rt/wrolpi
1
0
Fork 0
mirror of https://github.com/lrnselfreliance/wrolpi.git synced 2026-10-26 10:51:36 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions 5

repair.sh: heal Postgres cluster ownership non-destructively.

Browse source 
The only self-service fix for a permission-broken cluster was
reset_api_db.sh, which drops the database.  Repair now restores
postgres ownership of /var/lib/postgresql and /etc/postgresql and
restarts Postgres before configuring the database, same as the reset
script but keeping the data.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Roland 2026-07-06 20:06:21 -06:00
parent 3180a29e82
commit fdf65e218e
No known key found for this signature in database
GPG key ID: 1BA620CE6C9A4680
1 changed files with 11 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

11
repair.sh
View file

@ -143,6 +143,17 @@ chmod 0440 /etc/sudoers.d/90-wrolpi
# Verify this new file is valid.
visudo -c -f /etc/sudoers.d/90-wrolpi
# Heal Postgres cluster ownership before configuring the database. A stray
# recursive chown can leave cluster files unreadable by the server: the
# postmaster stays up but new backends die with FATAL: could not open file
# "global/pg_filenode.map": Permission denied. Same healing as
# reset_api_db.sh, but without dropping the database.
if id postgres >/dev/null 2>&1; then
[ -d /var/lib/postgresql ] && chown -R postgres:postgres /var/lib/postgresql 2>/dev/null || :
[ -d /etc/postgresql ] && chown -R postgres:postgres /etc/postgresql 2>/dev/null || :
systemctl restart postgresql 2>/dev/null || :
fi
# Configure Postgresql. Do this after the API is stopped. Never fatal: a
# repair/upgrade must complete its system-level work even when the database
# is down (e.g. it is the reboot after this repair that revives Postgres).