Install been running for 7 years, mostly trouble free. Centos 7 Decided to delete some user accounts and make them into aliases Looks like it worked, except server started to refuse incoming email webhost postfix/error[11235]: E21C043F64FA: to=<[email protected]>, relay=none, delay=4, delays=4/0/0/0.02, dsn=4.4.1, status=deferred (delivery temporarily suspended: connect to 127.0.0.1[127.0.0.1]:10024: Connection refused) Ran around in circles. couldnt understand the problem Restored a backup All good again Woke up today, did the same account delete - guess what?!?! AGAIN, same thing. All I'm deleting is some not really used accounts that should have been made as aliases, and turning them into aliases. but instead I get webhost postfix/error[11235]: E21C043F64FA: to=<[email protected]>, relay=none, delay=4, delays=4/0/0/0.02, dsn=4.4.1, status=deferred (delivery temporarily suspended: connect to 127.0.0.1[127.0.0.1]:10024: Connection refused) I'm so frustrated that I dont understand whats happening.
I guess I assumed its amavis thats angry, but I cant beging to find a log/error/etc that makes any sense (to me) Oct 04 12:31:10 webhost.domain.com systemd[1]: Failed to start Amavis mail content checker. Oct 04 12:31:10 webhost.domain.com systemd[1]: Unit amavisd.service entered failed state. Oct 04 12:31:10 webhost.domain.com systemd[1]: amavisd.service failed. Oct 04 12:31:10 webhost.domain.com systemd[1]: amavisd.service holdoff time over, scheduling restart. Oct 04 12:31:10 webhost.domain.com systemd[1]: Stopped Amavis mail content checker. Oct 04 12:31:10 webhost.domain.com systemd[1]: start request repeated too quickly for amavisd.service Oct 04 12:31:10 webhost.domain.com systemd[1]: Failed to start Amavis mail content checker. Oct 04 12:31:10 webhost.domain.com systemd[1]: Unit amavisd.service entered failed state. Oct 04 12:31:10 webhost.domain.com systemd[1]: amavisd.service failed.
Thats where I started , and I just dont see anything that tells me anything To confirm, I left that file tailing, and restarted amavisd from another session, and I get pretty much zero. I dont understand how (TWICE) deleting some users and creating aliases broke it.
should systemctl restart amavisd dump longs into /var/log/maillog? I see postfix scrolling along (doing nothing) and dovecot working, but nothing from amavis
I don't think that this is the direct reason why it broke. here a few things that you can try from ChatGPT:
Ok, I made progress. I did not think of watching messages, and that had the hint that helped me fix it. So, I think there is a bug, but heres what happened. I was clearing out emails not used, and 1 domain. The messages logs make mention of some missing files in the config of amavis It was referring to files in /var/lib/amavis/dkim (public/private keys) and sure enuf killing that domain removed those files, but not removing them from the config. So as a quick fix I copied the files in question back over, and amavis was able to restart. I guess the script that removes the domain doesnt remove the config entries. *OR* I did that YEARS ago by hand??!?!? Trying to make sure its all working, but I just got 3 emails to pass..
That's quite unlikely. The script removes everything correctly, just verified that on three different systems (Ubuntu, Debain and centOS). You likely added config manually by hand on your system and such config will not get removed by ISPConfig.
I guess thats the downside of having an install running since 2016. I keep upgrading the virt hardware, and just moving the old centos 7 vm over... Thanks for the help. I really need to spend some time to learn more
