You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
From usability perspective, we need a CLI which user can issue to query to know history of failover/ failback with timestamp and its status(success/failure) along with entire sequence of operations and its status during failover/ failback
As of now there is no go-to place for user to look for present status and past occurrences of HA and which would make very difficult for user to understand present state of system if it would have failed over overnight/ in weekend.
Also recording this to a journal is best to debug. No ceph status detects this as of now, except searching along huge logs which is difficult as at #545, there is no other way user can be alerted on this
Further, alerting the same on dashboard would be more beneficial
Output of CLI or logs should look like below-(basically it will be good if we include the sequence and its status) Few of below parameters can be marked optional upon discussion
**Operation- Failover
Status: Success/ Failure**
Time of occurrence - dd/mm/yyyy hrs:min:sec
Source Gateway - <GW_NAME>
Destination Gateway - <GW_NAME>
output of ceph nvme-gw show nvmeof ''
Blocklisting - success/Failure
New monmap boradcast
Time of End - dd/mm/yyyy hrs:min:sec
**Operation- Failback
Status: Success/ Failure**
Time of occurrence - dd/mm/yyyy hrs:min:sec
Source Gateway - <GW_NAME>
Destination Gateway - <GW_NAME>
output of ceph nvme-gw show nvmeof ''
Blocklisting - success/Failure
New monmap boradcast
Time of End - dd/mm/yyyy hrs:min:sec
and / or
all above information should be recorded a journal to keep track off and an user can refer at any time.
Open for discussion.
The text was updated successfully, but these errors were encountered:
From usability perspective, we need a CLI which user can issue to query to know history of failover/ failback with timestamp and its status(success/failure) along with entire sequence of operations and its status during failover/ failback
As of now there is no go-to place for user to look for present status and past occurrences of HA and which would make very difficult for user to understand present state of system if it would have failed over overnight/ in weekend.
Also recording this to a journal is best to debug. No ceph status detects this as of now, except searching along huge logs which is difficult as at #545, there is no other way user can be alerted on this
Further, alerting the same on dashboard would be more beneficial
Output of CLI or logs should look like below-(basically it will be good if we include the sequence and its status) Few of below parameters can be marked optional upon discussion
and / or
all above information should be recorded a journal to keep track off and an user can refer at any time.
Open for discussion.
The text was updated successfully, but these errors were encountered: