fix(db): restore postgres command dropped by entrypoint override

Overriding entrypoint in compose clears the image default CMD, leaving
pgautoupgrade with no $1=postgres, so its version logic ran with unset
PGTARGET_MAJOR and aborted. Restore command: postgres.

.env.sample

@@ -21,9 +21,3 @@ DISCOURSE_DEVELOPER_EMAILS=admin@example.com
 #SECRET_SMTP_PASSWORD_VERSION=v1
 
 SECRET_DB_PASSWORD_VERSION=v1
-
-# Postgres bootstrap superuser (the cluster's "install user"). Defaults to
-# `postgres`, which matches fresh installs and bitnami-origin clusters. Only set
-# this if you are upgrading a cluster that was bootstrapped with a different
-# superuser (e.g. `discourse`) — a postgres major upgrade fails unless it matches.
-#POSTGRES_USER=postgres

README.md

@@ -43,22 +43,6 @@ override) so it works behind the reverse proxy.
 abra app run YOURAPPDOMAIN app discourse admin create
 ```
 
-## Postgres major version upgrades
-
-Handled automatically by the [`discourse/postgres`] image (pgvector + an
-auto-upgrade layer). On deploy it finds an older cluster, installs the old
-binaries and runs `pg_upgrade` into the new versioned data directory. No manual
-dump/restore needed.
-
-`pg_upgrade` must run as the old cluster's bootstrap superuser (its "install
-user"). The recipe uses `POSTGRES_USER`, which defaults to `postgres` — the right
-value for fresh installs and for clusters that came from the old bitnami recipe.
-If your cluster was bootstrapped with a different superuser (e.g. `discourse`),
-set `POSTGRES_USER` in the app `.env` before upgrading, otherwise `pg_upgrade`
-will refuse with an install-user mismatch.
-
-[`discourse/postgres`]: https://github.com/discourse/discourse-postgres
-
 ## Migrating from the previous (bitnami) recipe
 
 The official image stores uploads under `/shared` rather than bitnami's

abra.sh

@@ -1,4 +1,5 @@
-export PG_BACKUP_VERSION=v4
+export DB_ENTRYPOINT_VERSION=v4
+export PG_BACKUP_VERSION=v2
 export APP_ENTRYPOINT_VERSION=v2
 export APP_INSTALL_SSL_VERSION=v1
 export APP_MIGRATE_UPLOADS_VERSION=v1

cc-db-entrypoint.sh

@@ -0,0 +1,35 @@
+#!/bin/bash
+# Co-op Cloud wrapper around pgautoupgrade's entrypoint.
+#
+# pgautoupgrade runs pg_upgrade assuming the old cluster's bootstrap superuser
+# (the "install user", oid 10) equals $POSTGRES_USER. That is not always true
+# (e.g. a cluster bootstrapped with the default "postgres" superuser and a
+# separate "discourse" app role), and a mismatch makes pg_upgrade's consistency
+# check fail (pgautoupgrade issue #115). When a major-version upgrade is pending,
+# detect the old cluster's real install user and run the upgrade as that user;
+# pgautoupgrade handles everything else (old binaries, initdb, pg_upgrade).
+set -e
+
+if [ -s "$PGDATA/PG_VERSION" ]; then
+  OLD_VERSION="$(cat "$PGDATA/PG_VERSION")"
+  TARGET_VERSION="${PGTARGET%%.*}"
+
+  if [ -n "$OLD_VERSION" ] && [ -n "$TARGET_VERSION" ] && [ "$OLD_VERSION" != "$TARGET_VERSION" ]; then
+    # query the old cluster in single-user mode (no socket/auth needed) with
+    # pgautoupgrade's bundled old-version binaries
+    OLD_POSTGRES="/usr/local-pg${OLD_VERSION}/bin/postgres"
+    if [ -x "$OLD_POSTGRES" ]; then
+      INSTALL_USER="$(echo 'select rolname from pg_authid where oid = 10;' \
+        | gosu postgres "$OLD_POSTGRES" --single -D "$PGDATA" template1 2>/dev/null \
+        | sed -n 's/.*rolname = "\([^"]*\)".*/\1/p' | head -n1)"
+      if [ -n "$INSTALL_USER" ]; then
+        echo "cc-db-entrypoint: old cluster (pg$OLD_VERSION) install user is '$INSTALL_USER'; running the pg$TARGET_VERSION upgrade as that user"
+        export POSTGRES_USER="$INSTALL_USER"
+      else
+        echo "cc-db-entrypoint: WARNING could not detect old install user; letting pgautoupgrade use POSTGRES_USER=$POSTGRES_USER"
+      fi
+    fi
+  fi
+fi
+
+exec /usr/local/bin/docker-entrypoint.sh "$@"

compose.yml

@@ -63,51 +63,39 @@ services:
       start_period: 25m
 
   db:
-    # discourse/postgres = pgvector + discourse's postgres management layer, which
+    image: pgautoupgrade/pgautoupgrade:17-trixie
-    # auto-upgrades an older cluster in place on boot (pg_upgrade into the versioned
-    # PGDATA /var/lib/postgresql/${MAJOR}/docker); everything is driven by the env below.
-    image: discourse/postgres:pg18
     networks:
       - internal
     secrets:
       - db_password
     volumes:
-      # the image expects the whole cluster tree mounted here (not the data subdir);
+      - 'postgresql_data:/var/lib/postgresql/data'
-      # an existing pg17 cluster at the volume root is found and upgraded into /18/docker
-      - 'postgresql_data:/var/lib/postgresql'
     configs:
+      - source: db_entrypoint
+        target: /usr/local/bin/cc-db-entrypoint.sh
+        mode: 0555
       - source: pg_backup
         target: /pg_backup.sh
         mode: 0555
-    entrypoint:
+    entrypoint: /usr/local/bin/cc-db-entrypoint.sh
-      - /bin/bash
+    # overriding entrypoint drops the image's default CMD, so restore it; the
-      - -c
+    # wrapper passes "$@" through and pgautoupgrade keys its logic on $1=postgres
-      - |
+    command: postgres
-        if [ -f /run/secrets/db_password ]; then
-          DB_PASSWORD="$$(cat /run/secrets/db_password)"
-          export DB_PASSWORD POSTGRES_PASSWORD="$$DB_PASSWORD"
-        fi
-        exec run-postgres.sh postgres
     environment:
-      # internal-only overlay network; keep all-trust so the app and the
-      # backup/restore hooks connect without juggling the superuser password
       - POSTGRES_HOST_AUTH_METHOD=trust
+      - POSTGRES_USER=discourse
       - POSTGRES_DB=discourse
-      - DB_USER=discourse
+      - POSTGRES_PASSWORD_FILE=/run/secrets/db_password
-      # pg_upgrade runs as this role and initdb's the new cluster with it; it must
+      - PGTARGET=17
-      # match the OLD cluster's bootstrap superuser (oid 10). The image default
-      # `postgres` matches fresh installs and bitnami-origin clusters. Override in
-      # the app .env (POSTGRES_USER=...) only for a cluster bootstrapped differently.
-      - POSTGRES_USER=${POSTGRES_USER:-postgres}
-      # pg18's initdb enables data checksums by default, but pg13-17 clusters here
-      # have them off and pg_upgrade requires a match -> initialise without them.
-      - POSTGRES_INITDB_ARGS=--no-data-checksums
     healthcheck:
       test: "pg_isready -U discourse -d discourse"
       interval: 30s
       timeout: 10s
       retries: 5
-      start_period: 15m
+      # generous: pgautoupgrade may run an in-place pg_upgrade on deploy before
+      # the server accepts connections — don't let the healthcheck kill an
+      # in-progress migration
+      start_period: 10m
     deploy:
       labels:
         backupbot.backup: "true"
@@ -154,6 +142,9 @@ configs:
   app_migrate_uploads:
     name: ${STACK_NAME}_app_migrate_uploads_${APP_MIGRATE_UPLOADS_VERSION}
     file: migrate-uploads.sh
+  db_entrypoint:
+    name: ${STACK_NAME}_db_entrypoint_${DB_ENTRYPOINT_VERSION}
+    file: cc-db-entrypoint.sh
   pg_backup:
     name: ${STACK_NAME}_pg_backup_${PG_BACKUP_VERSION}
     file: pg_backup.sh

pg_backup.sh

@@ -1,47 +1,44 @@
 #!/bin/bash
 
-# Postgres backup/restore hook for the discourse `db` service (discourse/postgres image).
+# Postgres backup/restore hook for the discourse `db` service.
 
 set -e
 
-# dump goes at the volume root so backupbot's backup.sql label finds it
+BACKUP_FILE='/var/lib/postgresql/data/backup.sql'
-BACKUP_FILE='/var/lib/postgresql/backup.sql'
+export PGPASSWORD=$(cat "${POSTGRES_PASSWORD_FILE:-/run/secrets/db_password}")
-DATADIR="${PGDATA:-/var/lib/postgresql/18/docker}"
+DB_USER="${POSTGRES_USER:-discourse}"
 DB_NAME="${POSTGRES_DB:-discourse}"
 
-# bootstrap superuser for the dump/drop/recreate; same POSTGRES_USER the db service sets
-SU="${POSTGRES_USER:-postgres}"
-
 function backup {
-  pg_dump -U "$SU" "$DB_NAME" | gzip > "$BACKUP_FILE"
+  pg_dump -U "$DB_USER" "$DB_NAME" | gzip > "$BACKUP_FILE"
 }
 
 function restore {
-  cd "$DATADIR"
+  cd /var/lib/postgresql/data/
 
   # Block all non-local connections so the running discourse app + sidekiq cannot reconnect and
   # interfere with the drop/recreate/reimport. Restored on exit.
   restore_hba() {
     cat pg_hba.conf.bak > pg_hba.conf
     rm -f pg_hba.conf.bak
-    su postgres -c "pg_ctl -D '$DATADIR' reload"
+    su postgres -c 'pg_ctl reload'
   }
   cp pg_hba.conf pg_hba.conf.bak
   echo 'local all all trust' > pg_hba.conf
-  su postgres -c "pg_ctl -D '$DATADIR' reload"
+  su postgres -c 'pg_ctl reload'
   trap restore_hba EXIT INT TERM
 
   # terminate any lingering local sessions before recreate
   # see https://stackoverflow.com/questions/5108876/kill-a-postgresql-session-connection
-  psql -U "$SU" -d postgres -c \
+  psql -U "$DB_USER" -d postgres -c \
     "SELECT pg_terminate_backend(pid) FROM pg_stat_activity WHERE datname='${DB_NAME}' AND pid<>pg_backend_pid();"
 
   # drop database and then recreate it
-  psql -U "$SU" -d postgres -c "DROP DATABASE ${DB_NAME} WITH (FORCE);"
+  psql -U "$DB_USER" -d postgres -c "DROP DATABASE ${DB_NAME} WITH (FORCE);"
-  createdb -U "$SU" "$DB_NAME"
+  createdb -U "$DB_USER" "$DB_NAME"
 
-  # reimport data
+  # reimport data 
-  gunzip -c "$BACKUP_FILE" | psql -U "$SU" -d "$DB_NAME" -1 -v ON_ERROR_STOP=1 -f -
+  gunzip -c "$BACKUP_FILE" | psql -U "$DB_USER" -d "$DB_NAME" -1 -v ON_ERROR_STOP=1 -f -
 }
 
 $@

release/1.0.0+3.5.3

@@ -8,12 +8,3 @@ Rename these in your app's .env (the values carry over):
   DISCOURSE_SMTP_USER     --> DISCOURSE_SMTP_USER_NAME
   DISCOURSE_SMTP_AUTH     --> DISCOURSE_SMTP_AUTHENTICATION
   DISCOURSE_SMTP_PROTOCOL --> DISCOURSE_SMTP_ENABLE_START_TLS (takes a boolean true/false, not the old tls/ssl value, so translate it rather than copying it straight across)
-
-WARNING: if your deployment's database has an "install user" other than `postgres`
-(some older deployments do), you must set the POSTGRES_USER env var in your .env
-for this migration, otherwise the postgres upgrade aborts with an install-user
-mismatch.
-
-Check your old deployment's install user before upgrading (if this command returns postgres, then you do not need to set this env):
-
-  abra app run YOURAPPDOMAIN db -- psql -U discourse -tAc 'select rolname from pg_roles where oid = 10'