Replies: 3 comments 16 replies
-
@derekcollison Could you please help? |
Beta Was this translation helpful? Give feedback.
0 replies
-
do you have at least 4 cpus / GOMAXPROCS setup in the cluster? |
Beta Was this translation helpful? Give feedback.
1 reply
-
Also could you share the stream info from the affected stream? |
Beta Was this translation helpful? Give feedback.
15 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi!
We have the configuration with about 40 streams and about 400 consumers.
Server version: 2.10.20
Using KV storage for some kind of role election mechanism based on key ttl.
Besides the election normal message exchange is in processing.
At some point in time, we see timeout exceptions from the client side and as a result problems with the election mechanism(losing the elected role). This happens at 1 node from 40+.
This one node is some kind of prime node with a dedicated role(more loaded with messages).
We use one shared connection for each node.
And there are my questions:
Beta Was this translation helpful? Give feedback.
All reactions