"state changing too quickly" after juju upgrade and redeploy of application
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Canonical Juju |
Expired
|
High
|
Unassigned |
Bug Description
After performing a snap refresh of Juju from 2.8 to 2.9, Juju has an error deploying some charms. Steps to reproduce:
- Started with fresh Ubuntu 18.04 VM (occurs on 20.04 too)
sudo snap install juju --channel 2.8 --classic
sudo snap install microk8s --classic
microk8s.enable storage dns ingress
echo <IP Address>-<IP Address> | microk8s.enable metallb
juju bootstrap microk8s microk8s --config controller-
juju add-model model microk8s
juju deploy cs:~charmed-
- Wait for deployment to complete
juju remove-application kafka-k8s
- Wait for application to be removed. In my original case, the application removed succesfully, but sometimes a --force is needed if it remains stuck in `unknown`.
sudo snap refresh juju --channel 2.9/stable
juju upgrade-controller
juju upgrade-model
- Once upgraded, attempt to deploy the application
juju deploy kafka-k8s --channel latest/edge
At this point the following error message is seen:
Located charm "kafka-k8s" in charm-hub, revision 3
Deploying "kafka-k8s" from charm-hub charm "kafka-k8s", revision 3 in channel edge
ERROR cannot add application "kafka-k8s": state changing too quickly; try again soon
Changed in juju: | |
status: | New → Triaged |
importance: | Undecided → High |
Changed in juju: | |
milestone: | none → 2.9.29 |
Changed in juju: | |
milestone: | 2.9.29 → 2.9.30 |
Changed in juju: | |
milestone: | 2.9.30 → 2.9-next |
Changed in juju: | |
milestone: | 2.9-next → none |
I think this may have been fixed with work done in 2.9.28 and 2.9.29.
I'll mark as incomplete and we can reopen if the issue can be reproduced.