adamant-kite-43734
09/03/2022, 9:46 AMsticky-summer-13450
09/03/2022, 10:50 AMI'm thinking, kill the pod and see what happens...I did that, and the logs seem to be showing the same looping behaviour.
...
Upgrading Harvester
<http://managedchart.management.cattle.io/harvester-crd|managedchart.management.cattle.io/harvester-crd> patched (no change)
<http://managedchart.management.cattle.io/harvester|managedchart.management.cattle.io/harvester> unchanged
<http://managedchart.management.cattle.io/harvester|managedchart.management.cattle.io/harvester> patched
<http://managedchart.management.cattle.io/harvester-crd|managedchart.management.cattle.io/harvester-crd> patched
Waiting for ManagedChart fleet-local/harvester from generation 11
Target version: 1.0.3, Target state: modified
Current version: 1.0.3, Current state: ErrApplied, Current generation: 12
Sleep for 5 seconds to retry
Current version: 1.0.3, Current state: ErrApplied, Current generation: 12
Sleep for 5 seconds to retry
...
Any ideas what I do now?sticky-summer-13450
09/04/2022, 8:49 AMkubectl get <http://managedcharts.management.cattle.io|managedcharts.management.cattle.io> harvester -n fleet-local -o yaml
and extracting some values.
The summary at the end of the yaml says:
summary:
desiredReady: 1
errApplied: 1
nonReadyResources:
- bundleState: ErrApplied
message: another operation (install/upgrade/rollback) is in progress
name: fleet-local/local
ready: 0
unavailable: 1
unavailablePartitions: 0
Is there anything else I can kill or kick in order to get past this issue?prehistoric-balloon-31801
09/05/2022, 1:26 AMsticky-summer-13450
09/05/2022, 10:21 AMhelm
command I needed was helm rollback harvester
.
After running that and waiting a few minutes, the pod logs of the upgrade job I was watching started changing, as the deployment happened, the Upgrading System Service
has successfully completed and one of the Harvester nods is in the pre-drained
state.
Hopefully the node won't get stuck in this state... It's only been 30 minutes...sticky-summer-13450
09/05/2022, 1:55 PMPre-drained
for over 4 hours, the upgrade job has completed, but there is no reboot happening.
$ kubectl get jobs -n harvester-system -l <http://harvesterhci.io/upgradeComponent=node|harvesterhci.io/upgradeComponent=node> --context=harvester003
NAME COMPLETIONS DURATION AGE
hvst-upgrade-bz6gs-pre-drain-harvester003 1/1 5m 4h7m
Maybe I should reboot the node manually and see what happens...sticky-summer-13450
09/06/2022, 11:14 AMsticky-summer-13450
09/06/2022, 11:14 AM