This post has NOT been accepted by the mailing list yet.
This post was updated on .
On a 3-node cluster of rabbitmq 3.6.10, erlang 9.3 queue declare operation takes 2.5 seconds
This is happening only in our prod env and no other env and need some help to figure out what to attribute this to
testing with rabbitmqadmin tried following parameters and they call consistently take ~2.5 seconds
tried durable=false/true, --host=localhost
without running trace, is there any other troubleshooting we can do ?
====update to this post
so after much brainstorming .. found that applying queue-master-locator= client-local OR random speeds up queue declaration to 70ms
it makes some sense that min-masters is more costly than random.. but 2500 vs 70 ms .. something pathological seems to be going on
we were recently able to turn off stats in 3.6.10 version.. is it maybe related to this?