I am running Mailcow Dockerized on a VPS.
Its a KVM with 4 dedicated cores, 12 gb ram and SSD storage. the server runs super fast (and I haven't noticed any performance drops at all).
Everything was fine but then I did my first update. since then, the server crashes. the whole vps isnt accessible anymore and via vnc i can only see a black screen. sometimes after an hour or so sometimes it runs for a few more hours.
has this something to do with docker and the fact that it runs on a vps? If so, i would consider renting a dedicated server or maybe another vps? (I am new to this docker stuff.)
After a restart I unlock my container and start the docker deamon. the webui is available but I cant connect to the mailserver. 'docker-compose dovecot-mailcow_1' says 'ERROR: No such service: dovecot-mailcow_1' without a service name and grep I am able to view the logsfiles.
EDIT: Now I know, that the service names are different to the names showing in 'docker ps'.
And I know, why I could not connect to the dovecot. there was still the master.pid in /var/run/dovecot.
After starting dovecot I could watch my server crash. I started the dovecot-container after deleting the master.pid inside of it, tested if I could connect with my mailclient. It worked. a minute later -> kernel panic. coincidence? This is everything I can see trough vnc.
EDIT2: The crash was not caused by dovecot. because now it runs just fine. for now. I need to know why it keeps crashing and where to start analyzing the problem. can a kernel panic be caused by the host-system?
(it runs for 3 hours now but i'll bet tomorrow it crashed again.)
I have to admit, this is my first serious kernel panic and the kernel is not the only one having a "panic attack" right now