Grafana dashboard shows cluster degraded but it doesn't seem so. #4487
Unanswered
alessandroros
asked this question in
Q&A
Replies: 1 comment
-
I've also stumbled upon this, for the Replication status, its consist of this query:
which according to the docs
and then the threshold I suspect the team consider the cluster healthy if it has 2 replica or more (1 primary, 2 standby minimum) |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello,
I have a question.
Everything was working properly, so I tried to do a "failover test" :). After testing the failover I also "downsized" the cluster from 2 instance to 1 and viceversa for some times.
Right now I have 2 isntances 1 called prod-db-1 and the other called prod-db-6.
Here the status of my cluster
Here my grafana dashboard
If I look for the variables used by those 2 tiles I see:
Here it seems like db-6 is in recovery.
Here I see that the wal reciever is not up on the db-6, but I'm not sure if it's ok or not.
Do you see any issues?
Thanks for anyone which could clarify something on these topics where I'm not totally familiar.
If I try to connect to DB6 (read only service) everything is fine, even if I retrigger a failover everything is ok (indipendently from the instance I'm killing). Any thoughts?
Alessandro
Beta Was this translation helpful? Give feedback.
All reactions