We have a Kubernetes cluster running on Google Container Engine. The Container clusters section of the Google cloud console shows that the cluster api version is 1.06.
I tried to do a rolling update to upgrade to the latest api version (1.1.1 is currently most up to date available).
I ran:
gcloud container clusters upgrade my-cluster-name
and received an error (identifiers removed)
(gcloud.container.clusters.upgrade) Operation [https://container.googleapis.com/v1/projects/' status: StatusValueValuesEnum(DONE, 3) statusMessage: u'Error upgrading nodes: Cluster nodes did not update in time. Rolling update with ID "1" last had status: "2/3 instances updated; Errors: CONDITION_NOT_MET (Too many (1 > 0) instances failed.)".' targetLink: u'https://container.googleapis.com/v1/projects/' zone: u'1'>] finished with error: Error upgrading nodes: Cluster nodes did not update in time. Rolling update with ID "1" last had status: "2/3 instances updated; Errors: CONDITION_NOT_MET (Too many (1 > 0) instances failed.)".
I tried running the command again and received this error:
(gcloud.container.clusters.upgrade) Operation [https://container.googleapis.com/v1/projects/1' status: StatusValueValuesEnum(DONE, 3) statusMessage: u'Error upgrading nodes: A problem occurred during the rolling update.' targetLink: u'https://container.googleapis.com/v1/projects/1' zone: u'1'>] finished with error: Error upgrading nodes: A problem occurred during the rolling update.
any ideas?