Am 17.04.21 um 07:55 schrieb Stefan Botter:
Hi munix9,
Am Freitag, den 16.04.2021, 13:02 +0200 schrieb munix9:
The annoying human monitor reports again :-\
Server Status i586 & x86_64 dead
https://pmbs.links2linux.de/monitor
thx
I have rebooted the server and extended the swap partition.
The schedulers died due to an OOM-condition. I am not really sure, where
that comes from, as the processes run smooth for an extended period of
time, and suddenly request RAM like mad. Probably a bug in the scheduler
software.
Unfortunately there is no (at least not known to me) possibility to
monitor the processes from checkmk, where I could build automation for
restarting the missing processes. Also, as all scheduler processes for
all architectures are started with one systemd service, systemd is not
able to watch the processes and possibly restart them.
At least at the moment I am not able to read mails during the day and
have to postpone any maintenance to the weekends.
Greetings,
Stefan
Hi Stefan,
thanks again and again for your effort.
I keep an eye on monitoring in between, at the latest when there are
major TW updates and zypper grumbles about problems resolving packages.
I dimly remember using monitoring at some point somewhere for some
project that checked for differences on a site - similar to
https://visualping.io/
Maybe this is something usable (e.g. if you draw a frame just above the
service blocks on the status page and then let it monitor them, an info
mail should go out if an icon changes from "running" to "dead").
ciao,
Paolo
_______________________________________________
Packman mailing list
[email protected]
https://lists.links2linux.de/cgi-bin/mailman/listinfo/packman
_______________________________________________
Packman mailing list
[email protected]
https://lists.links2linux.de/cgi-bin/mailman/listinfo/packman
