[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: emba.gnu.org non-functional
From: |
Michael Albinus |
Subject: |
Re: emba.gnu.org non-functional |
Date: |
Sun, 14 Jul 2024 11:41:50 +0200 |
User-agent: |
Gnus/5.13 (Gnus v5.13) |
Michael Albinus <michael.albinus@gmx.de> writes:
> Hi,
> emba.gnu.org does not cooperate any longer. The web access returns with
> "504 Gateway Time-out".
>
> The machine itself is accessible via ssh. Uptime looks normal.
>
> root@emba:~# uptime
> 12:48:15 up 8 days, 2:17, 1 user, load average: 0.22, 0.17, 0.15
>
>
> Furthermore, it runs out of space:
>
> root@emba:~# df
> Filesystem 1K-blocks Used Available Use% Mounted on
> tmpfs 812848 25388 787460 4% /run
> /dev/mapper/emba.gnu.org-crypt0 188691636 188691352 284 100% /
> tmpfs 4064236 52 4064184 1% /dev/shm
> tmpfs 5120 0 5120 0% /run/lock
> overlay 188691636 188691352 284 100%
> /var/lib/docker/overlay2/258b80131509190e6d09dfa7e11cb146c1d4ce74442c8db60ecb5f3d232c74ed/merged
> tmpfs 812844 0 812844 0% /run/user/0
>
> Could somebody pls check what's up?
I've run the docker prune commands from crontab manually, and this gave
me some room on the disks. However, the used space is increasing, again,
and I have no idea why the entries in crontab are not sufficient in a
longer run.
Furthermore, we still have the problem that gitlab jobs fail due to
authentication timeout. One datapoint is that there are many ufw entries
in /var/log/messages for blocked connections. And you can see them added
in live by calling
--8<---------------cut here---------------start------------->8---
root@emba:~# tail -f /var/log/messages
--8<---------------cut here---------------end--------------->8---
and then, in parallel, accessing <https://emba.gnu.org/emacs/emacs/-/pipelines>.
Could it be, that the firewall has some undesired deny or limit rules
for the gitlab instance? Note, that I have closed to nothing knowledge
about firewalls.
Could we try an experiment, and disable the firewall for a short time,
and see whether gitlab is responding as expected then?
Best regards, Michael.