Proxmox VE and Kubernetes Network

[debtolee]

H i
I created two Proxmox VE environments. And I installed Kubernetes to create master 3 worker 6.
Then I installed bitnami/postgresql-ha with replica 6 using helm.

helm install postgresql-ha -n ${nsc} db-postgres/postgresql-ha -f db-postgres/postgresql-ha/values.yaml \
  --set postgresql.password=MyPassword \
  --set postgresql.repmgrPassword=MyPassword \
  --set pgpool.adminPassword=MyPassword \
  --set postgresql.pgHbaTrustAll=true \
  --set postgresql.maxConnections=900 \
  --set postgresql.replicaCount=6 \
  --set postgresql.usePgRewind=true \
  --set postgresql.extendedConf="wal_log_hints = on" \
  --set postgresql.image.debug=true \
  --set volumePermissions.enabled=true \
  --set pgpool.image.debug=true \
  --set metrics.image.debug=true \
  --set backup.enabled=true \
  --set backup.cronjob.storage.existingClaim="nfs-db-postgres-ha-backup" \
  --set backup.cronjob.storage.size=4Gi \
  --set service.type=NodePort \
  --set service.nodePorts.postgresql=MyPort
  # --set postgresql.upgradeRepmgrExtension=true

helm upgrade  postgresql-ha -n ${nsc} db-postgres/postgresql-ha -f db-postgres/postgresql-ha/values.yaml \
  --set postgresql.password=MyPassword \
  --set postgresql.repmgrPassword=MyPassword \
  --set pgpool.adminPassword=MyPassword \
  --set postgresql.pgHbaTrustAll=true \
  --set postgresql.maxConnections=900 \
  --set postgresql.replicaCount=6 \
  --set postgresql.usePgRewind=true \
  --set postgresql.extendedConf="wal_log_hints = on" \
  --set postgresql.image.debug=true \
  --set volumePermissions.enabled=true \
  --set pgpool.image.debug=true \
  --set metrics.image.debug=true \
  --set backup.enabled=true \
  --set backup.cronjob.storage.existingClaim="nfs-db-postgres-ha-backup" \
  --set backup.cronjob.storage.size=4Gi \
  --set service.type=NodePort \
  --set service.nodePorts.postgresql=MyPort \
  --set postgresql.upgradeRepmgrExtension=true

But only one pod works and the rest goes to 0/1 Running state.

Pod List is

postgresql-ha-pgpool-75c885bf49-hwgs4   1/1     Running   1 (28s ago)     2m7s
postgresql-ha-postgresql-0              1/1     Running   0               27m
postgresql-ha-postgresql-1              0/1     Running   0               2m6s
postgresql-ha-postgresql-2              0/1     Running   0               2m6s
postgresql-ha-postgresql-3              0/1     Running   0               2m7s
postgresql-ha-postgresql-4              0/1     Running   0               2m7s
postgresql-ha-postgresql-5              0/1     Running   0               2m7s

Pod Event Log is

Events:
  Type     Reason     Age                From               Message
  ----     ------     ----               ----               -------
  Normal   Scheduled  70s                default-scheduler  Successfully assigned icas/postgresql-ha-postgresql-5 to pve3-k8s-worker-6
  Normal   Pulled     69s                kubelet            Container image "docker.io/bitnami/os-shell:12-debian-12-r31" already present on machine
  Normal   Created    69s                kubelet            Created container init-chmod-data
  Normal   Started    69s                kubelet            Started container init-chmod-data
  Normal   Pulled     58s                kubelet            Container image "docker.io/bitnami/postgresql-repmgr:16.4.0-debian-12-r27" already present on machine
  Normal   Created    58s                kubelet            Created container postgresql
  Normal   Started    58s                kubelet            Started container postgresql
  Warning  Unhealthy  10s (x5 over 50s)  kubelet            Readiness probe failed: 127.0.0.1:5432 - no response
  Warning  Unhealthy  10s (x2 over 20s)  kubelet            Liveness probe failed: psql: error: connection to server at "127.0.0.1", port 5432 failed: Connection refused

It worked without any problems when configured with nodes in a single pc, but this time, when I tried to install postgresql after configuring Kubernetes master node 3 and workernode 6 in two nodes of the proxmox ve environment, respectively, this issue arose.

I don't know what I'm gonna do.

 

[dherzig]

Hi debtolee!

I created two Proxmox VE environments. And I installed Kubernetes to create master 3 worker 6.

Does this mean you have a PVE cluster consisting of 2 servers? And you placed your kubernetes vms (3x controlplanes, 6x workers) spread over this cluster?

Could you please post the output of:

pvecm status

from one of your PVE hosts?

And also the output of:

kubectl get nodes

from within one of your kubernetes master nodes?

Also, what hardware are you using on your PVE cluster (CPU threads, RAM, storage, network connections)?

Best,
Daniel