting page hints.
## Prefetching
### Why?
As far as pages in Neon are loaded on demand, to reduce node startup time
and also speedup some massive queries we need some mechanism for bulk loading to
reduce page request round-trip overhead.
Currently Postgres is supporting prefetching only for bitmap scan.
In Neon we should also use prefetch for sequential and index scans, because the OS is not doing it for us.
For sequential scan we could prefetch some number of following pages. For index scan we could prefetch pages
of heap relation addressed by TIDs.
## Prewarming
### Why?
Short downtime (or, in other words, fast compute node restart time) is one of the key feature of Neon.
Short downtime (or, in other words, fast compute node restart time) is one of the key feature of Zenith.
But overhead of request-response round-trip for loading pages on demand can make started node warm-up quite slow.
We can capture state of compute node buffer cache and send bulk request for this pages at startup.

View File

@@ -4,18 +4,18 @@
Currently we build two main images:
- [neondatabase/neon](https://hub.docker.com/repository/docker/neondatabase/neon) — image with pre-built `pageserver`, `safekeeper` and `proxy` binaries and all the required runtime dependencies. Built from [/Dockerfile](/Dockerfile).
- [neondatabase/compute-node-v16](https://hub.docker.com/repository/docker/neondatabase/compute-node-v16) — compute node image with pre-built Postgres binaries from [neondatabase/postgres](https://github.com/neondatabase/postgres). Similar images exist for v15 and v14.
- [neondatabase/neon](https://hub.docker.com/repository/docker/zenithdb/zenith) — image with pre-built `pageserver`, `safekeeper` and `proxy` binaries and all the required runtime dependencies. Built from [/Dockerfile](/Dockerfile).
- [neondatabase/compute-node](https://hub.docker.com/repository/docker/zenithdb/compute-node) — compute node image with pre-built Postgres binaries from [neondatabase/postgres](https://github.com/neondatabase/postgres).
And additional intermediate image:
- [neondatabase/compute-tools](https://hub.docker.com/repository/docker/neondatabase/compute-tools) — compute node configuration management tools.
## Build pipeline
## Building pipeline
We build all images after a successful `release` tests run and push automatically to Docker Hub with two parallel CI jobs
1. `neondatabase/compute-tools` and `neondatabase/compute-node-v16` (and -v15 and -v14)
1. `neondatabase/compute-tools` and `neondatabase/compute-node`
2. `neondatabase/neon`
@@ -34,12 +34,12 @@ You can see a [docker compose](https://docs.docker.com/compose/) example to crea
1. create containers
You can specify version of neon cluster using following environment values.
- PG_VERSION: postgres version for compute (default is 16 as of this writing)
- TAG: the tag version of [docker image](https://registry.hub.docker.com/r/neondatabase/neon/tags), which is tagged in [CI test](/.github/workflows/build_and_test.yml). Default is 'latest'
- PG_VERSION: postgres version for compute (default is 14)
- TAG: the tag version of [docker image](https://registry.hub.docker.com/r/neondatabase/neon/tags) (default is latest), which is tagged in [CI test](/.github/workflows/build_and_test.yml)
```
$ cd docker-compose/
$ docker-compose down # remove the containers if exists
$ PG_VERSION=16 TAG=latest docker-compose up --build -d # You can specify the postgres and image version
$ PG_VERSION=15 TAG=2937 docker-compose up --build -d # You can specify the postgres and image version
Creating network "dockercompose_default" with the default driver
Creating docker-compose_storage_broker_1 ... done
(...omit...)
@@ -47,31 +47,29 @@ Creating docker-compose_storage_broker_1 ... done
2. connect compute node
```
$ psql postgresql://cloud_admin:cloud_admin@localhost:55433/postgres
psql (16.3)
Type "help" for help.
$ echo "localhost:55433:postgres:cloud_admin:cloud_admin" >> ~/.pgpass
$ chmod 600 ~/.pgpass
$ psql -h localhost -p 55433 -U cloud_admin
postgres=# CREATE TABLE t(key int primary key, value text);
CREATE TABLE
postgres=# insert into t values(1, 1);
postgres=# insert into t values(1,1);
INSERT 0 1
postgres=# select * from t;
key | value
key | value
-----+-------
1 | 1
(1 row)
```
3. If you want to see the log, you can use `docker-compose logs` command.
```
# check the container name you want to see
$ docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
3582f6d76227 docker-compose_compute "/shell/compute.sh" 2 minutes ago Up 2 minutes 0.0.0.0:3080->3080/tcp, :::3080->3080/tcp, 0.0.0.0:55433->55433/tcp, :::55433->55433/tcp docker-compose_compute_1
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
d6968a5ae912 dockercompose_compute "/shell/compute.sh" 5 minutes ago Up 5 minutes 0.0.0.0:3080->3080/tcp, 0.0.0.0:55433->55433/tcp dockercompose_compute_1
(...omit...)
$ docker logs -f docker-compose_compute_1
$ docker logs -f dockercompose_compute_1
2022-10-21 06:15:48.757 GMT [56] LOG: connection authorized: user=cloud_admin database=postgres application_name=psql
2022-10-21 06:17:00.307 GMT [56] LOG: [NEON_SMGR] libpagestore: connected to 'host=pageserver port=6400'
(...omit...)

View File

@@ -5,3 +5,4 @@ TODO:
- shared across tenants
- store pages from layer files
- store pages from "in-memory layer"
- store materialized pages

View File

@@ -101,12 +101,11 @@ or
```toml
[remote_storage]
container_name = 'some-container-name'
storage_account = 'somestorageaccnt'
container_region = 'us-east'
prefix_in_container = '/test-prefix/'
```
The `AZURE_STORAGE_ACCESS_KEY` env variable can be used to specify the azure credentials if needed.
`AZURE_STORAGE_ACCOUNT` and `AZURE_STORAGE_ACCESS_KEY` env variables can be used to specify the azure credentials if needed.
## Repository background tasks

View File

@@ -134,7 +134,7 @@ depends on that, so if you change it, bad things will happen.
#### page_cache_size
Size of the page cache. Unit is
Size of the page cache, to hold materialized page versions. Unit is
number of 8 kB blocks. The default is 8192, which means 64 MB.
#### max_file_descriptors

View File

@@ -96,6 +96,12 @@ pub struct ComputeSpec {
// Stripe size for pageserver sharding, in pages
#[serde(default)]
pub shard_stripe_size: Option<usize>,
// When we are starting a new replica in hot standby mode,
// we need to know if the primary is running.
// This is used to determine if replica should wait for
// RUNNING_XACTS from primary or not.
pub primary_is_running: Option<bool>,
}
/// Feature flag to signal `compute_ctl` to enable certain experimental functionality.

View File

@@ -209,7 +209,6 @@ pub enum NodeSchedulingPolicy {
Active,
Filling,
Pause,
PauseForRestart,
Draining,
}
@@ -221,7 +220,6 @@ impl FromStr for NodeSchedulingPolicy {
"active" => Ok(Self::Active),
"filling" => Ok(Self::Filling),
"pause" => Ok(Self::Pause),
"pause_for_restart" => Ok(Self::PauseForRestart),
"draining" => Ok(Self::Draining),
_ => Err(anyhow::anyhow!("Unknown scheduling state '{s}'")),
}
@@ -235,7 +233,6 @@ impl From<NodeSchedulingPolicy> for String {
Active => "active",
Filling => "filling",
Pause => "pause",
PauseForRestart => "pause_for_restart",
Draining => "draining",
}
.to_string()

View File

@@ -1,7 +1,7 @@
use anyhow::{bail, Result};
use byteorder::{ByteOrder, BE};
use bytes::BufMut;
use postgres_ffi::relfile_utils::{FSM_FORKNUM, VISIBILITYMAP_FORKNUM};
use postgres_ffi::RepOriginId;
use postgres_ffi::{Oid, TransactionId};
use serde::{Deserialize, Serialize};
use std::{fmt, ops::Range};
@@ -39,9 +39,6 @@ pub const RELATION_SIZE_PREFIX: u8 = 0x61;
/// The key prefix of AUX file keys.
pub const AUX_KEY_PREFIX: u8 = 0x62;
/// The key prefix of ReplOrigin keys.
pub const REPL_ORIGIN_KEY_PREFIX: u8 = 0x63;
/// Check if the key falls in the range of metadata keys.
pub const fn is_metadata_key_slice(key: &[u8]) -> bool {
key[0] >= METADATA_KEY_BEGIN_PREFIX && key[0] < METADATA_KEY_END_PREFIX
@@ -56,8 +53,14 @@ impl Key {
/// Encode a metadata key to a storage key.
pub fn from_metadata_key_fixed_size(key: &[u8; METADATA_KEY_SIZE]) -> Self {
assert!(is_metadata_key_slice(key), "key not in metadata key range");
// Metadata key space ends at 0x7F so it's fine to directly convert it to i128.
Self::from_i128(i128::from_be_bytes(*key))
Key {
field1: key[0],
field2: u16::from_be_bytes(key[1..3].try_into().unwrap()) as u32,
field3: u32::from_be_bytes(key[3..7].try_into().unwrap()),
field4: u32::from_be_bytes(key[7..11].try_into().unwrap()),
field5: key[11],
field6: u32::from_be_bytes(key[12..16].try_into().unwrap()),
}
}
/// Encode a metadata key to a storage key.
@@ -65,6 +68,17 @@ impl Key {
Self::from_metadata_key_fixed_size(key.try_into().expect("expect 16 byte metadata key"))
}
/// Extract a metadata key to a writer. The result should always be 16 bytes.
pub fn extract_metadata_key_to_writer(&self, mut writer: impl BufMut) {
writer.put_u8(self.field1);
assert!(self.field2 <= 0xFFFF);
writer.put_u16(self.field2 as u16);
writer.put_u32(self.field3);
writer.put_u32(self.field4);
writer.put_u8(self.field5);
writer.put_u32(self.field6);
}
/// Get the range of metadata keys.
pub const fn metadata_key_range() -> Range<Self> {
Key {
@@ -107,7 +121,7 @@ impl Key {
/// As long as Neon does not support tablespace (because of lack of access to local file system),
/// we can assume that only some predefined namespace OIDs are used which can fit in u16
pub fn to_i128(&self) -> i128 {
assert!(self.field2 <= 0xFFFF || self.field2 == 0xFFFFFFFF || self.field2 == 0x22222222);
assert!(self.field2 < 0xFFFF || self.field2 == 0xFFFFFFFF || self.field2 == 0x22222222);
(((self.field1 & 0x7F) as i128) << 120)
| (((self.field2 & 0xFFFF) as i128) << 104)
| ((self.field3 as i128) << 72)
@@ -160,9 +174,8 @@ impl Key {
key
}
/// Convert a 18B slice to a key. This function should not be used for 16B metadata keys because `field2` is handled differently.
/// Use [`Key::from_i128`] instead if you want to handle 16B keys (i.e., metadata keys). There are some restrictions on `field2`,
/// and therefore not all 18B slices are valid page server keys.
/// Convert a 18B slice to a key. This function should not be used for metadata keys because field2 is handled differently.
/// Use [`Key::from_metadata_key`] instead.
pub fn from_slice(b: &[u8]) -> Self {
Key {
field1: b[0],
@@ -174,8 +187,8 @@ impl Key {
}
}
/// Convert a key to a 18B slice. This function should not be used for getting a 16B metadata key because `field2` is handled differently.
/// Use [`Key::to_i128`] instead if you want to get a 16B key (i.e., metadata keys).
/// Convert a key to a 18B slice. This function should not be used for metadata keys because field2 is handled differently.
/// Use [`Key::extract_metadata_key_to_writer`] instead.
pub fn write_to_byte_slice(&self, buf: &mut [u8]) {
buf[0] = self.field1;
BE::write_u32(&mut buf[1..5], self.field2);
@@ -386,14 +399,7 @@ pub fn rel_size_to_key(rel: RelTag) -> Key {
field3: rel.dbnode,
field4: rel.relnode,
field5: rel.forknum,
field6: 0xffff_ffff,
}
}
impl Key {
#[inline(always)]
pub fn is_rel_size_key(&self) -> bool {
self.field1 == 0 && self.field6 == u32::MAX
field6: 0xffffffff,
}
}
@@ -434,25 +440,6 @@ pub fn slru_dir_to_key(kind: SlruKind) -> Key {
}
}
#[inline(always)]
pub fn slru_dir_kind(key: &Key) -> Option<Result<SlruKind, u32>> {
if key.field1 == 0x01
&& key.field3 == 0
&& key.field4 == 0
&& key.field5 == 0
&& key.field6 == 0
{
match key.field2 {
0 => Some(Ok(SlruKind::Clog)),
1 => Some(Ok(SlruKind::MultiXactMembers)),
2 => Some(Ok(SlruKind::MultiXactOffsets)),
x => Some(Err(x)),
}
} else {
None
}
}
#[inline(always)]
pub fn slru_block_to_key(kind: SlruKind, segno: u32, blknum: BlockNumber) -> Key {
Key {
@@ -481,17 +468,7 @@ pub fn slru_segment_size_to_key(kind: SlruKind, segno: u32) -> Key {
field3: 1,
field4: segno,
field5: 0,
field6: 0xffff_ffff,
}
}
impl Key {
pub fn is_slru_segment_size_key(&self) -> bool {
self.field1 == 0x01
&& self.field2 < 0x03
&& self.field3 == 0x01
&& self.field5 == 0
&& self.field6 == u32::MAX
field6: 0xffffffff,
}
}
@@ -592,37 +569,6 @@ pub const AUX_FILES_KEY: Key = Key {
field6: 2,
};
#[inline(always)]
pub fn repl_origin_key(origin_id: RepOriginId) -> Key {
Key {
field1: REPL_ORIGIN_KEY_PREFIX,
field2: 0,
field3: 0,
field4: 0,
field5: 0,
field6: origin_id as u32,
}
}
/// Get the range of replorigin keys.
pub fn repl_origin_key_range() -> Range<Key> {
Key {
field1: REPL_ORIGIN_KEY_PREFIX,
field2: 0,
field3: 0,
field4: 0,
field5: 0,
field6: 0,
}..Key {
field1: REPL_ORIGIN_KEY_PREFIX,
field2: 0,
field3: 0,
field4: 0,
field5: 0,
field6: 0x10000,
}
}
// Reverse mappings for a few Keys.
// These are needed by WAL redo manager.
@@ -631,78 +577,73 @@ pub const NON_INHERITED_RANGE: Range<Key> = AUX_FILES_KEY..AUX_FILES_KEY.next();
/// Sparse keyspace range for vectored get. Missing key error will be ignored for this range.
pub const NON_INHERITED_SPARSE_RANGE: Range<Key> = Key::metadata_key_range();
impl Key {
// AUX_FILES currently stores only data for logical replication (slots etc), and
// we don't preserve these on a branch because safekeepers can't follow timeline
// switch (and generally it likely should be optional), so ignore these.
#[inline(always)]
pub fn is_inherited_key(self) -> bool {
!NON_INHERITED_RANGE.contains(&self) && !NON_INHERITED_SPARSE_RANGE.contains(&self)
}
// AUX_FILES currently stores only data for logical replication (slots etc), and
// we don't preserve these on a branch because safekeepers can't follow timeline
// switch (and generally it likely should be optional), so ignore these.
#[inline(always)]
pub fn is_inherited_key(key: Key) -> bool {
!NON_INHERITED_RANGE.contains(&key) && !NON_INHERITED_SPARSE_RANGE.contains(&key)
}
#[inline(always)]
pub fn is_rel_fsm_block_key(self) -> bool {
self.field1 == 0x00
&& self.field4 != 0
&& self.field5 == FSM_FORKNUM
&& self.field6 != 0xffffffff
}
#[inline(always)]
pub fn is_rel_fsm_block_key(key: Key) -> bool {
key.field1 == 0x00 && key.field4 != 0 && key.field5 == FSM_FORKNUM && key.field6 != 0xffffffff
}
#[inline(always)]
pub fn is_rel_vm_block_key(self) -> bool {
self.field1 == 0x00
&& self.field4 != 0
&& self.field5 == VISIBILITYMAP_FORKNUM
&& self.field6 != 0xffffffff
}
#[inline(always)]
pub fn is_rel_vm_block_key(key: Key) -> bool {
key.field1 == 0x00
&& key.field4 != 0
&& key.field5 == VISIBILITYMAP_FORKNUM
&& key.field6 != 0xffffffff
}
#[inline(always)]
pub fn to_slru_block(self) -> anyhow::Result<(SlruKind, u32, BlockNumber)> {
Ok(match self.field1 {
0x01 => {
let kind = match self.field2 {
0x00 => SlruKind::Clog,
0x01 => SlruKind::MultiXactMembers,
0x02 => SlruKind::MultiXactOffsets,
_ => anyhow::bail!("unrecognized slru kind 0x{:02x}", self.field2),
};
let segno = self.field4;
let blknum = self.field6;
#[inline(always)]
pub fn key_to_slru_block(key: Key) -> anyhow::Result<(SlruKind, u32, BlockNumber)> {
Ok(match key.field1 {
0x01 => {
let kind = match key.field2 {
0x00 => SlruKind::Clog,
0x01 => SlruKind::MultiXactMembers,
0x02 => SlruKind::MultiXactOffsets,
_ => anyhow::bail!("unrecognized slru kind 0x{:02x}", key.field2),
};
let segno = key.field4;
let blknum = key.field6;
(kind, segno, blknum)
}
_ => anyhow::bail!("unexpected value kind 0x{:02x}", self.field1),
})
}
(kind, segno, blknum)
}
_ => anyhow::bail!("unexpected value kind 0x{:02x}", key.field1),
})
}
#[inline(always)]
pub fn is_slru_block_key(self) -> bool {
self.field1 == 0x01 // SLRU-related
&& self.field3 == 0x00000001 // but not SlruDir
&& self.field6 != 0xffffffff // and not SlruSegSize
}
#[inline(always)]
pub fn is_slru_block_key(key: Key) -> bool {
key.field1 == 0x01 // SLRU-related
&& key.field3 == 0x00000001 // but not SlruDir
&& key.field6 != 0xffffffff // and not SlruSegSize
}
#[inline(always)]
pub fn is_rel_block_key(&self) -> bool {
self.field1 == 0x00 && self.field4 != 0 && self.field6 != 0xffffffff
}
#[inline(always)]
pub fn is_rel_block_key(key: &Key) -> bool {
key.field1 == 0x00 && key.field4 != 0 && key.field6 != 0xffffffff
}
/// Guaranteed to return `Ok()` if [`Self::is_rel_block_key`] returns `true` for `key`.
#[inline(always)]
pub fn to_rel_block(self) -> anyhow::Result<(RelTag, BlockNumber)> {
Ok(match self.field1 {
0x00 => (
RelTag {
spcnode: self.field2,
dbnode: self.field3,
relnode: self.field4,
forknum: self.field5,
},
self.field6,
),
_ => anyhow::bail!("unexpected value kind 0x{:02x}", self.field1),
})
}
/// Guaranteed to return `Ok()` if [[is_rel_block_key]] returns `true` for `key`.
#[inline(always)]
pub fn key_to_rel_block(key: Key) -> anyhow::Result<(RelTag, BlockNumber)> {
Ok(match key.field1 {
0x00 => (
RelTag {
spcnode: key.field2,
dbnode: key.field3,
relnode: key.field4,
forknum: key.field5,
},
key.field6,
),
_ => anyhow::bail!("unexpected value kind 0x{:02x}", key.field1),
})
}
impl std::str::FromStr for Key {
@@ -746,15 +687,10 @@ mod tests {
let mut metadata_key = vec![AUX_KEY_PREFIX];
metadata_key.extend_from_slice(&[0xFF; 15]);
let encoded_key = Key::from_metadata_key(&metadata_key);
let output_key = encoded_key.to_i128().to_be_bytes();
let mut output_key = Vec::new();
encoded_key.extract_metadata_key_to_writer(&mut output_key);
assert_eq!(metadata_key, output_key);
assert!(encoded_key.is_metadata_key());
assert!(is_metadata_key_slice(&metadata_key));
}
#[test]
fn test_possible_largest_key() {
Key::from_i128(0x7FFF_FFFF_FFFF_FFFF_FFFF_FFFF_FFFF_FFFF);
// TODO: put this key into the system and see if anything breaks.
}
}

View File

@@ -558,12 +558,6 @@ impl KeySpaceRandomAccum {
self.ranges.push(range);
}
pub fn add_keyspace(&mut self, keyspace: KeySpace) {
for range in keyspace.ranges {
self.add_range(range);
}
}
pub fn to_keyspace(mut self) -> KeySpace {
let mut ranges = Vec::new();
if !self.ranges.is_empty() {

View File

@@ -9,6 +9,7 @@ use std::{
collections::HashMap,
io::{BufRead, Read},
num::{NonZeroU64, NonZeroUsize},
str::FromStr,
sync::atomic::AtomicUsize,
time::{Duration, SystemTime},
};
@@ -177,20 +178,6 @@ serde_with::serde_conv!(
|value: String| -> Result<_, humantime::TimestampError> { humantime::parse_rfc3339(&value) }
);
impl LsnLease {
/// The default length for an explicit LSN lease request (10 minutes).
pub const DEFAULT_LENGTH: Duration = Duration::from_secs(10 * 60);
/// The default length for an implicit LSN lease granted during
/// `get_lsn_by_timestamp` request (1 minutes).
pub const DEFAULT_LENGTH_FOR_TS: Duration = Duration::from_secs(60);
/// Checks whether the lease is expired.
pub fn is_expired(&self, now: &SystemTime) -> bool {
now > &self.valid_until
}
}
/// The only [`TenantState`] variants we could be `TenantState::Activating` from.
#[derive(Clone, Copy, Debug, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
pub enum ActivatingFrom {
@@ -293,6 +280,22 @@ pub struct TenantCreateRequest {
pub config: TenantConfig, // as we have a flattened field, we should reject all unknown fields in it
}
#[derive(Deserialize, Debug)]
#[serde(deny_unknown_fields)]
pub struct TenantLoadRequest {
#[serde(default)]
#[serde(skip_serializing_if = "Option::is_none")]
pub generation: Option<u32>,
}
impl std::ops::Deref for TenantCreateRequest {
type Target = TenantConfig;
fn deref(&self) -> &Self::Target {
&self.config
}
}
/// An alternative representation of `pageserver::tenant::TenantConf` with
/// simpler types.
#[derive(Serialize, Deserialize, Debug, Default, Clone, Eq, PartialEq)]
@@ -303,7 +306,7 @@ pub struct TenantConfig {
pub compaction_period: Option<String>,
pub compaction_threshold: Option<usize>,
// defer parsing compaction_algorithm, like eviction_policy
pub compaction_algorithm: Option<CompactionAlgorithmSettings>,
pub compaction_algorithm: Option<CompactionAlgorithm>,
pub gc_horizon: Option<u64>,
pub gc_period: Option<String>,
pub image_creation_threshold: Option<usize>,
@@ -320,8 +323,6 @@ pub struct TenantConfig {
pub timeline_get_throttle: Option<ThrottleConfig>,
pub image_layer_creation_check_threshold: Option<u8>,
pub switch_aux_file_policy: Option<AuxFilePolicy>,
pub lsn_lease_length: Option<String>,
pub lsn_lease_length_for_ts: Option<String>,
}
/// The policy for the aux file storage. It can be switched through `switch_aux_file_policy`
@@ -333,28 +334,14 @@ pub struct TenantConfig {
/// Unset -> V1
/// -> V2
/// -> CrossValidation -> V2
#[derive(
Eq,
PartialEq,
Debug,
Copy,
Clone,
strum_macros::EnumString,
strum_macros::Display,
serde_with::DeserializeFromStr,
serde_with::SerializeDisplay,
)]
#[strum(serialize_all = "kebab-case")]
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
pub enum AuxFilePolicy {
/// V1 aux file policy: store everything in AUX_FILE_KEY
#[strum(ascii_case_insensitive)]
V1,
/// V2 aux file policy: store in the AUX_FILE keyspace
#[strum(ascii_case_insensitive)]
V2,
/// Cross validation runs both formats on the write path and does validation
/// on the read path.
#[strum(ascii_case_insensitive)]
CrossValidation,
}
@@ -420,6 +407,23 @@ impl AuxFilePolicy {
}
}
impl FromStr for AuxFilePolicy {
type Err = anyhow::Error;
fn from_str(s: &str) -> Result<Self, Self::Err> {
let s = s.to_lowercase();
if s == "v1" {
Ok(Self::V1)
} else if s == "v2" {
Ok(Self::V2)
} else if s == "crossvalidation" || s == "cross_validation" {
Ok(Self::CrossValidation)
} else {
anyhow::bail!("cannot parse {} to aux file policy", s)
}
}
}
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
#[serde(tag = "kind")]
pub enum EvictionPolicy {
@@ -438,28 +442,13 @@ impl EvictionPolicy {
}
}
#[derive(
Eq,
PartialEq,
Debug,
Copy,
Clone,
strum_macros::EnumString,
strum_macros::Display,
serde_with::DeserializeFromStr,
serde_with::SerializeDisplay,
)]
#[strum(serialize_all = "kebab-case")]
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
#[serde(tag = "kind")]
pub enum CompactionAlgorithm {
Legacy,
Tiered,
}
#[derive(Eq, PartialEq, Debug, Clone, Serialize, Deserialize)]
pub struct CompactionAlgorithmSettings {
pub kind: CompactionAlgorithm,
}
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
pub struct EvictionPolicyLayerAccessThreshold {
#[serde(with = "humantime_serde")]
@@ -607,6 +596,31 @@ impl TenantConfigRequest {
}
}
#[derive(Debug, Deserialize)]
pub struct TenantAttachRequest {
#[serde(default)]
pub config: TenantAttachConfig,
#[serde(default)]
pub generation: Option<u32>,
}
/// Newtype to enforce deny_unknown_fields on TenantConfig for
/// its usage inside `TenantAttachRequest`.
#[derive(Debug, Serialize, Deserialize, Default)]
#[serde(deny_unknown_fields)]
pub struct TenantAttachConfig {
#[serde(flatten)]
allowing_unknown_fields: TenantConfig,
}
impl std::ops::Deref for TenantAttachConfig {
type Target = TenantConfig;
fn deref(&self) -> &Self::Target {
&self.allowing_unknown_fields
}
}
/// See [`TenantState::attachment_status`] and the OpenAPI docs for context.
#[derive(Serialize, Deserialize, Clone)]
#[serde(tag = "slug", content = "data", rename_all = "snake_case")]
@@ -625,7 +639,8 @@ pub struct TenantInfo {
/// If a layer is present in both local FS and S3, it counts only once.
pub current_physical_size: Option<u64>, // physical size is only included in `tenant_status` endpoint
pub attachment_status: TenantAttachmentStatus,
pub generation: u32,
#[serde(skip_serializing_if = "Option::is_none")]
pub generation: Option<u32>,
}
#[derive(Serialize, Deserialize, Clone)]
@@ -1390,7 +1405,6 @@ impl PagestreamBeMessage {
#[cfg(test)]
mod tests {
use serde_json::json;
use std::str::FromStr;
use super::*;
@@ -1452,7 +1466,7 @@ mod tests {
state: TenantState::Active,
current_physical_size: Some(42),
attachment_status: TenantAttachmentStatus::Attached,
generation: 1,
generation: None,
};
let expected_active = json!({
"id": original_active.id.to_string(),
@@ -1462,8 +1476,7 @@ mod tests {
"current_physical_size": 42,
"attachment_status": {
"slug":"attached",
},
"generation" : 1
}
});
let original_broken = TenantInfo {
@@ -1474,7 +1487,7 @@ mod tests {
},
current_physical_size: Some(42),
attachment_status: TenantAttachmentStatus::Attached,
generation: 1,
generation: None,
};
let expected_broken = json!({
"id": original_broken.id.to_string(),
@@ -1488,8 +1501,7 @@ mod tests {
"current_physical_size": 42,
"attachment_status": {
"slug":"attached",
},
"generation" : 1
}
});
assert_eq!(
@@ -1530,6 +1542,18 @@ mod tests {
"expect unknown field `unknown_field` error, got: {}",
err
);
let attach_request = json!({
"config": {
"unknown_field": "unknown_value".to_string(),
},
});
let err = serde_json::from_value::<TenantAttachRequest>(attach_request).unwrap_err();
assert!(
err.to_string().contains("unknown field `unknown_field`"),
"expect unknown field `unknown_field` error, got: {}",
err
);
}
#[test]
@@ -1643,14 +1667,4 @@ mod tests {
AuxFilePolicy::V2
));
}
#[test]
fn test_aux_parse() {
assert_eq!(AuxFilePolicy::from_str("V2").unwrap(), AuxFilePolicy::V2);
assert_eq!(AuxFilePolicy::from_str("v2").unwrap(), AuxFilePolicy::V2);
assert_eq!(
AuxFilePolicy::from_str("cross-validation").unwrap(),
AuxFilePolicy::CrossValidation
);
}
}

View File

@@ -3,7 +3,7 @@ use std::cmp::Ordering;
use std::fmt;
use postgres_ffi::pg_constants::GLOBALTABLESPACE_OID;
use postgres_ffi::relfile_utils::{forkname_to_number, forknumber_to_name, MAIN_FORKNUM};
use postgres_ffi::relfile_utils::forknumber_to_name;
use postgres_ffi::Oid;
///
@@ -68,57 +68,6 @@ impl fmt::Display for RelTag {
}
}
#[derive(Debug, thiserror::Error)]
pub enum ParseRelTagError {
#[error("invalid forknum")]
InvalidForknum(#[source] std::num::ParseIntError),
#[error("missing triplet member {}", .0)]
MissingTripletMember(usize),
#[error("invalid triplet member {}", .0)]
InvalidTripletMember(usize, #[source] std::num::ParseIntError),
}
impl std::str::FromStr for RelTag {
type Err = ParseRelTagError;
fn from_str(s: &str) -> Result<Self, Self::Err> {
use ParseRelTagError::*;
// FIXME: in postgres logs this separator is dot
// Example:
// could not read block 2 in rel 1663/208101/2620.1 from page server at lsn 0/2431E6F0
// with a regex we could get this more painlessly
let (triplet, forknum) = match s.split_once('_').or_else(|| s.split_once('.')) {
Some((t, f)) => {
let forknum = forkname_to_number(Some(f));
let forknum = if let Ok(f) = forknum {
f
} else {
f.parse::<u8>().map_err(InvalidForknum)?
};
(t, Some(forknum))
}
None => (s, None),
};
let mut split = triplet
.splitn(3, '/')
.enumerate()
.map(|(i, s)| s.parse::<u32>().map_err(|e| InvalidTripletMember(i, e)));
let spcnode = split.next().ok_or(MissingTripletMember(0))??;
let dbnode = split.next().ok_or(MissingTripletMember(1))??;
let relnode = split.next().ok_or(MissingTripletMember(2))??;
Ok(RelTag {
spcnode,
forknum: forknum.unwrap_or(MAIN_FORKNUM),
dbnode,
relnode,
})
}
}
impl RelTag {
pub fn to_segfile_name(&self, segno: u32) -> String {
let mut name = if self.spcnode == GLOBALTABLESPACE_OID {

View File

@@ -1,6 +1,9 @@
use std::{ops::RangeInclusive, str::FromStr};
use crate::{key::Key, models::ShardParameters};
use crate::{
key::{is_rel_block_key, Key},
models::ShardParameters,
};
use hex::FromHex;
use postgres_ffi::relfile_utils::INIT_FORKNUM;
use serde::{Deserialize, Serialize};
@@ -425,12 +428,6 @@ impl<'de> Deserialize<'de> for TenantShardId {
#[derive(Clone, Copy, Serialize, Deserialize, Eq, PartialEq, Debug)]
pub struct ShardStripeSize(pub u32);
impl Default for ShardStripeSize {
fn default() -> Self {
DEFAULT_STRIPE_SIZE
}
}
/// Layout version: for future upgrades where we might change how the key->shard mapping works
#[derive(Clone, Copy, Serialize, Deserialize, Eq, PartialEq, Debug)]
pub struct ShardLayout(u8);
@@ -669,7 +666,7 @@ fn key_is_shard0(key: &Key) -> bool {
// because they must be included in basebackups.
let is_initfork = key.field5 == INIT_FORKNUM;
!key.is_rel_block_key() || is_initfork
!is_rel_block_key(key) || is_initfork
}
/// Provide the same result as the function in postgres `hashfn.h` with the same name
@@ -716,25 +713,6 @@ fn key_to_shard_number(count: ShardCount, stripe_size: ShardStripeSize, key: &Ke
ShardNumber((hash % count.0 as u32) as u8)
}
/// For debugging, while not exposing the internals.
#[derive(Debug)]
#[allow(unused)] // used by debug formatting by pagectl
struct KeyShardingInfo {
shard0: bool,
shard_number: ShardNumber,
}
pub fn describe(
key: &Key,
shard_count: ShardCount,
stripe_size: ShardStripeSize,
) -> impl std::fmt::Debug {
KeyShardingInfo {
shard0: key_is_shard0(key),
shard_number: key_to_shard_number(shard_count, stripe_size, key),
}
}
#[cfg(test)]
mod tests {
use utils::Hex;

View File

@@ -178,13 +178,6 @@ impl PgConnectionConfig {
}
}
impl fmt::Display for PgConnectionConfig {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
// The password is intentionally hidden and not part of this display string.
write!(f, "postgresql://{}:{}", self.host, self.port)
}
}
impl fmt::Debug for PgConnectionConfig {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
// We want `password: Some(REDACTED-STRING)`, not `password: Some("REDACTED-STRING")`

View File

@@ -126,7 +126,6 @@ fn main() -> anyhow::Result<()> {
.allowlist_type("PageHeaderData")
.allowlist_type("DBState")
.allowlist_type("RelMapFile")
.allowlist_type("RepOriginId")
// Because structs are used for serialization, tell bindgen to emit
// explicit padding fields.
.explicit_padding(true)

View File

@@ -110,7 +110,6 @@ pub mod pg_constants;
pub mod relfile_utils;
// Export some widely used datatypes that are unlikely to change across Postgres versions
pub use v14::bindings::RepOriginId;
pub use v14::bindings::{uint32, uint64, Oid};
pub use v14::bindings::{BlockNumber, OffsetNumber};
pub use v14::bindings::{MultiXactId, TransactionId};

View File

@@ -102,7 +102,7 @@ pub const XACT_XINFO_HAS_SUBXACTS: u32 = 1u32 << 1;
pub const XACT_XINFO_HAS_RELFILENODES: u32 = 1u32 << 2;
pub const XACT_XINFO_HAS_INVALS: u32 = 1u32 << 3;
pub const XACT_XINFO_HAS_TWOPHASE: u32 = 1u32 << 4;
pub const XACT_XINFO_HAS_ORIGIN: u32 = 1u32 << 5;
// pub const XACT_XINFO_HAS_ORIGIN: u32 = 1u32 << 5;
// pub const XACT_XINFO_HAS_AE_LOCKS: u32 = 1u32 << 6;
// pub const XACT_XINFO_HAS_GID: u32 = 1u32 << 7;
@@ -167,7 +167,6 @@ pub const RM_RELMAP_ID: u8 = 7;
pub const RM_STANDBY_ID: u8 = 8;
pub const RM_HEAP2_ID: u8 = 9;
pub const RM_HEAP_ID: u8 = 10;
pub const RM_REPLORIGIN_ID: u8 = 19;
pub const RM_LOGICALMSG_ID: u8 = 21;
// from neon_rmgr.h
@@ -224,10 +223,6 @@ pub const XLOG_CHECKPOINT_ONLINE: u8 = 0x10;
pub const XLP_FIRST_IS_CONTRECORD: u16 = 0x0001;
pub const XLP_LONG_HEADER: u16 = 0x0002;
/* From xlog.h */
pub const XLOG_REPLORIGIN_SET: u8 = 0x00;
pub const XLOG_REPLORIGIN_DROP: u8 = 0x10;
/* From replication/slot.h */
pub const REPL_SLOT_ON_DISK_OFFSETOF_RESTART_LSN: usize = 4*4 /* offset of `slotdata` in ReplicationSlotOnDisk */
+ 64 /* NameData */ + 4*4;
@@ -242,9 +237,6 @@ pub const SLOTS_PER_FSM_PAGE: u32 = FSM_LEAF_NODES_PER_PAGE as u32;
pub const VM_HEAPBLOCKS_PER_PAGE: u32 =
(BLCKSZ as usize - SIZEOF_PAGE_HEADER_DATA) as u32 * (8 / 2); // MAPSIZE * (BITS_PER_BYTE / BITS_PER_HEAPBLOCK)
/* From origin.c */
pub const REPLICATION_STATE_MAGIC: u32 = 0x1257DADE;
// List of subdirectories inside pgdata.
// Copied from src/bin/initdb/initdb.c
pub const PGDATA_SUBDIRS: [&str; 22] = [

View File

@@ -7,7 +7,6 @@ license.workspace = true
[dependencies]
bytes.workspace = true
byteorder.workspace = true
itertools.workspace = true
pin-project-lite.workspace = true
postgres-protocol.workspace = true
rand.workspace = true

View File

@@ -7,9 +7,8 @@ pub mod framed;
use byteorder::{BigEndian, ReadBytesExt};
use bytes::{Buf, BufMut, Bytes, BytesMut};
use itertools::Itertools;
use serde::{Deserialize, Serialize};
use std::{borrow::Cow, fmt, io, str};
use std::{borrow::Cow, collections::HashMap, fmt, io, str};
// re-export for use in utils pageserver_feedback.rs
pub use postgres_protocol::PG_EPOCH;
@@ -51,37 +50,15 @@ pub enum FeStartupPacket {
},
}
#[derive(Debug, Clone, Default)]
pub struct StartupMessageParamsBuilder {
params: BytesMut,
}
impl StartupMessageParamsBuilder {
/// Set parameter's value by its name.
/// name and value must not contain a \0 byte
pub fn insert(&mut self, name: &str, value: &str) {
self.params.put(name.as_bytes());
self.params.put(&b"\0"[..]);
self.params.put(value.as_bytes());
self.params.put(&b"\0"[..]);
}
pub fn freeze(self) -> StartupMessageParams {
StartupMessageParams {
params: self.params.freeze(),
}
}
}
#[derive(Debug, Clone, Default)]
#[derive(Debug)]
pub struct StartupMessageParams {
params: Bytes,
params: HashMap<String, String>,
}
impl StartupMessageParams {
/// Get parameter's value by its name.
pub fn get(&self, name: &str) -> Option<&str> {
self.iter().find_map(|(k, v)| (k == name).then_some(v))
self.params.get(name).map(|s| s.as_str())
}
/// Split command-line options according to PostgreSQL's logic,
@@ -135,19 +112,15 @@ impl StartupMessageParams {
/// Iterate through key-value pairs in an arbitrary order.
pub fn iter(&self) -> impl Iterator<Item = (&str, &str)> {
let params =
std::str::from_utf8(&self.params).expect("should be validated as utf8 already");
params.split_terminator('\0').tuples()
self.params.iter().map(|(k, v)| (k.as_str(), v.as_str()))
}
// This function is mostly useful in tests.
#[doc(hidden)]
pub fn new<'a, const N: usize>(pairs: [(&'a str, &'a str); N]) -> Self {
let mut b = StartupMessageParamsBuilder::default();
for (k, v) in pairs {
b.insert(k, v)
Self {
params: pairs.map(|(k, v)| (k.to_owned(), v.to_owned())).into(),
}
b.freeze()
}
}
@@ -372,21 +345,35 @@ impl FeStartupPacket {
(major_version, minor_version) => {
// StartupMessage
let s = str::from_utf8(&msg).map_err(|_e| {
ProtocolError::BadMessage("StartupMessage params: invalid utf-8".to_owned())
})?;
let s = s.strip_suffix('\0').ok_or_else(|| {
ProtocolError::Protocol(
"StartupMessage params: missing null terminator".to_string(),
)
})?;
// Parse pairs of null-terminated strings (key, value).
// See `postgres: ProcessStartupPacket, build_startup_packet`.
let mut tokens = str::from_utf8(&msg)
.map_err(|_e| {
ProtocolError::BadMessage("StartupMessage params: invalid utf-8".to_owned())
})?
.strip_suffix('\0') // drop packet's own null
.ok_or_else(|| {
ProtocolError::Protocol(
"StartupMessage params: missing null terminator".to_string(),
)
})?
.split_terminator('\0');
let mut params = HashMap::new();
while let Some(name) = tokens.next() {
let value = tokens.next().ok_or_else(|| {
ProtocolError::Protocol(
"StartupMessage params: key without value".to_string(),
)
})?;
params.insert(name.to_owned(), value.to_owned());
}
FeStartupPacket::StartupMessage {
major_version,
minor_version,
params: StartupMessageParams {
params: msg.slice_ref(s.as_bytes()),
},
params: StartupMessageParams { params },
}
}
};

View File

@@ -14,9 +14,8 @@ aws-config.workspace = true
aws-sdk-s3.workspace = true
aws-credential-types.workspace = true
bytes.workspace = true
camino = { workspace = true, features = ["serde1"] }
camino.workspace = true
humantime.workspace = true
humantime-serde.workspace = true
hyper = { workspace = true, features = ["stream"] }
futures.workspace = true
rand.workspace = true

View File

@@ -3,7 +3,6 @@
use std::borrow::Cow;
use std::collections::HashMap;
use std::env;
use std::fmt::Display;
use std::io;
use std::num::NonZeroU32;
use std::pin::Pin;
@@ -27,15 +26,14 @@ use futures::stream::Stream;
use futures_util::StreamExt;
use futures_util::TryStreamExt;
use http_types::{StatusCode, Url};
use scopeguard::ScopeGuard;
use tokio_util::sync::CancellationToken;
use tracing::debug;
use utils::backoff;
use crate::metrics::{start_measuring_requests, AttemptOutcome, RequestKind};
use crate::RemoteStorageActivity;
use crate::{
config::AzureConfig, error::Cancelled, ConcurrencyLimiter, Download, DownloadError, Listing,
ListingMode, RemotePath, RemoteStorage, StorageMetadata, TimeTravelError, TimeoutOrCancel,
error::Cancelled, s3_bucket::RequestKind, AzureConfig, ConcurrencyLimiter, Download,
DownloadError, Listing, ListingMode, RemotePath, RemoteStorage, StorageMetadata,
TimeTravelError, TimeoutOrCancel,
};
pub struct AzureBlobStorage {
@@ -54,10 +52,7 @@ impl AzureBlobStorage {
azure_config.container_name
);
// Use the storage account from the config by default, fall back to env var if not present.
let account = azure_config.storage_account.clone().unwrap_or_else(|| {
env::var("AZURE_STORAGE_ACCOUNT").expect("missing AZURE_STORAGE_ACCOUNT")
});
let account = env::var("AZURE_STORAGE_ACCOUNT").expect("missing AZURE_STORAGE_ACCOUNT");
// If the `AZURE_STORAGE_ACCESS_KEY` env var has an access key, use that,
// otherwise try the token based credentials.
@@ -143,8 +138,6 @@ impl AzureBlobStorage {
let mut last_modified = None;
let mut metadata = HashMap::new();
let started_at = start_measuring_requests(kind);
let download = async {
let response = builder
// convert to concrete Pageable
@@ -208,22 +201,13 @@ impl AzureBlobStorage {
})
};
let download = tokio::select! {
tokio::select! {
bufs = download => bufs,
cancel_or_timeout = cancel_or_timeout => match cancel_or_timeout {
TimeoutOrCancel::Timeout => return Err(DownloadError::Timeout),
TimeoutOrCancel::Cancel => return Err(DownloadError::Cancelled),
TimeoutOrCancel::Timeout => Err(DownloadError::Timeout),
TimeoutOrCancel::Cancel => Err(DownloadError::Cancelled),
},
};
let started_at = ScopeGuard::into_inner(started_at);
let outcome = match &download {
Ok(_) => AttemptOutcome::Ok,
Err(_) => AttemptOutcome::Err,
};
crate::metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, outcome, started_at);
download
}
}
async fn permit(
@@ -357,10 +341,7 @@ impl RemoteStorage for AzureBlobStorage {
metadata: Option<StorageMetadata>,
cancel: &CancellationToken,
) -> anyhow::Result<()> {
let kind = RequestKind::Put;
let _permit = self.permit(kind, cancel).await?;
let started_at = start_measuring_requests(kind);
let _permit = self.permit(RequestKind::Put, cancel).await?;
let op = async {
let blob_client = self.client.blob_client(self.relative_path_to_name(to));
@@ -384,25 +365,14 @@ impl RemoteStorage for AzureBlobStorage {
match fut.await {
Ok(Ok(_response)) => Ok(()),
Ok(Err(azure)) => Err(azure.into()),
Err(_timeout) => Err(TimeoutOrCancel::Timeout.into()),
Err(_timeout) => Err(TimeoutOrCancel::Cancel.into()),
}
};
let res = tokio::select! {
tokio::select! {
res = op => res,
_ = cancel.cancelled() => return Err(TimeoutOrCancel::Cancel.into()),
};
let outcome = match res {
Ok(_) => AttemptOutcome::Ok,
Err(_) => AttemptOutcome::Err,
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, outcome, started_at);
res
_ = cancel.cancelled() => Err(TimeoutOrCancel::Cancel.into()),
}
}
async fn download(
@@ -448,79 +418,40 @@ impl RemoteStorage for AzureBlobStorage {
paths: &'a [RemotePath],
cancel: &CancellationToken,
) -> anyhow::Result<()> {
let kind = RequestKind::Delete;
let _permit = self.permit(kind, cancel).await?;
let started_at = start_measuring_requests(kind);
let _permit = self.permit(RequestKind::Delete, cancel).await?;
let op = async {
// TODO batch requests are not supported by the SDK
// TODO batch requests are also not supported by the SDK
// https://github.com/Azure/azure-sdk-for-rust/issues/1068
// https://github.com/Azure/azure-sdk-for-rust/issues/1249
for path in paths {
#[derive(Debug)]
enum AzureOrTimeout {
AzureError(azure_core::Error),
Timeout,
Cancel,
}
impl Display for AzureOrTimeout {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
write!(f, "{self:?}")
}
}
let warn_threshold = 3;
let max_retries = 5;
backoff::retry(
|| async {
let blob_client = self.client.blob_client(self.relative_path_to_name(path));
let blob_client = self.client.blob_client(self.relative_path_to_name(path));
let request = blob_client.delete().into_future();
let request = blob_client.delete().into_future();
let res = tokio::time::timeout(self.timeout, request).await;
let res = tokio::time::timeout(self.timeout, request).await;
match res {
Ok(Ok(_v)) => Ok(()),
Ok(Err(azure_err)) => {
if let Some(http_err) = azure_err.as_http_error() {
if http_err.status() == StatusCode::NotFound {
return Ok(());
}
}
Err(AzureOrTimeout::AzureError(azure_err))
match res {
Ok(Ok(_response)) => continue,
Ok(Err(e)) => {
if let Some(http_err) = e.as_http_error() {
if http_err.status() == StatusCode::NotFound {
continue;
}
Err(_elapsed) => Err(AzureOrTimeout::Timeout),
}
},
|err| match err {
AzureOrTimeout::AzureError(_) | AzureOrTimeout::Timeout => false,
AzureOrTimeout::Cancel => true,
},
warn_threshold,
max_retries,
"deleting remote object",
cancel,
)
.await
.ok_or_else(|| AzureOrTimeout::Cancel)
.and_then(|x| x)
.map_err(|e| match e {
AzureOrTimeout::AzureError(err) => anyhow::Error::from(err),
AzureOrTimeout::Timeout => TimeoutOrCancel::Timeout.into(),
AzureOrTimeout::Cancel => TimeoutOrCancel::Cancel.into(),
})?;
return Err(e.into());
}
Err(_elapsed) => return Err(TimeoutOrCancel::Timeout.into()),
}
}
Ok(())
};
let res = tokio::select! {
tokio::select! {
res = op => res,
_ = cancel.cancelled() => return Err(TimeoutOrCancel::Cancel.into()),
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, &res, started_at);
res
_ = cancel.cancelled() => Err(TimeoutOrCancel::Cancel.into()),
}
}
async fn copy(
@@ -529,9 +460,7 @@ impl RemoteStorage for AzureBlobStorage {
to: &RemotePath,
cancel: &CancellationToken,
) -> anyhow::Result<()> {
let kind = RequestKind::Copy;
let _permit = self.permit(kind, cancel).await?;
let started_at = start_measuring_requests(kind);
let _permit = self.permit(RequestKind::Copy, cancel).await?;
let timeout = tokio::time::sleep(self.timeout);
@@ -575,21 +504,15 @@ impl RemoteStorage for AzureBlobStorage {
}
};
let res = tokio::select! {
tokio::select! {
res = op => res,
_ = cancel.cancelled() => return Err(anyhow::Error::new(TimeoutOrCancel::Cancel)),
_ = cancel.cancelled() => Err(anyhow::Error::new(TimeoutOrCancel::Cancel)),
_ = timeout => {
let e = anyhow::Error::new(TimeoutOrCancel::Timeout);
let e = e.context(format!("Timeout, last status: {copy_status:?}"));
Err(e)
},
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, &res, started_at);
res
}
}
async fn time_travel_recover(
@@ -603,6 +526,10 @@ impl RemoteStorage for AzureBlobStorage {
// https://learn.microsoft.com/en-us/azure/storage/blobs/point-in-time-restore-overview
Err(TimeTravelError::Unimplemented)
}
fn activity(&self) -> RemoteStorageActivity {
self.concurrency_limiter.activity()
}
}
pin_project_lite::pin_project! {

View File

@@ -1,277 +0,0 @@
use std::{fmt::Debug, num::NonZeroUsize, str::FromStr, time::Duration};
use anyhow::bail;
use aws_sdk_s3::types::StorageClass;
use camino::Utf8PathBuf;
use serde::{Deserialize, Serialize};
use crate::{
DEFAULT_MAX_KEYS_PER_LIST_RESPONSE, DEFAULT_REMOTE_STORAGE_AZURE_CONCURRENCY_LIMIT,
DEFAULT_REMOTE_STORAGE_S3_CONCURRENCY_LIMIT,
};
/// External backup storage configuration, enough for creating a client for that storage.
#[derive(Debug, Clone, PartialEq, Eq, Deserialize, Serialize)]
pub struct RemoteStorageConfig {
/// The storage connection configuration.
#[serde(flatten)]
pub storage: RemoteStorageKind,
/// A common timeout enforced for all requests after concurrency limiter permit has been
/// acquired.
#[serde(
with = "humantime_serde",
default = "default_timeout",
skip_serializing_if = "is_default_timeout"
)]
pub timeout: Duration,
}
fn default_timeout() -> Duration {
RemoteStorageConfig::DEFAULT_TIMEOUT
}
fn is_default_timeout(d: &Duration) -> bool {
*d == RemoteStorageConfig::DEFAULT_TIMEOUT
}
/// A kind of a remote storage to connect to, with its connection configuration.
#[derive(Debug, Clone, PartialEq, Eq, Deserialize, Serialize)]
#[serde(untagged)]
pub enum RemoteStorageKind {
/// Storage based on local file system.
/// Specify a root folder to place all stored files into.
LocalFs { local_path: Utf8PathBuf },
/// AWS S3 based storage, storing all files in the S3 bucket
/// specified by the config
AwsS3(S3Config),
/// Azure Blob based storage, storing all files in the container
/// specified by the config
AzureContainer(AzureConfig),
}
/// AWS S3 bucket coordinates and access credentials to manage the bucket contents (read and write).
#[derive(Clone, PartialEq, Eq, Deserialize, Serialize)]
pub struct S3Config {
/// Name of the bucket to connect to.
pub bucket_name: String,
/// The region where the bucket is located at.
pub bucket_region: String,
/// A "subfolder" in the bucket, to use the same bucket separately by multiple remote storage users at once.
pub prefix_in_bucket: Option<String>,
/// A base URL to send S3 requests to.
/// By default, the endpoint is derived from a region name, assuming it's
/// an AWS S3 region name, erroring on wrong region name.
/// Endpoint provides a way to support other S3 flavors and their regions.
///
/// Example: `http://127.0.0.1:5000`
pub endpoint: Option<String>,
/// AWS S3 has various limits on its API calls, we need not to exceed those.
/// See [`DEFAULT_REMOTE_STORAGE_S3_CONCURRENCY_LIMIT`] for more details.
#[serde(default = "default_remote_storage_s3_concurrency_limit")]
pub concurrency_limit: NonZeroUsize,
#[serde(default = "default_max_keys_per_list_response")]
pub max_keys_per_list_response: Option<i32>,
#[serde(
deserialize_with = "deserialize_storage_class",
serialize_with = "serialize_storage_class",
default
)]
pub upload_storage_class: Option<StorageClass>,
}
fn default_remote_storage_s3_concurrency_limit() -> NonZeroUsize {
DEFAULT_REMOTE_STORAGE_S3_CONCURRENCY_LIMIT
.try_into()
.unwrap()
}
fn default_max_keys_per_list_response() -> Option<i32> {
DEFAULT_MAX_KEYS_PER_LIST_RESPONSE
}
impl Debug for S3Config {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.debug_struct("S3Config")
.field("bucket_name", &self.bucket_name)
.field("bucket_region", &self.bucket_region)
.field("prefix_in_bucket", &self.prefix_in_bucket)
.field("concurrency_limit", &self.concurrency_limit)
.field(
"max_keys_per_list_response",
&self.max_keys_per_list_response,
)
.finish()
}
}
/// Azure bucket coordinates and access credentials to manage the bucket contents (read and write).
#[derive(Clone, PartialEq, Eq, Serialize, Deserialize)]
pub struct AzureConfig {
/// Name of the container to connect to.
pub container_name: String,
/// Name of the storage account the container is inside of
pub storage_account: Option<String>,
/// The region where the bucket is located at.
pub container_region: String,
/// A "subfolder" in the container, to use the same container separately by multiple remote storage users at once.
pub prefix_in_container: Option<String>,
/// Azure has various limits on its API calls, we need not to exceed those.
/// See [`DEFAULT_REMOTE_STORAGE_AZURE_CONCURRENCY_LIMIT`] for more details.
#[serde(default = "default_remote_storage_azure_concurrency_limit")]
pub concurrency_limit: NonZeroUsize,
#[serde(default = "default_max_keys_per_list_response")]
pub max_keys_per_list_response: Option<i32>,
}
fn default_remote_storage_azure_concurrency_limit() -> NonZeroUsize {
NonZeroUsize::new(DEFAULT_REMOTE_STORAGE_AZURE_CONCURRENCY_LIMIT).unwrap()
}
impl Debug for AzureConfig {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.debug_struct("AzureConfig")
.field("bucket_name", &self.container_name)
.field("storage_account", &self.storage_account)
.field("bucket_region", &self.container_region)
.field("prefix_in_container", &self.prefix_in_container)
.field("concurrency_limit", &self.concurrency_limit)
.field(
"max_keys_per_list_response",
&self.max_keys_per_list_response,
)
.finish()
}
}
fn deserialize_storage_class<'de, D: serde::Deserializer<'de>>(
deserializer: D,
) -> Result<Option<StorageClass>, D::Error> {
Option::<String>::deserialize(deserializer).and_then(|s| {
if let Some(s) = s {
use serde::de::Error;
let storage_class = StorageClass::from_str(&s).expect("infallible");
#[allow(deprecated)]
if matches!(storage_class, StorageClass::Unknown(_)) {
return Err(D::Error::custom(format!(
"Specified storage class unknown to SDK: '{s}'. Allowed values: {:?}",
StorageClass::values()
)));
}
Ok(Some(storage_class))
} else {
Ok(None)
}
})
}
fn serialize_storage_class<S: serde::Serializer>(
val: &Option<StorageClass>,
serializer: S,
) -> Result<S::Ok, S::Error> {
let val = val.as_ref().map(StorageClass::as_str);
Option::<&str>::serialize(&val, serializer)
}
impl RemoteStorageConfig {
pub const DEFAULT_TIMEOUT: Duration = std::time::Duration::from_secs(120);
pub fn from_toml(toml: &toml_edit::Item) -> anyhow::Result<Option<RemoteStorageConfig>> {
let document: toml_edit::Document = match toml {
toml_edit::Item::Table(toml) => toml.clone().into(),
toml_edit::Item::Value(toml_edit::Value::InlineTable(toml)) => {
toml.clone().into_table().into()
}
_ => bail!("toml not a table or inline table"),
};
if document.is_empty() {
return Ok(None);
}
Ok(Some(toml_edit::de::from_document(document)?))
}
}
#[cfg(test)]
mod tests {
use super::*;
fn parse(input: &str) -> anyhow::Result<Option<RemoteStorageConfig>> {
let toml = input.parse::<toml_edit::Document>().unwrap();
RemoteStorageConfig::from_toml(toml.as_item())
}
#[test]
fn parse_localfs_config_with_timeout() {
let input = "local_path = '.'
timeout = '5s'";
let config = parse(input).unwrap().expect("it exists");
assert_eq!(
config,
RemoteStorageConfig {
storage: RemoteStorageKind::LocalFs {
local_path: Utf8PathBuf::from(".")
},
timeout: Duration::from_secs(5)
}
);
}
#[test]
fn test_s3_parsing() {
let toml = "\
bucket_name = 'foo-bar'
bucket_region = 'eu-central-1'
upload_storage_class = 'INTELLIGENT_TIERING'
timeout = '7s'
";
let config = parse(toml).unwrap().expect("it exists");
assert_eq!(
config,
RemoteStorageConfig {
storage: RemoteStorageKind::AwsS3(S3Config {
bucket_name: "foo-bar".into(),
bucket_region: "eu-central-1".into(),
prefix_in_bucket: None,
endpoint: None,
concurrency_limit: default_remote_storage_s3_concurrency_limit(),
max_keys_per_list_response: DEFAULT_MAX_KEYS_PER_LIST_RESPONSE,
upload_storage_class: Some(StorageClass::IntelligentTiering),
}),
timeout: Duration::from_secs(7)
}
);
}
#[test]
fn test_azure_parsing() {
let toml = "\
container_name = 'foo-bar'
container_region = 'westeurope'
upload_storage_class = 'INTELLIGENT_TIERING'
timeout = '7s'
";
let config = parse(toml).unwrap().expect("it exists");
assert_eq!(
config,
RemoteStorageConfig {
storage: RemoteStorageKind::AzureContainer(AzureConfig {
container_name: "foo-bar".into(),
storage_account: None,
container_region: "westeurope".into(),
prefix_in_container: None,
concurrency_limit: default_remote_storage_azure_concurrency_limit(),
max_keys_per_list_response: DEFAULT_MAX_KEYS_PER_LIST_RESPONSE,
}),
timeout: Duration::from_secs(7)
}
);
}
}

View File

@@ -10,19 +10,24 @@
#![deny(clippy::undocumented_unsafe_blocks)]
mod azure_blob;
mod config;
mod error;
mod local_fs;
mod metrics;
mod s3_bucket;
mod simulate_failures;
mod support;
use std::{
collections::HashMap, fmt::Debug, num::NonZeroU32, pin::Pin, sync::Arc, time::SystemTime,
collections::HashMap,
fmt::Debug,
num::{NonZeroU32, NonZeroUsize},
pin::Pin,
str::FromStr,
sync::Arc,
time::{Duration, SystemTime},
};
use anyhow::Context;
use anyhow::{bail, Context};
use aws_sdk_s3::types::StorageClass;
use camino::{Utf8Path, Utf8PathBuf};
use bytes::Bytes;
@@ -30,6 +35,7 @@ use futures::stream::Stream;
use serde::{Deserialize, Serialize};
use tokio::sync::Semaphore;
use tokio_util::sync::CancellationToken;
use toml_edit::Item;
use tracing::info;
pub use self::{
@@ -38,8 +44,6 @@ pub use self::{
};
use s3_bucket::RequestKind;
pub use crate::config::{AzureConfig, RemoteStorageConfig, RemoteStorageKind, S3Config};
/// Azure SDK's ETag type is a simple String wrapper: we use this internally instead of repeating it here.
pub use azure_core::Etag;
@@ -117,8 +121,8 @@ impl RemotePath {
self.0.file_name()
}
pub fn join(&self, path: impl AsRef<Utf8Path>) -> Self {
Self(self.0.join(path))
pub fn join(&self, segment: &Utf8Path) -> Self {
Self(self.0.join(segment))
}
pub fn get_path(&self) -> &Utf8PathBuf {
@@ -259,6 +263,17 @@ pub trait RemoteStorage: Send + Sync + 'static {
done_if_after: SystemTime,
cancel: &CancellationToken,
) -> Result<(), TimeTravelError>;
/// Query how busy we currently are: may be used by callers which wish to politely
/// back off if there are already a lot of operations underway.
fn activity(&self) -> RemoteStorageActivity;
}
pub struct RemoteStorageActivity {
pub read_available: usize,
pub read_total: usize,
pub write_available: usize,
pub write_total: usize,
}
/// DownloadStream is sensitive to the timeout and cancellation used with the original
@@ -440,13 +455,22 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
}
}
}
pub fn activity(&self) -> RemoteStorageActivity {
match self {
Self::LocalFs(s) => s.activity(),
Self::AwsS3(s) => s.activity(),
Self::AzureBlob(s) => s.activity(),
Self::Unreliable(s) => s.activity(),
}
}
}
impl GenericRemoteStorage {
pub fn from_config(storage_config: &RemoteStorageConfig) -> anyhow::Result<Self> {
let timeout = storage_config.timeout;
Ok(match &storage_config.storage {
RemoteStorageKind::LocalFs { local_path: path } => {
RemoteStorageKind::LocalFs(path) => {
info!("Using fs root '{path}' as a remote storage");
Self::LocalFs(LocalFs::new(path.clone(), timeout)?)
}
@@ -461,11 +485,7 @@ impl GenericRemoteStorage {
Self::AwsS3(Arc::new(S3Bucket::new(s3_config, timeout)?))
}
RemoteStorageKind::AzureContainer(azure_config) => {
let storage_account = azure_config
.storage_account
.as_deref()
.unwrap_or("<AZURE_STORAGE_ACCOUNT>");
info!("Using azure container '{}' in account '{storage_account}' in region '{}' as a remote storage, prefix in container: '{:?}'",
info!("Using azure container '{}' in region '{}' as a remote storage, prefix in container: '{:?}'",
azure_config.container_name, azure_config.container_region, azure_config.prefix_in_container);
Self::AzureBlob(Arc::new(AzureBlobStorage::new(azure_config, timeout)?))
}
@@ -521,12 +541,262 @@ impl<const N: usize> From<[(&str, &str); N]> for StorageMetadata {
}
}
/// External backup storage configuration, enough for creating a client for that storage.
#[derive(Debug, Clone, PartialEq, Eq)]
pub struct RemoteStorageConfig {
/// The storage connection configuration.
pub storage: RemoteStorageKind,
/// A common timeout enforced for all requests after concurrency limiter permit has been
/// acquired.
pub timeout: Duration,
}
/// A kind of a remote storage to connect to, with its connection configuration.
#[derive(Debug, Clone, PartialEq, Eq)]
pub enum RemoteStorageKind {
/// Storage based on local file system.
/// Specify a root folder to place all stored files into.
LocalFs(Utf8PathBuf),
/// AWS S3 based storage, storing all files in the S3 bucket
/// specified by the config
AwsS3(S3Config),
/// Azure Blob based storage, storing all files in the container
/// specified by the config
AzureContainer(AzureConfig),
}
/// AWS S3 bucket coordinates and access credentials to manage the bucket contents (read and write).
#[derive(Clone, PartialEq, Eq)]
pub struct S3Config {
/// Name of the bucket to connect to.
pub bucket_name: String,
/// The region where the bucket is located at.
pub bucket_region: String,
/// A "subfolder" in the bucket, to use the same bucket separately by multiple remote storage users at once.
pub prefix_in_bucket: Option<String>,
/// A base URL to send S3 requests to.
/// By default, the endpoint is derived from a region name, assuming it's
/// an AWS S3 region name, erroring on wrong region name.
/// Endpoint provides a way to support other S3 flavors and their regions.
///
/// Example: `http://127.0.0.1:5000`
pub endpoint: Option<String>,
/// AWS S3 has various limits on its API calls, we need not to exceed those.
/// See [`DEFAULT_REMOTE_STORAGE_S3_CONCURRENCY_LIMIT`] for more details.
pub concurrency_limit: NonZeroUsize,
pub max_keys_per_list_response: Option<i32>,
pub upload_storage_class: Option<StorageClass>,
}
impl Debug for S3Config {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.debug_struct("S3Config")
.field("bucket_name", &self.bucket_name)
.field("bucket_region", &self.bucket_region)
.field("prefix_in_bucket", &self.prefix_in_bucket)
.field("concurrency_limit", &self.concurrency_limit)
.field(
"max_keys_per_list_response",
&self.max_keys_per_list_response,
)
.finish()
}
}
/// Azure bucket coordinates and access credentials to manage the bucket contents (read and write).
#[derive(Clone, PartialEq, Eq)]
pub struct AzureConfig {
/// Name of the container to connect to.
pub container_name: String,
/// The region where the bucket is located at.
pub container_region: String,
/// A "subfolder" in the container, to use the same container separately by multiple remote storage users at once.
pub prefix_in_container: Option<String>,
/// Azure has various limits on its API calls, we need not to exceed those.
/// See [`DEFAULT_REMOTE_STORAGE_AZURE_CONCURRENCY_LIMIT`] for more details.
pub concurrency_limit: NonZeroUsize,
pub max_keys_per_list_response: Option<i32>,
}
impl Debug for AzureConfig {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.debug_struct("AzureConfig")
.field("bucket_name", &self.container_name)
.field("bucket_region", &self.container_region)
.field("prefix_in_bucket", &self.prefix_in_container)
.field("concurrency_limit", &self.concurrency_limit)
.field(
"max_keys_per_list_response",
&self.max_keys_per_list_response,
)
.finish()
}
}
impl RemoteStorageConfig {
pub const DEFAULT_TIMEOUT: Duration = std::time::Duration::from_secs(120);
pub fn from_toml(toml: &toml_edit::Item) -> anyhow::Result<Option<RemoteStorageConfig>> {
let local_path = toml.get("local_path");
let bucket_name = toml.get("bucket_name");
let bucket_region = toml.get("bucket_region");
let container_name = toml.get("container_name");
let container_region = toml.get("container_region");
let use_azure = container_name.is_some() && container_region.is_some();
let default_concurrency_limit = if use_azure {
DEFAULT_REMOTE_STORAGE_AZURE_CONCURRENCY_LIMIT
} else {
DEFAULT_REMOTE_STORAGE_S3_CONCURRENCY_LIMIT
};
let concurrency_limit = NonZeroUsize::new(
parse_optional_integer("concurrency_limit", toml)?.unwrap_or(default_concurrency_limit),
)
.context("Failed to parse 'concurrency_limit' as a positive integer")?;
let max_keys_per_list_response =
parse_optional_integer::<i32, _>("max_keys_per_list_response", toml)
.context("Failed to parse 'max_keys_per_list_response' as a positive integer")?
.or(DEFAULT_MAX_KEYS_PER_LIST_RESPONSE);
let endpoint = toml
.get("endpoint")
.map(|endpoint| parse_toml_string("endpoint", endpoint))
.transpose()?;
let timeout = toml
.get("timeout")
.map(|timeout| {
timeout
.as_str()
.ok_or_else(|| anyhow::Error::msg("timeout was not a string"))
})
.transpose()
.and_then(|timeout| {
timeout
.map(humantime::parse_duration)
.transpose()
.map_err(anyhow::Error::new)
})
.context("parse timeout")?
.unwrap_or(Self::DEFAULT_TIMEOUT);
if timeout < Duration::from_secs(1) {
bail!("timeout was specified as {timeout:?} which is too low");
}
let storage = match (
local_path,
bucket_name,
bucket_region,
container_name,
container_region,
) {
// no 'local_path' nor 'bucket_name' options are provided, consider this remote storage disabled
(None, None, None, None, None) => return Ok(None),
(_, Some(_), None, ..) => {
bail!("'bucket_region' option is mandatory if 'bucket_name' is given ")
}
(_, None, Some(_), ..) => {
bail!("'bucket_name' option is mandatory if 'bucket_region' is given ")
}
(None, Some(bucket_name), Some(bucket_region), ..) => {
RemoteStorageKind::AwsS3(S3Config {
bucket_name: parse_toml_string("bucket_name", bucket_name)?,
bucket_region: parse_toml_string("bucket_region", bucket_region)?,
prefix_in_bucket: toml
.get("prefix_in_bucket")
.map(|prefix_in_bucket| {
parse_toml_string("prefix_in_bucket", prefix_in_bucket)
})
.transpose()?,
endpoint,
concurrency_limit,
max_keys_per_list_response,
upload_storage_class: toml
.get("upload_storage_class")
.map(|prefix_in_bucket| -> anyhow::Result<_> {
let s = parse_toml_string("upload_storage_class", prefix_in_bucket)?;
let storage_class = StorageClass::from_str(&s).expect("infallible");
#[allow(deprecated)]
if matches!(storage_class, StorageClass::Unknown(_)) {
bail!("Specified storage class unknown to SDK: '{s}'. Allowed values: {:?}", StorageClass::values());
}
Ok(storage_class)
})
.transpose()?,
})
}
(_, _, _, Some(_), None) => {
bail!("'container_name' option is mandatory if 'container_region' is given ")
}
(_, _, _, None, Some(_)) => {
bail!("'container_name' option is mandatory if 'container_region' is given ")
}
(None, None, None, Some(container_name), Some(container_region)) => {
RemoteStorageKind::AzureContainer(AzureConfig {
container_name: parse_toml_string("container_name", container_name)?,
container_region: parse_toml_string("container_region", container_region)?,
prefix_in_container: toml
.get("prefix_in_container")
.map(|prefix_in_container| {
parse_toml_string("prefix_in_container", prefix_in_container)
})
.transpose()?,
concurrency_limit,
max_keys_per_list_response,
})
}
(Some(local_path), None, None, None, None) => RemoteStorageKind::LocalFs(
Utf8PathBuf::from(parse_toml_string("local_path", local_path)?),
),
(Some(_), Some(_), ..) => {
bail!("'local_path' and 'bucket_name' are mutually exclusive")
}
(Some(_), _, _, Some(_), Some(_)) => {
bail!("local_path and 'container_name' are mutually exclusive")
}
};
Ok(Some(RemoteStorageConfig { storage, timeout }))
}
}
// Helper functions to parse a toml Item
fn parse_optional_integer<I, E>(name: &str, item: &toml_edit::Item) -> anyhow::Result<Option<I>>
where
I: TryFrom<i64, Error = E>,
E: std::error::Error + Send + Sync + 'static,
{
let toml_integer = match item.get(name) {
Some(item) => item
.as_integer()
.with_context(|| format!("configure option {name} is not an integer"))?,
None => return Ok(None),
};
I::try_from(toml_integer)
.map(Some)
.with_context(|| format!("configure option {name} is too large"))
}
fn parse_toml_string(name: &str, item: &Item) -> anyhow::Result<String> {
let s = item
.as_str()
.with_context(|| format!("configure option {name} is not a string"))?;
Ok(s.to_string())
}
struct ConcurrencyLimiter {
// Every request to S3 can be throttled or cancelled, if a certain number of requests per second is exceeded.
// Same goes to IAM, which is queried before every S3 request, if enabled. IAM has even lower RPS threshold.
// The helps to ensure we don't exceed the thresholds.
write: Arc<Semaphore>,
read: Arc<Semaphore>,
write_total: usize,
read_total: usize,
}
impl ConcurrencyLimiter {
@@ -555,10 +825,21 @@ impl ConcurrencyLimiter {
Arc::clone(self.for_kind(kind)).acquire_owned().await
}
fn activity(&self) -> RemoteStorageActivity {
RemoteStorageActivity {
read_available: self.read.available_permits(),
read_total: self.read_total,
write_available: self.write.available_permits(),
write_total: self.write_total,
}
}
fn new(limit: usize) -> ConcurrencyLimiter {
Self {
read: Arc::new(Semaphore::new(limit)),
write: Arc::new(Semaphore::new(limit)),
read_total: limit,
write_total: limit,
}
}
}
@@ -588,4 +869,24 @@ mod tests {
let err = RemotePath::new(Utf8Path::new("/")).expect_err("Should fail on absolute paths");
assert_eq!(err.to_string(), "Path \"/\" is not relative");
}
#[test]
fn parse_localfs_config_with_timeout() {
let input = "local_path = '.'
timeout = '5s'";
let toml = input.parse::<toml_edit::Document>().unwrap();
let config = RemoteStorageConfig::from_toml(toml.as_item())
.unwrap()
.expect("it exists");
assert_eq!(
config,
RemoteStorageConfig {
storage: RemoteStorageKind::LocalFs(Utf8PathBuf::from(".")),
timeout: Duration::from_secs(5)
}
);
}
}

View File

@@ -23,8 +23,8 @@ use tokio_util::{io::ReaderStream, sync::CancellationToken};
use utils::crashsafe::path_with_suffix_extension;
use crate::{
Download, DownloadError, Listing, ListingMode, RemotePath, TimeTravelError, TimeoutOrCancel,
REMOTE_STORAGE_PREFIX_SEPARATOR,
Download, DownloadError, Listing, ListingMode, RemotePath, RemoteStorageActivity,
TimeTravelError, TimeoutOrCancel, REMOTE_STORAGE_PREFIX_SEPARATOR,
};
use super::{RemoteStorage, StorageMetadata};
@@ -605,6 +605,16 @@ impl RemoteStorage for LocalFs {
) -> Result<(), TimeTravelError> {
Err(TimeTravelError::Unimplemented)
}
fn activity(&self) -> RemoteStorageActivity {
// LocalFS has no concurrency limiting: give callers the impression that plenty of units are available
RemoteStorageActivity {
read_available: 16,
read_total: 16,
write_available: 16,
write_total: 16,
}
}
}
fn storage_metadata_path(original_path: &Utf8Path) -> Utf8PathBuf {

View File

@@ -46,16 +46,15 @@ use utils::backoff;
use super::StorageMetadata;
use crate::{
config::S3Config,
error::Cancelled,
metrics::{start_counting_cancelled_wait, start_measuring_requests},
support::PermitCarrying,
ConcurrencyLimiter, Download, DownloadError, Listing, ListingMode, RemotePath, RemoteStorage,
error::Cancelled, support::PermitCarrying, ConcurrencyLimiter, Download, DownloadError,
Listing, ListingMode, RemotePath, RemoteStorage, RemoteStorageActivity, S3Config,
TimeTravelError, TimeoutOrCancel, MAX_KEYS_PER_DELETE, REMOTE_STORAGE_PREFIX_SEPARATOR,
};
use crate::metrics::AttemptOutcome;
pub(super) use crate::metrics::RequestKind;
pub(super) mod metrics;
use self::metrics::AttemptOutcome;
pub(super) use self::metrics::RequestKind;
/// AWS S3 storage.
pub struct S3Bucket {
@@ -228,7 +227,7 @@ impl S3Bucket {
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.wait_seconds
.observe_elapsed(kind, started_at);
@@ -249,7 +248,7 @@ impl S3Bucket {
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.wait_seconds
.observe_elapsed(kind, started_at);
Ok(permit)
@@ -288,7 +287,7 @@ impl S3Bucket {
// Count this in the AttemptOutcome::Ok bucket, because 404 is not
// an error: we expect to sometimes fetch an object and find it missing,
// e.g. when probing for timeline indices.
crate::metrics::BUCKET_METRICS.req_seconds.observe_elapsed(
metrics::BUCKET_METRICS.req_seconds.observe_elapsed(
kind,
AttemptOutcome::Ok,
started_at,
@@ -296,7 +295,7 @@ impl S3Bucket {
return Err(DownloadError::NotFound);
}
Err(e) => {
crate::metrics::BUCKET_METRICS.req_seconds.observe_elapsed(
metrics::BUCKET_METRICS.req_seconds.observe_elapsed(
kind,
AttemptOutcome::Err,
started_at,
@@ -372,12 +371,12 @@ impl S3Bucket {
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, &resp, started_at);
let resp = resp.context("request deletion")?;
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.deleted_objects_total
.inc_by(chunk.len() as u64);
@@ -436,14 +435,14 @@ pin_project_lite::pin_project! {
/// Times and tracks the outcome of the request.
struct TimedDownload<S> {
started_at: std::time::Instant,
outcome: AttemptOutcome,
outcome: metrics::AttemptOutcome,
#[pin]
inner: S
}
impl<S> PinnedDrop for TimedDownload<S> {
fn drop(mut this: Pin<&mut Self>) {
crate::metrics::BUCKET_METRICS.req_seconds.observe_elapsed(RequestKind::Get, this.outcome, this.started_at);
metrics::BUCKET_METRICS.req_seconds.observe_elapsed(RequestKind::Get, this.outcome, this.started_at);
}
}
}
@@ -452,7 +451,7 @@ impl<S> TimedDownload<S> {
fn new(started_at: std::time::Instant, inner: S) -> Self {
TimedDownload {
started_at,
outcome: AttemptOutcome::Cancelled,
outcome: metrics::AttemptOutcome::Cancelled,
inner,
}
}
@@ -469,8 +468,8 @@ impl<S: Stream<Item = std::io::Result<Bytes>>> Stream for TimedDownload<S> {
let res = ready!(this.inner.poll_next(cx));
match &res {
Some(Ok(_)) => {}
Some(Err(_)) => *this.outcome = AttemptOutcome::Err,
None => *this.outcome = AttemptOutcome::Ok,
Some(Err(_)) => *this.outcome = metrics::AttemptOutcome::Err,
None => *this.outcome = metrics::AttemptOutcome::Ok,
}
Poll::Ready(res)
@@ -544,7 +543,7 @@ impl RemoteStorage for S3Bucket {
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, &response, started_at);
@@ -626,7 +625,7 @@ impl RemoteStorage for S3Bucket {
if let Ok(inner) = &res {
// do not incl. timeouts as errors in metrics but cancellations
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, inner, started_at);
}
@@ -674,7 +673,7 @@ impl RemoteStorage for S3Bucket {
};
let started_at = ScopeGuard::into_inner(started_at);
crate::metrics::BUCKET_METRICS
metrics::BUCKET_METRICS
.req_seconds
.observe_elapsed(kind, &res, started_at);
@@ -976,6 +975,32 @@ impl RemoteStorage for S3Bucket {
}
Ok(())
}
fn activity(&self) -> RemoteStorageActivity {
self.concurrency_limiter.activity()
}
}
/// On drop (cancellation) count towards [`metrics::BucketMetrics::cancelled_waits`].
fn start_counting_cancelled_wait(
kind: RequestKind,
) -> ScopeGuard<std::time::Instant, impl FnOnce(std::time::Instant), scopeguard::OnSuccess> {
scopeguard::guard_on_success(std::time::Instant::now(), move |_| {
metrics::BUCKET_METRICS.cancelled_waits.get(kind).inc()
})
}
/// On drop (cancellation) add time to [`metrics::BucketMetrics::req_seconds`].
fn start_measuring_requests(
kind: RequestKind,
) -> ScopeGuard<std::time::Instant, impl FnOnce(std::time::Instant), scopeguard::OnSuccess> {
scopeguard::guard_on_success(std::time::Instant::now(), move |started_at| {
metrics::BUCKET_METRICS.req_seconds.observe_elapsed(
kind,
AttemptOutcome::Cancelled,
started_at,
)
})
}
// Save RAM and only store the needed data instead of the entire ObjectVersion/DeleteMarkerEntry

View File

@@ -15,7 +15,6 @@ pub(crate) enum RequestKind {
TimeTravel = 5,
}
use scopeguard::ScopeGuard;
use RequestKind::*;
impl RequestKind {
@@ -34,10 +33,10 @@ impl RequestKind {
}
}
pub(crate) struct RequestTyped<C>([C; 6]);
pub(super) struct RequestTyped<C>([C; 6]);
impl<C> RequestTyped<C> {
pub(crate) fn get(&self, kind: RequestKind) -> &C {
pub(super) fn get(&self, kind: RequestKind) -> &C {
&self.0[kind.as_index()]
}
@@ -59,19 +58,19 @@ impl<C> RequestTyped<C> {
}
impl RequestTyped<Histogram> {
pub(crate) fn observe_elapsed(&self, kind: RequestKind, started_at: std::time::Instant) {
pub(super) fn observe_elapsed(&self, kind: RequestKind, started_at: std::time::Instant) {
self.get(kind).observe(started_at.elapsed().as_secs_f64())
}
}
pub(crate) struct PassFailCancelledRequestTyped<C> {
pub(super) struct PassFailCancelledRequestTyped<C> {
success: RequestTyped<C>,
fail: RequestTyped<C>,
cancelled: RequestTyped<C>,
}
#[derive(Debug, Clone, Copy)]
pub(crate) enum AttemptOutcome {
pub(super) enum AttemptOutcome {
Ok,
Err,
Cancelled,
@@ -87,7 +86,7 @@ impl<T, E> From<&Result<T, E>> for AttemptOutcome {
}
impl AttemptOutcome {
pub(crate) fn as_str(&self) -> &'static str {
pub(super) fn as_str(&self) -> &'static str {
match self {
AttemptOutcome::Ok => "ok",
AttemptOutcome::Err => "err",
@@ -97,7 +96,7 @@ impl AttemptOutcome {
}
impl<C> PassFailCancelledRequestTyped<C> {
pub(crate) fn get(&self, kind: RequestKind, outcome: AttemptOutcome) -> &C {
pub(super) fn get(&self, kind: RequestKind, outcome: AttemptOutcome) -> &C {
let target = match outcome {
AttemptOutcome::Ok => &self.success,
AttemptOutcome::Err => &self.fail,
@@ -120,7 +119,7 @@ impl<C> PassFailCancelledRequestTyped<C> {
}
impl PassFailCancelledRequestTyped<Histogram> {
pub(crate) fn observe_elapsed(
pub(super) fn observe_elapsed(
&self,
kind: RequestKind,
outcome: impl Into<AttemptOutcome>,
@@ -131,44 +130,19 @@ impl PassFailCancelledRequestTyped<Histogram> {
}
}
/// On drop (cancellation) count towards [`BucketMetrics::cancelled_waits`].
pub(crate) fn start_counting_cancelled_wait(
kind: RequestKind,
) -> ScopeGuard<std::time::Instant, impl FnOnce(std::time::Instant), scopeguard::OnSuccess> {
scopeguard::guard_on_success(std::time::Instant::now(), move |_| {
crate::metrics::BUCKET_METRICS
.cancelled_waits
.get(kind)
.inc()
})
}
/// On drop (cancellation) add time to [`BucketMetrics::req_seconds`].
pub(crate) fn start_measuring_requests(
kind: RequestKind,
) -> ScopeGuard<std::time::Instant, impl FnOnce(std::time::Instant), scopeguard::OnSuccess> {
scopeguard::guard_on_success(std::time::Instant::now(), move |started_at| {
crate::metrics::BUCKET_METRICS.req_seconds.observe_elapsed(
kind,
AttemptOutcome::Cancelled,
started_at,
)
})
}
pub(crate) struct BucketMetrics {
pub(super) struct BucketMetrics {
/// Full request duration until successful completion, error or cancellation.
pub(crate) req_seconds: PassFailCancelledRequestTyped<Histogram>,
pub(super) req_seconds: PassFailCancelledRequestTyped<Histogram>,
/// Total amount of seconds waited on queue.
pub(crate) wait_seconds: RequestTyped<Histogram>,
pub(super) wait_seconds: RequestTyped<Histogram>,
/// Track how many semaphore awaits were cancelled per request type.
///
/// This is in case cancellations are happening more than expected.
pub(crate) cancelled_waits: RequestTyped<IntCounter>,
pub(super) cancelled_waits: RequestTyped<IntCounter>,
/// Total amount of deleted objects in batches or single requests.
pub(crate) deleted_objects_total: IntCounter,
pub(super) deleted_objects_total: IntCounter,
}
impl Default for BucketMetrics {

View File

@@ -12,7 +12,7 @@ use tokio_util::sync::CancellationToken;
use crate::{
Download, DownloadError, GenericRemoteStorage, Listing, ListingMode, RemotePath, RemoteStorage,
StorageMetadata, TimeTravelError,
RemoteStorageActivity, StorageMetadata, TimeTravelError,
};
pub struct UnreliableWrapper {
@@ -213,4 +213,8 @@ impl RemoteStorage for UnreliableWrapper {
.time_travel_recover(prefix, timestamp, done_if_after, cancel)
.await
}
fn activity(&self) -> RemoteStorageActivity {
self.inner.activity()
}
}

View File

@@ -78,10 +78,6 @@ where
let e = Err(std::io::Error::from(e));
return Poll::Ready(Some(e));
}
} else {
// this would be perfectly valid behaviour for doing a graceful completion on the
// download for example, but not one we expect to do right now.
tracing::warn!("continuing polling after having cancelled or timeouted");
}
this.inner.poll_next(cx)
@@ -93,22 +89,13 @@ where
}
/// Fires only on the first cancel or timeout, not on both.
pub(crate) fn cancel_or_timeout(
pub(crate) async fn cancel_or_timeout(
timeout: Duration,
cancel: CancellationToken,
) -> impl std::future::Future<Output = TimeoutOrCancel> + 'static {
// futures are lazy, they don't do anything before being polled.
//
// "precalculate" the wanted deadline before returning the future, so that we can use pause
// failpoint to trigger a timeout in test.
let deadline = tokio::time::Instant::now() + timeout;
async move {
tokio::select! {
_ = tokio::time::sleep_until(deadline) => TimeoutOrCancel::Timeout,
_ = cancel.cancelled() => {
TimeoutOrCancel::Cancel
},
}
) -> TimeoutOrCancel {
tokio::select! {
_ = tokio::time::sleep(timeout) => TimeoutOrCancel::Timeout,
_ = cancel.cancelled() => TimeoutOrCancel::Cancel,
}
}
@@ -185,31 +172,4 @@ mod tests {
_ = tokio::time::sleep(Duration::from_secs(121)) => {},
}
}
#[tokio::test]
async fn notified_but_pollable_after() {
let inner = futures::stream::once(futures::future::ready(Ok(bytes::Bytes::from_static(
b"hello world",
))));
let timeout = Duration::from_secs(120);
let cancel = CancellationToken::new();
cancel.cancel();
let stream = DownloadStream::new(cancel_or_timeout(timeout, cancel.clone()), inner);
let mut stream = std::pin::pin!(stream);
let next = stream.next().await;
let ioe = next.unwrap().unwrap_err();
assert!(
matches!(
ioe.get_ref().unwrap().downcast_ref::<DownloadError>(),
Some(&DownloadError::Cancelled)
),
"{ioe:?}"
);
let next = stream.next().await;
let bytes = next.unwrap().unwrap();
assert_eq!(&b"hello world"[..], bytes);
}
}

View File

@@ -212,7 +212,6 @@ fn create_azure_client(
let remote_storage_config = RemoteStorageConfig {
storage: RemoteStorageKind::AzureContainer(AzureConfig {
container_name: remote_storage_azure_container,
storage_account: None,
container_region: remote_storage_azure_region,
prefix_in_container: Some(format!("test_{millis}_{random:08x}/")),
concurrency_limit: NonZeroUsize::new(100).unwrap(),

View File

@@ -7,7 +7,7 @@ license.workspace = true
[dependencies]
hyper.workspace = true
opentelemetry = { workspace = true, features=["rt-tokio"] }
opentelemetry-otlp = { workspace = true, default-features=false, features = ["http-proto", "trace", "http", "reqwest-client"] }
opentelemetry-otlp = { workspace = true, default_features=false, features = ["http-proto", "trace", "http", "reqwest-client"] }
opentelemetry-semantic-conventions.workspace = true
reqwest = { workspace = true, default-features = false, features = ["rustls-tls"] }
tokio = { workspace = true, features = ["rt", "rt-multi-thread"] }

View File

@@ -9,33 +9,6 @@ use serde::{Deserialize, Serialize};
use tokio_util::sync::CancellationToken;
use tracing::*;
/// Declare a failpoint that can use the `pause` failpoint action.
/// We don't want to block the executor thread, hence, spawn_blocking + await.
#[macro_export]
macro_rules! pausable_failpoint {
($name:literal) => {
if cfg!(feature = "testing") {
tokio::task::spawn_blocking({
let current = tracing::Span::current();
move || {
let _entered = current.entered();
tracing::info!("at failpoint {}", $name);
fail::fail_point!($name);
}
})
.await
.expect("spawn_blocking");
}
};
($name:literal, $cond:expr) => {
if cfg!(feature = "testing") {
if $cond {
pausable_failpoint!($name)
}
}
};
}
/// use with fail::cfg("$name", "return(2000)")
///
/// The effect is similar to a "sleep(2000)" action, i.e. we sleep for the

View File

@@ -3,9 +3,6 @@ use std::{fs, io, path::Path};
use anyhow::Context;
mod rename_noreplace;
pub use rename_noreplace::rename_noreplace;
pub trait PathExt {
/// Returns an error if `self` is not a directory.
fn is_empty_dir(&self) -> io::Result<bool>;

View File

@@ -1,109 +0,0 @@
use nix::NixPath;
/// Rename a file without replacing an existing file.
///
/// This is a wrapper around platform-specific APIs.
pub fn rename_noreplace<P1: ?Sized + NixPath, P2: ?Sized + NixPath>(
src: &P1,
dst: &P2,
) -> nix::Result<()> {
{
#[cfg(target_os = "linux")]
{
nix::fcntl::renameat2(
None,
src,
None,
dst,
nix::fcntl::RenameFlags::RENAME_NOREPLACE,
)
}
#[cfg(target_os = "macos")]
{
let res = src.with_nix_path(|src| {
dst.with_nix_path(|dst|
// SAFETY: `src` and `dst` are valid C strings as per the NixPath trait and they outlive the call to renamex_np.
unsafe {
nix::libc::renamex_np(src.as_ptr(), dst.as_ptr(), nix::libc::RENAME_EXCL)
})
})??;
nix::errno::Errno::result(res).map(drop)
}
#[cfg(not(any(target_os = "linux", target_os = "macos")))]
{
std::compile_error!("OS does not support no-replace renames");
}
}
}
#[cfg(test)]
mod test {
use std::{fs, path::PathBuf};
use super::*;
fn testdir() -> camino_tempfile::Utf8TempDir {
match crate::env::var("NEON_UTILS_RENAME_NOREPLACE_TESTDIR") {
Some(path) => {
let path: camino::Utf8PathBuf = path;
camino_tempfile::tempdir_in(path).unwrap()
}
None => camino_tempfile::tempdir().unwrap(),
}
}
#[test]
fn test_absolute_paths() {
let testdir = testdir();
println!("testdir: {}", testdir.path());
let src = testdir.path().join("src");
let dst = testdir.path().join("dst");
fs::write(&src, b"").unwrap();
fs::write(&dst, b"").unwrap();
let src = src.canonicalize().unwrap();
assert!(src.is_absolute());
let dst = dst.canonicalize().unwrap();
assert!(dst.is_absolute());
let result = rename_noreplace(&src, &dst);
assert_eq!(result.unwrap_err(), nix::Error::EEXIST);
}
#[test]
fn test_relative_paths() {
let testdir = testdir();
println!("testdir: {}", testdir.path());
// this is fine because we run in nextest => process per test
std::env::set_current_dir(testdir.path()).unwrap();
let src = PathBuf::from("src");
let dst = PathBuf::from("dst");
fs::write(&src, b"").unwrap();
fs::write(&dst, b"").unwrap();
let result = rename_noreplace(&src, &dst);
assert_eq!(result.unwrap_err(), nix::Error::EEXIST);
}
#[test]
fn test_works_when_not_exists() {
let testdir = testdir();
println!("testdir: {}", testdir.path());
let src = testdir.path().join("src");
let dst = testdir.path().join("dst");
fs::write(&src, b"content").unwrap();
rename_noreplace(src.as_std_path(), dst.as_std_path()).unwrap();
assert_eq!(
"content",
String::from_utf8(std::fs::read(&dst).unwrap()).unwrap()
);
}
}

View File

@@ -9,11 +9,20 @@ use serde::{Deserialize, Serialize};
/// numbers are used.
#[derive(Copy, Clone, Eq, PartialEq, PartialOrd, Ord, Hash)]
pub enum Generation {
// The None Generation is used in the metadata of layers written before generations were
// introduced. A running Tenant always has a valid generation, but the layer metadata may
// include None generations.
// Generations with this magic value will not add a suffix to S3 keys, and will not
// be included in persisted index_part.json. This value is only to be used
// during migration from pre-generation metadata to generation-aware metadata,
// and should eventually go away.
//
// A special Generation is used rather than always wrapping Generation in an Option,
// so that code handling generations doesn't have to be aware of the legacy
// case everywhere it touches a generation.
None,
// Generations with this magic value may never be used to construct S3 keys:
// we will panic if someone tries to. This is for Tenants in the "Broken" state,
// so that we can satisfy their constructor with a Generation without risking
// a code bug using it in an S3 write (broken tenants should never write)
Broken,
Valid(u32),
}
@@ -33,6 +42,11 @@ impl Generation {
Self::None
}
// Create a new generation that will panic if you try to use get_suffix
pub fn broken() -> Self {
Self::Broken
}
pub const fn new(v: u32) -> Self {
Self::Valid(v)
}
@@ -46,6 +60,9 @@ impl Generation {
match self {
Self::Valid(v) => GenerationFileSuffix(Some(*v)),
Self::None => GenerationFileSuffix(None),
Self::Broken => {
panic!("Tried to use a broken generation");
}
}
}
@@ -69,6 +86,7 @@ impl Generation {
}
}
Self::None => Self::None,
Self::Broken => panic!("Attempted to use a broken generation"),
}
}
@@ -77,6 +95,7 @@ impl Generation {
match self {
Self::Valid(n) => Self::Valid(*n + 1),
Self::None => Self::Valid(1),
Self::Broken => panic!("Attempted to use a broken generation"),
}
}
@@ -109,7 +128,7 @@ impl Serialize for Generation {
if let Self::Valid(v) = self {
v.serialize(serializer)
} else {
// We should never be asked to serialize a None. Structures
// We should never be asked to serialize a None or Broken. Structures
// that include an optional generation should convert None to an
// Option<Generation>::None
Err(serde::ser::Error::custom(
@@ -140,6 +159,9 @@ impl Debug for Generation {
Self::None => {
write!(f, "<none>")
}
Self::Broken => {
write!(f, "<broken>")
}
}
}
}

View File

@@ -19,13 +19,13 @@
/// // right: [0x68; 1]
/// # fn serialize_something() -> Vec<u8> { "hello world".as_bytes().to_vec() }
/// ```
pub struct Hex<S>(pub S);
#[derive(PartialEq)]
pub struct Hex<'a>(pub &'a [u8]);
impl<S: AsRef<[u8]>> std::fmt::Debug for Hex<S> {
impl std::fmt::Debug for Hex<'_> {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
write!(f, "[")?;
let chunks = self.0.as_ref().chunks(16);
for (i, c) in chunks.enumerate() {
for (i, c) in self.0.chunks(16).enumerate() {
if i > 0 && !c.is_empty() {
writeln!(f, ", ")?;
}
@@ -36,15 +36,6 @@ impl<S: AsRef<[u8]>> std::fmt::Debug for Hex<S> {
write!(f, "0x{b:02x}")?;
}
}
write!(f, "; {}]", self.0.as_ref().len())
}
}
impl<R: AsRef<[u8]>, L: AsRef<[u8]>> PartialEq<Hex<R>> for Hex<L> {
fn eq(&self, other: &Hex<R>) -> bool {
let left = self.0.as_ref();
let right = other.0.as_ref();
left == right
write!(f, "; {}]", self.0.len())
}
}

View File

@@ -34,9 +34,6 @@ pub enum ApiError {
#[error("Timeout")]
Timeout(Cow<'static, str>),
#[error("Request cancelled")]
Cancelled,
#[error(transparent)]
InternalServerError(anyhow::Error),
}
@@ -77,10 +74,6 @@ impl ApiError {
err.to_string(),
StatusCode::REQUEST_TIMEOUT,
),
ApiError::Cancelled => HttpErrorBody::response_from_msg_and_status(
self.to_string(),
StatusCode::INTERNAL_SERVER_ERROR,
),
ApiError::InternalServerError(err) => HttpErrorBody::response_from_msg_and_status(
err.to_string(),
StatusCode::INTERNAL_SERVER_ERROR,
@@ -140,7 +133,6 @@ pub fn api_error_handler(api_error: ApiError) -> Response<Body> {
ApiError::InternalServerError(_) => error!("Error processing HTTP request: {api_error:?}"),
ApiError::ShuttingDown => info!("Shut down while processing HTTP request"),
ApiError::Timeout(_) => info!("Timeout while processing HTTP request: {api_error:#}"),
ApiError::Cancelled => info!("Request cancelled while processing HTTP request"),
_ => info!("Error processing HTTP request: {api_error:#}"),
}

View File

@@ -8,15 +8,22 @@ use super::error::ApiError;
pub async fn json_request<T: for<'de> Deserialize<'de>>(
request: &mut Request<Body>,
) -> Result<T, ApiError> {
json_request_or_empty_body(request)
.await?
.context("missing request body")
.map_err(ApiError::BadRequest)
}
/// Will be removed as part of <https://github.com/neondatabase/neon/issues/4282>
pub async fn json_request_or_empty_body<T: for<'de> Deserialize<'de>>(
request: &mut Request<Body>,
) -> Result<Option<T>, ApiError> {
let body = hyper::body::aggregate(request.body_mut())
.await
.context("Failed to read request body")
.map_err(ApiError::BadRequest)?;
if body.remaining() == 0 {
return Err(ApiError::BadRequest(anyhow::anyhow!(
"missing request body"
)));
return Ok(None);
}
let mut deser = serde_json::de::Deserializer::from_reader(body.reader());
@@ -24,6 +31,7 @@ pub async fn json_request<T: for<'de> Deserialize<'de>>(
serde_path_to_error::deserialize(&mut deser)
// intentionally stringify because the debug version is not helpful in python logs
.map_err(|e| anyhow::anyhow!("Failed to parse json request: {e}"))
.map(Some)
.map_err(ApiError::BadRequest)
}

View File

@@ -135,8 +135,7 @@ impl Gate {
let started_at = std::time::Instant::now();
let mut do_close = std::pin::pin!(self.do_close());
// with 1s we rarely saw anything, let's try if we get more gate closing reasons with 100ms
let nag_after = Duration::from_millis(100);
let nag_after = Duration::from_secs(1);
let Err(_timeout) = tokio::time::timeout(nag_after, &mut do_close).await else {
return;

View File

@@ -25,8 +25,6 @@ pub struct Config {
///
/// For simplicity, this value must be greater than or equal to `memory_history_len`.
memory_history_log_interval: usize,
/// The max number of iterations to skip before logging the next iteration
memory_history_log_noskip_interval: Duration,
}
impl Default for Config {
@@ -35,7 +33,6 @@ impl Default for Config {
memory_poll_interval: Duration::from_millis(100),
memory_history_len: 5, // use 500ms of history for decision-making
memory_history_log_interval: 20, // but only log every ~2s (otherwise it's spammy)
memory_history_log_noskip_interval: Duration::from_secs(15), // but only if it's changed, or 60 seconds have passed
}
}
}
@@ -88,12 +85,7 @@ impl CgroupWatcher {
// buffer for samples that will be logged. once full, it remains so.
let history_log_len = self.config.memory_history_log_interval;
let max_skip = self.config.memory_history_log_noskip_interval;
let mut history_log_buf = vec![MemoryStatus::zeroed(); history_log_len];
let mut last_logged_memusage = MemoryStatus::zeroed();
// Ensure that we're tracking a value that's definitely in the past, as Instant::now is only guaranteed to be non-decreasing on Rust's T1-supported systems.
let mut can_skip_logs_until = Instant::now() - max_skip;
for t in 0_u64.. {
ticker.tick().await;
@@ -123,24 +115,12 @@ impl CgroupWatcher {
// equal to the logging interval, we can just log the entire buffer every time we set
// the last entry, which also means that for this log line, we can ignore that it's a
// ring buffer (because all the entries are in order of increasing time).
//
// We skip logging the data if data hasn't meaningfully changed in a while, unless
// we've already ignored previous iterations for the last max_skip period.
if i == history_log_len - 1
&& (now > can_skip_logs_until
|| !history_log_buf
.iter()
.all(|usage| last_logged_memusage.status_is_close_or_similar(usage)))
{
if i == history_log_len - 1 {
info!(
history = ?MemoryStatus::debug_slice(&history_log_buf),
summary = ?summary,
"Recent cgroup memory statistics history"
);
can_skip_logs_until = now + max_skip;
last_logged_memusage = *history_log_buf.last().unwrap();
}
updates
@@ -252,24 +232,6 @@ impl MemoryStatus {
DS(slice)
}
/// Check if the other memory status is a close or similar result.
/// Returns true if the larger value is not larger than the smaller value
/// by 1/8 of the smaller value, and within 128MiB.
/// See tests::check_similarity_behaviour for examples of behaviour
fn status_is_close_or_similar(&self, other: &MemoryStatus) -> bool {
let margin;
let diff;
if self.non_reclaimable >= other.non_reclaimable {
margin = other.non_reclaimable / 8;
diff = self.non_reclaimable - other.non_reclaimable;
} else {
margin = self.non_reclaimable / 8;
diff = other.non_reclaimable - self.non_reclaimable;
}
diff < margin && diff < 128 * 1024 * 1024
}
}
#[cfg(test)]
@@ -299,65 +261,4 @@ mod tests {
assert_eq!(values(2, 4), [9, 0, 1, 2]);
assert_eq!(values(2, 10), [3, 4, 5, 6, 7, 8, 9, 0, 1, 2]);
}
#[test]
fn check_similarity_behaviour() {
// This all accesses private methods, so we can't actually run this
// as doctests, because doctests run as an external crate.
let mut small = super::MemoryStatus {
non_reclaimable: 1024,
};
let mut large = super::MemoryStatus {
non_reclaimable: 1024 * 1024 * 1024 * 1024,
};
// objects are self-similar, no matter the size
assert!(small.status_is_close_or_similar(&small));
assert!(large.status_is_close_or_similar(&large));
// inequality is symmetric
assert!(!small.status_is_close_or_similar(&large));
assert!(!large.status_is_close_or_similar(&small));
small.non_reclaimable = 64;
large.non_reclaimable = (small.non_reclaimable / 8) * 9;
// objects are self-similar, no matter the size
assert!(small.status_is_close_or_similar(&small));
assert!(large.status_is_close_or_similar(&large));
// values are similar if the larger value is larger by less than
// 12.5%, i.e. 1/8 of the smaller value.
// In the example above, large is exactly 12.5% larger, so this doesn't
// match.
assert!(!small.status_is_close_or_similar(&large));
assert!(!large.status_is_close_or_similar(&small));
large.non_reclaimable -= 1;
assert!(large.status_is_close_or_similar(&large));
assert!(small.status_is_close_or_similar(&large));
assert!(large.status_is_close_or_similar(&small));
// The 1/8 rule only applies up to 128MiB of difference
small.non_reclaimable = 1024 * 1024 * 1024 * 1024;
large.non_reclaimable = small.non_reclaimable / 8 * 9;
assert!(small.status_is_close_or_similar(&small));
assert!(large.status_is_close_or_similar(&large));
assert!(!small.status_is_close_or_similar(&large));
assert!(!large.status_is_close_or_similar(&small));
// the large value is put just above the threshold
large.non_reclaimable = small.non_reclaimable + 128 * 1024 * 1024;
assert!(large.status_is_close_or_similar(&large));
assert!(!small.status_is_close_or_similar(&large));
assert!(!large.status_is_close_or_similar(&small));
// now below
large.non_reclaimable -= 1;
assert!(large.status_is_close_or_similar(&large));
assert!(small.status_is_close_or_similar(&large));
assert!(large.status_is_close_or_similar(&small));
}
}

View File

@@ -12,11 +12,11 @@ use futures::{
stream::{SplitSink, SplitStream},
SinkExt, StreamExt,
};
use tracing::{debug, info};
use tracing::info;
use crate::protocol::{
OutboundMsg, OutboundMsgKind, ProtocolRange, ProtocolResponse, ProtocolVersion,
PROTOCOL_MAX_VERSION, PROTOCOL_MIN_VERSION,
OutboundMsg, ProtocolRange, ProtocolResponse, ProtocolVersion, PROTOCOL_MAX_VERSION,
PROTOCOL_MIN_VERSION,
};
/// The central handler for all communications in the monitor.
@@ -118,12 +118,7 @@ impl Dispatcher {
/// serialize the wrong thing and send it, since `self.sink.send` will take
/// any string.
pub async fn send(&mut self, message: OutboundMsg) -> anyhow::Result<()> {
if matches!(&message.inner, OutboundMsgKind::HealthCheck { .. }) {
debug!(?message, "sending message");
} else {
info!(?message, "sending message");
}
info!(?message, "sending message");
let json = serde_json::to_string(&message).context("failed to serialize message")?;
self.sink
.send(Message::Text(json))

View File

@@ -12,7 +12,7 @@ use axum::extract::ws::{Message, WebSocket};
use futures::StreamExt;
use tokio::sync::{broadcast, watch};
use tokio_util::sync::CancellationToken;
use tracing::{debug, error, info, warn};
use tracing::{error, info, warn};
use crate::cgroup::{self, CgroupWatcher};
use crate::dispatcher::Dispatcher;
@@ -474,29 +474,26 @@ impl Runner {
// there is a message from the agent
msg = self.dispatcher.source.next() => {
if let Some(msg) = msg {
match &msg {
// Don't use 'message' as a key as the string also uses
// that for its key
info!(?msg, "received message");
match msg {
Ok(msg) => {
let message: InboundMsg = match msg {
Message::Text(text) => {
serde_json::from_str(text).context("failed to deserialize text message")?
serde_json::from_str(&text).context("failed to deserialize text message")?
}
other => {
warn!(
// Don't use 'message' as a key as the
// string also uses that for its key
msg = ?other,
"problem processing incoming message: agent should only send text messages but received different type"
"agent should only send text messages but received different type"
);
continue
},
};
if matches!(&message.inner, InboundMsgKind::HealthCheck { .. }) {
debug!(?msg, "received message");
} else {
info!(?msg, "received message");
}
let out = match self.process_message(message.clone()).await {
Ok(Some(out)) => out,
Ok(None) => continue,
@@ -520,11 +517,7 @@ impl Runner {
.await
.context("failed to send message")?;
}
Err(e) => warn!(
error = format!("{e}"),
msg = ?msg,
"received error message"
),
Err(e) => warn!("{e}"),
}
} else {
anyhow::bail!("dispatcher connection closed")

Some files were not shown because too many files have changed in this diff Show More