Hi,
I have a cluster of 3 nodes on 3 different host in docker containers.
I have a stream (+2) with leader and two replicas.
When I restart one broker (docker restart) it takes several minutes, hours to reach back +2.
In config I have the properties set:
stream.advertised_host=node host name
stream.advertised_port=docker exposed port
5552
On the network FW I can see RabbitMQ trying to connect to docker IP addresses and strange ports like 58270, 6101, 6182 ...
In the log I get:
"level":"warning","msg":"rabbit_stream_coordinator: Error while starting replica for...
{{{case_clause,{error,{connection_refused,{child,...}}}},[{osiris_replica,init,1,[{...}|...]},
...
Am I missing something?
Regards,
BK