refactor(postgresql.conf): configure postgresql.conf to use 'include_dir' directive #1843
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…dir' directive and move logging.conf into the postgresql-customer dir
… preserve ordering
* 'develop' of github.com:supabase/postgres: (29 commits) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1818) fix: update Dockerfiles for changes to postgis multiversion (#1817) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1813) feat: support multiple versions of the postgis extension (#1667) Update pgsql-http to handle semver (#1816) feat: add retry policy for auth service routes (#1782) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1811) fix: move tmpdir for SAA to one that always exists (#1799) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1810) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1809) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1807) build: automatically cancel old tests/build on new push (#1808) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1804) docs: small adjust (#1806) docs: getting started guide in README (#1805) fix: templating of SAA service file fix: adjust the qemu vars filename refactor(ansible): bring our ansible up to modern ansible-lint standards feat: support multiple versions of the http extension (#1664) feat: supply a slightly different supabase-admin-agent configuration for qemu artifacts ...
…les to ensure ordering
* origin/develop: fix(nix): remove '%' character from Nix trusted-public-keys configuration (#1840) chore: systemd hardening (#1837) Revert "refactor(postgresq): switch to 'include_dir' and then rename conf files to ensure ordering" (#1835) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1821) fix(pgmq): replace drop_queue function if exists (#1828) refactor(postgresq): switch to 'include_dir' and then renames conf files to ensure ordering (#1820) feat: support multiple versions of the pg_jsonschema extension (#1757) feat: bump auth to v2.180.0 (#1829) feat: update supautils confs w/ new tables feat(migrations): predefined role grants (#1815) feat: run pg_regress tests after installing the last version of the extension (#1826) fix(pgmq): add missing helper function in migration script (#1825) feat: support multiple versions of the pgmq extension (#1668) feat: run pg_regress during extension tests (#1812) feat: support multiple versions of the pgrouting extension (#1687) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1819) feat: support multiple versions of the pg-graphql extension (#1761)
…e numeric prefixes as per Sam
* fix: refactor tests to work with include_dir approach and be more accurate with production conditions * tests: use distinct savepoint names in test
* origin: refactor(ansible): bring our ansible up to modern ansible-lint standards (#1852) chore: bump to release (#1855) docs: getting started nix install or config revised (#1853) docs: how to update new structure (#1851) feat: support multiple versions of the plpgsql_check extension (#1684) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1848) feat(wrappers): support more versions (#1831) refactor(ansible): bring our ansible up to modern ansible-lint standards (#1833) fix: fine tune protection rules to unblock wal-g functionality (#1846) fix: incorporate v3.0.0 supautils (#1844) fix: remove git revision from postgres package
Collaborator
|
@hunleyd this is still in draft mode of course. But requesting that once we are ready for review, we need @soedirgo @steve-chavez @za-arthur and myself to review before we merge and deploy just due to all of the changes. |
Contributor
|
getting supautils related errors |
Contributor
|
|
Contributor
Author
Collaborator
|
@hunleyd see if you agree with all of this Recommending remove the symlink entirely and maintain your current approach of configs in /etc/postgresql/ with lots of testing to make sure nothing breaks:
Just please test as much as you can in local infra, and we should run e2e tests against your branch. Please create a playbook to document running e2e tests, but also reach out for help if you get stuck trying to do it. Thanks for the careful work on this. |
* 'develop' of github.com:supabase/postgres: fix: don't run a nix installer on qemu image build (#1916) fix: restore ami release to us east 1 (#1913) feat: upgrade admin-mgr to 0.32.3 (#1915) fix: Prepare job (lines 23-25): Moved checkout before nix-install-ephemeral (#1914) feat: support multiple versions of the wal2json extension (#1875)
This change adds a task to remove `pg_hba.conf`, `pg_ident.conf`, and `postgresql.conf` from the PostgreSQL data directory. This ensures there is no confusion about what configuration file is active.
Symlinks PostgreSQL configuration files (`pg_hba.conf`, `pg_ident.conf`, `postgresql.conf`) from `/etc/postgresql/` into the PGDATA directory (`/var/lib/postgresql/data/`). This ensures that the configuration files within PGDATA are always managed centrally and point to the correct, actively used configuration. The `force: true` option is added to overwrite any existing files, guaranteeing the symlink is correctly established.
* origin: feat: support multiple versions of the pg_plan_filter extension (#1873)
- Explicitly set the data directory using `--pgdata` in `pg_ctl initdb` commands.
* 'develop' of github.com:supabase/postgres: fix: add alpn_protocols to envoy (#1503)
…t gets rid of the arg errors
…-152 * 'INDATA-152' of github.com:supabase/postgres:
- Standardize Ansible module usage for PostgreSQL configuration files and related tasks. - Update the default PostgreSQL data directory for `initdb` to `/data/pgdata`.
This commit updates the Ansible tasks for setting up PostgreSQL. The primary changes involve: - Shifting the `initdb` executable path from `/usr/lib/postgresql/bin/pg_ctl` to `/var/lib/postgresql/.nix-profile/bin/initdb`, reflecting a Nix-based installation. - Adjusting the arguments passed to `initdb` for clarity and correctness. - Updating the `creates` path in the Ansible `args` to match the new data directory structure.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Moved all config files that are included in the postgresql.conf into the /etc/postgresql-custom dir then updated postgresql.conf to use include_dir instead of include. To ensure the same ordering of the various GUCs (to ensure proper overriding of values), rename files with digits prepended.
Finally, split auto_explain and pg_cron out into their own config files instead of them being tacked on the end of the main PG config.
this is take 2, which should fix the supautils issue