Elastichsearch reindex failed after hours

After upgrading to Zammad 4.1 from 3.* and elastic search from 5.6 to 6.8 I’m unable to rebuild elastic search index. This is driving me crazy because employees cannot get their reports and find the tickets and I’m unable to resolve it.

After running

zammad run rake searchindex:rebuild --trace

rebuild runs for about 2 hours, then suddenly it always stops on what looks like on the same element, after that, elasticsearch says failed to connect. I attached a video of the evidence.

Please, I need help to resolve this. Everything was working before upgrading. Do I need to upgrade now to 7 ? I don’t know what to do.

1 Like

Your Elasticsearch died, you can see that on the connect part that’s failing.
Reduce the heap size (see elastic docs please) or ensure your host doesn’t kill / end ES processes.

Not really something we have power on.

Thanks, I already try that and nothing.

Can I install a new instance of Zammad and export the current database to a new virtual machine? So I can start with a freshly installed Zammad?

Yes that should be no problem at all.
Our documentation might help you on that approach:
https://docs.zammad.org/en/latest/appendix/backup-and-restore.html

This topic was automatically closed 120 days after the last reply. New replies are no longer allowed.