docs: describe instanceSidecarConfiguration

Document how the instance sidecar is configured and what happens when
resources definitions conflict between objectstores.

Signed-off-by: Francesco Canovai <francesco.canovai@enterprisedb.com>
This commit is contained in:
Francesco Canovai 2025-05-09 12:12:16 +02:00 committed by Marco Nenciarini
parent fd6f3e9b10
commit 86e7377ef9
2 changed files with 4 additions and 4 deletions

View File

@ -5,7 +5,7 @@ metadata:
spec: spec:
retentionPolicy: "1m" retentionPolicy: "1m"
instanceSidecarConfiguration: instanceSidecarConfiguration:
retentionPolicyIntervalSeconds: 30 retentionPolicyIntervalSeconds: 1800
resources: resources:
requests: requests:
memory: "64Mi" memory: "64Mi"
@ -34,4 +34,3 @@ spec:
- "--min-chunk-size=5MB" - "--min-chunk-size=5MB"
- "--read-timeout=60" - "--read-timeout=60"
- "-vv" - "-vv"

View File

@ -228,7 +228,7 @@ spec:
configuration: configuration:
# [...] # [...]
instanceSidecarConfiguration: instanceSidecarConfiguration:
retentionPolicyIntervalSeconds: 30 retentionPolicyIntervalSeconds: 1800
resources: resources:
requests: requests:
memory: "64Mi" memory: "64Mi"
@ -243,4 +243,5 @@ PostgreSQL instance pod. Even if you define multiple `ObjectStore` resources,
only one sidecar will run per instance. If a replica cluster also archives WALs only one sidecar will run per instance. If a replica cluster also archives WALs
to a different `ObjectStore`, the sidecar will use the resource settings from the to a different `ObjectStore`, the sidecar will use the resource settings from the
`ObjectStore` referenced by the archiving plugin, not the one in the `ObjectStore` referenced by the archiving plugin, not the one in the
`.spec.externalClusters` section. `.spec.externalClusters` section. Changes to the sidecar configuration will be
applied on the next reconciliation of the `Cluster` resources it.