Graylog not processing messages


started a thread on github (, got displaced to the community forum.

Im using the latest docker-compose from here example ( the only thing whats different in my setup is that i map the journal to the host machine instead of using a volume. Stacktraces and More debug information you can find in the github thread wont copy it here (I hate redundancy!).

When its getting stuck it helps to delete the journal, recreate the docker containers then it runs again (for a couple of days/hours), until its getting stuck again.

Switched now to a docker volume (like the example says) and it seems running till now, but i can clearly say its the reason only in a couple of days…

I do have my setup similar to the following since release 2.0 - I did never had any problem with the journal getting corrupt.

What kind of filesystem did you have on your host?

you mapping the journal itself as volume…


im using


the host has ext4 in the specific path.

the current cluster io is avg 1k messages/sek atm

it seems i can confirm now thats a bug when mapping a host directory directly to the journal.

can anyone reproduce the issue?

#dontcare but closing issues on github :+1:

maybe just nobody is able to reproduce?

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.