Cluster outage - Vault Certificate Expired
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
vault-charm |
Incomplete
|
Undecided
|
Unassigned |
Bug Description
We experienced a full cluster outage that appears to be caused by expired certificates.
The symptoms encountered were:
- kubectl commands returned: Unable to connect to the server: x509: certificate has expired or is not yet valid
- all etcd nodes reported: Errored with 0 known peers
- certs in /root/cdk on master and workers were expired with a lifespan of 30 days
The behavior I saw was identical to the issue described here:
https:/
We are using the standard charmed-kubernetes bundle with the vault overlay
cs:~containers/
cs:~containers/
cs:~containers/
cs:~containers/
cs:~containers/
cs:percona-cluster
cs:~openstack-
I had been under the impression that the commit for this ticket would have prevented this issue:
https:/
Since we were using vault-41, we thought we had the fix.
I was hoping someone could clarify if this is a known issue or if it has been fixed? I also saw this ticket: https:/
We are using cs:~openstack-
The revisions between cs:vault-41 and cs:~openstack- charmers- next/vault- 41 are not the same, they track independent lists of versions. Can you confirm what version of the charm you were using, using a fully qualified version?