Description
I just updated my chart to use the latest version (2.29.1) and every time it starts, the pihole container shows [✗] DNS resolution is currently unavailable for a while with failing probes, before restarting and then finally resolving DNS. However then it starts outputting the following error over and over
2025-03-08 21:48:43.090 UTC [49M] INFO: Blocking status is enabled
2025-03-08 21:48:43.189 UTC [49/T79] ERROR: gravityDB_count(SELECT value FROM info WHERE property = 'gravity_count';) - SQL error step no more rows available
2025-03-08 21:48:43.189 UTC [49/T79] WARNING: Count of gravity domains not available. Please run pihole -g
2025-03-08 21:48:43.191 UTC [49/T79] INFO: Compiled 0 allow and 0 deny regex for 1 client in 1.1 msec
2025-03-08 21:48:44.003 UTC [49/T79] ERROR: gravity_updated(): SELECT value FROM info WHERE property = 'updated'; - SQL error step: no more rows available
I got into the container shell and ran pihole -g which successfully updated gravity and stopped the errors, but that's not ideal as it won't persist through pod restarts.
Once I ran pihole -g it started working, although the UI now shows
Here are my values - maybe something changed in the new release that's causing my existing values to conflict?
nameOverride: pihole
serviceDns:
mixedService: true
type: ClusterIP
admin:
existingSecret: pihole-admin
doh:
enabled: true
replicaCount: 1
podDisruptionBudget:
enabled: false
I tried disabling DoH to see if that made a difference but it had the same effect. Rolling back the chart to 2.27.0 immediately resolved the issue and everything was working as expected.
If it helps, I'm running it on K8s v1.31.1 using arm64 architecture.