Crash on Docker/Kubernetes about rabbitmq-message-deduplication HOT 4 CLOSED

Hello and thanks for reporting this. Contrary to what you think, it seems indeed related to the plugin.

In both cases, it seems the crash appears when the broker queries the information about the exchange via the info callback. One is a periodic call and the other might be initiated by the MGMT WebUI.

The error seems to be due to the fact that the deduplication cache was not initialized at the moment of the check, hence the noproc error.

Is the bulk_uptime_checks_exchange existing? Was it correctly initialized or did you get an error already when creating it?

from rabbitmq-message-deduplication.

This and issue #30 are related. I am aware of the root cause and I am discussing about alternative solutions on the rabbitmq-users mailing list.

from rabbitmq-message-deduplication.

Hey, apologies for not responding to initial one. I had to remove the plugin for now and rely on client to de-dupe (or, in my case, ensure I only send one message).

Great to hear you've found the cause! I'll give it another shot when a fix rolls out.

from rabbitmq-message-deduplication.

Issue fixed in release 0.4.0.

Closing this ticket, please re-open it if the problem persists.

from rabbitmq-message-deduplication.