Constant watchdog alerts on rspamd

hlippke

I for a couple of hours I’m receiving watchdog notifications concerning rspamd with multiple lines of
Rspamd settings check failed
every 5-10 minutes.

I can’t find which setting might be a problem, the watchdog script that emits the message doesn’t clarify for me.

Any ideas?

diekuh

You need to post the settings map somewhere. You can find it here:

image.png

I can try to replicate it with an 1:1 copy of your settings map.

Else: Set debug_modules = ["settings"]; in data/conf/rspamd/override.d/logging.inc and restart Rspamd. Check and post docker-compose logs --tail=200 -f rspamd-mailcow logs. The logs must contain the moment the settings map failed.

diekuh

It can also be a broken firewall (ufw, firewalld, etc.). But I doubt it.

hlippke

It appeared around 24hours after first startup, but closely afterwards I experienced a complete systems failure, rendering the whole machine useless so I had to reset everything. Good thing the error’s gone, bad thing I can’t help with the analysis as all the logs and configs were gone. I was lucky I had a backup of my vmail volume so was able to restore the important bits.

Grissom

Hi,
I do have the very same problem. The moment I received another email, I checked the logs as per above and the following showed up:

rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_term_handler: catch termination signal, waiting for 5 children for 16.00 seconds
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_check_termination_clause: hs_helper process 49 terminated normally
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_cld_handler: do not respawn process hs_helper after found terminated process with pid 49
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_check_termination_clause: fuzzy process 45 terminated normally
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_cld_handler: do not respawn process fuzzy after found terminated process with pid 45
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_check_termination_clause: rspamd_proxy process 46 terminated normally
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_cld_handler: do not respawn process rspamd_proxy after found terminated process with pid 46
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_check_termination_clause: controller process 47 terminated normally
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_cld_handler: do not respawn process controller after found terminated process with pid 47
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_check_termination_clause: normal process 48 terminated normally
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; rspamd_cld_handler: do not respawn process normal after found terminated process with pid 48
rspamd-mailcow_1 | 2021-01-30 11:48:23 #1(main) <9aaf69>; main; main: terminating…
mailcowdockerized_rspamd-mailcow_1 exited with code 0

Any clue?