HA postgresql pod in crash loop #607

Closed
opened 2024-01-24 21:28:47 +00:00 by sli720 · 4 comments
sli720 commented 2024-01-24 21:28:47 +00:00 (Migrated from gitea.com)

I just fresh installed the newest (v10.1.0) helm chart on my k8s cluster but one of the postgress pods is not coming up. I also tried with completely reinstalling but always the same pod stays in a crash loop. If I check the pod logs I see no error. Are there maybe wrong settings in the liveness/readyness probes?

gitea-postgresql-ha-postgresql-1 0/1 CrashLoopBackOff 5 (57s ago) 4m28s

postgresql-repmgr 21:24:49.84 INFO ==>
postgresql-repmgr 21:24:49.84 INFO ==> Welcome to the Bitnami postgresql-repmgr container
postgresql-repmgr 21:24:49.84 INFO ==> Subscribe to project updates by watching https://github.com/bitnami/containers
postgresql-repmgr 21:24:49.84 INFO ==> Submit issues and feature requests at https://github.com/bitnami/containers/issues
postgresql-repmgr 21:24:49.84 INFO ==>
postgresql-repmgr 21:24:49.86 INFO ==> ** Starting PostgreSQL with Replication Manager setup **
postgresql-repmgr 21:24:49.88 INFO ==> Validating settings in REPMGR_* env vars...
postgresql-repmgr 21:24:49.88 INFO ==> Validating settings in POSTGRESQL_* env vars..
postgresql-repmgr 21:24:49.88 INFO ==> Querying all partner nodes for common upstream node...
postgresql-repmgr 21:24:50.47 INFO ==> Auto-detected primary node: 'gitea-postgresql-ha-postgresql-0.gitea-postgresql-ha-postgresql-headless.default.svc.cluster.local:5432'
postgresql-repmgr 21:24:50.47 INFO ==> Node configured as standby
postgresql-repmgr 21:24:50.47 INFO ==> Preparing PostgreSQL configuration...
postgresql-repmgr 21:24:50.48 INFO ==> postgresql.conf file not detected. Generating it...
postgresql-repmgr 21:24:50.53 INFO ==> Preparing repmgr configuration...
postgresql-repmgr 21:24:50.53 INFO ==> Initializing Repmgr...
postgresql-repmgr 21:24:50.53 INFO ==> Waiting for primary node...
postgresql-repmgr 21:24:50.71 INFO ==> Rejoining node...
postgresql-repmgr 21:24:50.72 INFO ==> Cloning data from primary node...
postgresql-repmgr 21:25:01.80 INFO ==> Initializing PostgreSQL database...
postgresql-repmgr 21:25:01.82 INFO ==> Custom configuration /opt/bitnami/postgresql/conf/postgresql.conf detected
postgresql-repmgr 21:25:01.82 INFO ==> Custom configuration /opt/bitnami/postgresql/conf/pg_hba.conf detected
postgresql-repmgr 21:25:01.84 INFO ==> Deploying PostgreSQL with persisted data...
postgresql-repmgr 21:25:01.85 INFO ==> Configuring replication parameters
postgresql-repmgr 21:25:01.87 INFO ==> Configuring fsync
postgresql-repmgr 21:25:01.88 INFO ==> Setting up streaming replication slave...
postgresql-repmgr 21:25:01.89 INFO ==> Starting PostgreSQL in background...
postgresql-repmgr 21:25:02.13 INFO ==> Unregistering standby node...
postgresql-repmgr 21:25:02.16 INFO ==> Registering Standby node...
postgresql-repmgr 21:25:04.22 INFO ==> Stopping PostgreSQL...

I just fresh installed the newest (v10.1.0) helm chart on my k8s cluster but one of the postgress pods is not coming up. I also tried with completely reinstalling but always the same pod stays in a crash loop. If I check the pod logs I see no error. Are there maybe wrong settings in the liveness/readyness probes? `gitea-postgresql-ha-postgresql-1 0/1 CrashLoopBackOff 5 (57s ago) 4m28s` > postgresql-repmgr 21:24:49.84 INFO ==> postgresql-repmgr 21:24:49.84 INFO ==> Welcome to the Bitnami postgresql-repmgr container postgresql-repmgr 21:24:49.84 INFO ==> Subscribe to project updates by watching https://github.com/bitnami/containers postgresql-repmgr 21:24:49.84 INFO ==> Submit issues and feature requests at https://github.com/bitnami/containers/issues postgresql-repmgr 21:24:49.84 INFO ==> postgresql-repmgr 21:24:49.86 INFO ==> ** Starting PostgreSQL with Replication Manager setup ** postgresql-repmgr 21:24:49.88 INFO ==> Validating settings in REPMGR_* env vars... postgresql-repmgr 21:24:49.88 INFO ==> Validating settings in POSTGRESQL_* env vars.. postgresql-repmgr 21:24:49.88 INFO ==> Querying all partner nodes for common upstream node... postgresql-repmgr 21:24:50.47 INFO ==> Auto-detected primary node: 'gitea-postgresql-ha-postgresql-0.gitea-postgresql-ha-postgresql-headless.default.svc.cluster.local:5432' postgresql-repmgr 21:24:50.47 INFO ==> Node configured as standby postgresql-repmgr 21:24:50.47 INFO ==> Preparing PostgreSQL configuration... postgresql-repmgr 21:24:50.48 INFO ==> postgresql.conf file not detected. Generating it... postgresql-repmgr 21:24:50.53 INFO ==> Preparing repmgr configuration... postgresql-repmgr 21:24:50.53 INFO ==> Initializing Repmgr... postgresql-repmgr 21:24:50.53 INFO ==> Waiting for primary node... postgresql-repmgr 21:24:50.71 INFO ==> Rejoining node... postgresql-repmgr 21:24:50.72 INFO ==> Cloning data from primary node... postgresql-repmgr 21:25:01.80 INFO ==> Initializing PostgreSQL database... postgresql-repmgr 21:25:01.82 INFO ==> Custom configuration /opt/bitnami/postgresql/conf/postgresql.conf detected postgresql-repmgr 21:25:01.82 INFO ==> Custom configuration /opt/bitnami/postgresql/conf/pg_hba.conf detected postgresql-repmgr 21:25:01.84 INFO ==> Deploying PostgreSQL with persisted data... postgresql-repmgr 21:25:01.85 INFO ==> Configuring replication parameters postgresql-repmgr 21:25:01.87 INFO ==> Configuring fsync postgresql-repmgr 21:25:01.88 INFO ==> Setting up streaming replication slave... postgresql-repmgr 21:25:01.89 INFO ==> Starting PostgreSQL in background... postgresql-repmgr 21:25:02.13 INFO ==> Unregistering standby node... postgresql-repmgr 21:25:02.16 INFO ==> Registering Standby node... postgresql-repmgr 21:25:04.22 INFO ==> Stopping PostgreSQL...
pat-s commented 2024-01-30 10:33:51 +00:00 (Migrated from gitea.com)

Please try to deploy the sub-chart in a standalone namespace. Does it work?

We don't do much modifications to it (see values.yml) and have not seen issues lately related to it. It also depends on your specific k8s env and more - but if you can deploy it standalone, it should also work within the chart.

I can't really give much more help based on the provided information.

Please try to deploy the sub-chart in a standalone namespace. Does it work? We don't do much modifications to it (see `values.yml`) and have not seen issues lately related to it. It also depends on your specific k8s env and more - but if you can deploy it standalone, it should also work within the chart. I can't really give much more help based on the provided information.
sli720 commented 2024-02-07 07:54:59 +00:00 (Migrated from gitea.com)

Is it possible to update the versions of the dependency charts e.g. postgres, redis to the latest versions somehow through a variable?

Is it possible to update the versions of the dependency charts e.g. postgres, redis to the latest versions somehow through a variable?
pat-s commented 2024-02-07 09:46:31 +00:00 (Migrated from gitea.com)

Sure, see this comment.

Sure, see [this comment](https://gitea.com/gitea/helm-chart/issues/572#issuecomment-807116).
sli720 commented 2024-02-07 22:57:46 +00:00 (Migrated from gitea.com)

This refers to the container image version. Is it also possible to select the chart version of postgres and redis?

This refers to the container image version. Is it also possible to select the chart version of postgres and redis?
Sign in to join this conversation.
No Milestone
No project
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: lunny/helm-chart#607
No description provided.