✨ Add a condition of clock sync. #296
Conversation
Signed-off-by: xuezhaojun <[email protected]>
|
/assign @qiujian16 |
| @@ -200,6 +200,8 @@ const ( | |||
| // cluster is available, the kube-apiserver is healthy and the Klusterlet agent is | |||
| // running with the minimum deployment on this managed cluster | |||
| ManagedClusterConditionAvailable string = "ManagedClusterConditionAvailable" | |||
| // ManagedClusterConditionClockSynced means the clock between the hub and the agent is synced. | |||
| ManagedClusterConditionClockSynced string = "ManagedClusterConditionClockSynced" | |||
There was a problem hiding this comment.
could it be a reason under available condition?
There was a problem hiding this comment.
The concern here is clock out-of-sync doesn't equal to Available = Unknown.
The trigger point of ClockSynced = False is that hub is 1 leaseDuration faster than the agent(the default gracePeriod is 5 * leaseDuration). At that point, the managed cluster is still available but we already want to get users attention and want to warn they to check on NTP status in their environment.
Making OutOfSync a reason of UnKnown is also a choice, but from user experience perspective, kind of feel there should be an independent condition to warn the user before things really go wrong and can't function.
|
/approve |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: qiujian16, xuezhaojun The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-merge-bot
bot
merged commit 8bf9424
into
open-cluster-management-io:main
Summary
Add a condition to represent whether the clock of hub and agents are synced.
Related issue(s)
Fixes #