Recently, we found many subscriptions failed due to the default catalogsource pod updating, as follows,
{
"message": "error using catalogsource openshift-marketplace/community-operators: failed to list bundles: rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing: dial tcp 172.30.228.106:50051: connect: connection refused\"",
"reason": "ErrorPreventedResolution",
"status": "True",
"type": "ResolutionFailed"
}
],
"lastUpdated": "2024-08-23T14:57:10Z"
I0823 15:00:10.519172 126112 olm_utils.go:1269] $oc get [pod -n openshift-marketplace], the returned resource:NAME READY STATUS RESTARTS AGE
...
certified-operators-5z7b8 0/1 Pending 0 122m
certified-operators-k2h95 1/1 Running 0 127m
community-operators-4fjfl 0/1 Init:1/2 0 5h16m
community-operators-lhskw 0/1 Init:1/2 0 127m
eef836905a36170ccb3a1f7b1b137f6ef530cbc7153e251c4dbab288f8fvhxg 0/1 Completed 0 3h58m
f45be18175a8996867af9317708a38d0a6373cd4e8d031535b1b435104m8bvm 0/1 Completed 0 5h24m
ffa31d2a7abf1fcb0aed18b7abc3beb698e410ec56c04790fb235221cbtqcb9 0/1 Completed 0 6h9m
marketplace-operator-5647669d88-zgdhm 1/1 Running 0 7h52m
qe-app-registry-b6fhm 1/1 Running 0 7h27m
redhat-marketplace-24bss 1/1 Running 0 7h42m
redhat-marketplace-m5445 0/1 Pending 0 122m
redhat-operators-5mtcd 0/1 Init:1/2 0 5h7m
redhat-operators-qsrl7 1/1 Running 0 7h42m
The catalogsource pod will pull the index image always when updating(every 10 mins as default), sometimes it costs time and blocks other operator subscriptions. So, I wonder if we can do HA for those default catalogsource pods, or any suggest? Thanks!
- relates to
-
CMP-3561 Errors encountered while listing bundles in compliance-operator
-
- New
-
