nsinnovations/erpnext
2
0
Fork 0
mirror of https://github.com/frappe/erpnext.git synced 2026-08-19 09:28:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

ci(postgres): fail setup if pg_ctl stop fails before baking the datadir

Browse Source 
The "Stop DB and stage datadir" step swallowed a failed `pg_ctl -m fast -w
stop` with `|| true`, then moved and tarred the PGDATA regardless. A stop
that times out or errors would bake a still-running, crash-inconsistent
cluster into the artifact every test shard consumes — and with
full_page_writes off, crash recovery can't repair torn pages. Drop the
`|| true` so a failed stop fails the job, mirroring the MariaDB sister's
"don't bake a dirty datadir" guard.

Also drop the redundant `ALTER SYSTEM SET fsync/synchronous_commit/
full_page_writes = off` block from install.sh. Its comment claimed the
postgres workflow "runs a service-container DB and never calls start-db.sh",
but it does call start-db.sh, which already applies those flags via `-o` on
every postgres start (setup job and each shard). The block was a no-op and
its justification was factually wrong.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
Mihir Kandoi
2026-06-29 21:29:22 +05:30
parent 6f97c7199c
commit b93a3bca16
2 changed files with 9 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
.github/helper/install.sh vendored
View File

@@ -297,14 +297,10 @@ if [ "$DB" == "postgres" ];then
echo "travis" | psql -h 127.0.0.1 -p 5432 -c "CREATE DATABASE test_frappe" -U postgres;
echo "travis" | psql -h 127.0.0.1 -p 5432 -c "CREATE USER test_frappe WITH PASSWORD 'test_frappe'" -U postgres;
# Disposable CI DB: durability off for speed (postgres fsyncs every commit by default, which
# dominates a commit-heavy suite). All reloadable, no restart. The postgres workflow runs a
# service-container DB and never calls start-db.sh, so the flags must be applied here.
echo "travis" | psql -h 127.0.0.1 -p 5432 -U postgres \
-c "ALTER SYSTEM SET synchronous_commit = 'off'" \
-c "ALTER SYSTEM SET fsync = 'off'" \
-c "ALTER SYSTEM SET full_page_writes = 'off'" \
-c "SELECT pg_reload_conf()";
# Durability-off for speed (no fsync/synchronous_commit/full_page_writes) is applied by
# start-db.sh's postgres `-o` flags on every start — setup job AND each test shard — so it is
# NOT repeated here. The postgres workflow runs in-runner via start-db.sh, not a service
# container.
fi
cd ~/frappe-bench || exit

6
.github/workflows/server-tests-postgres.yml vendored
View File

@@ -108,7 +108,11 @@ jobs:
- name: Stop DB and stage datadir
run: |
PG_BIN=$(ls -d /usr/lib/postgresql/*/bin | sort -V | tail -1)
"$PG_BIN/pg_ctl" -D /home/runner/pgdata -m fast -w stop || true
# Clean shutdown so the baked datadir is consistent. Do NOT swallow a failed stop with
# `|| true`: moving and tarring a still-running cluster ships a torn datadir the shards
# cannot crash-recover (full_page_writes is off). Fail the job instead — mirrors the
# MariaDB sister's "don't bake a dirty datadir" guard.
"$PG_BIN/pg_ctl" -D /home/runner/pgdata -m fast -w stop
mv /home/runner/pgdata /home/runner/frappe-bench/pgdata
- name: Package bench for test shards