Can't open/parse the config file /etc/clamav/freshclam.confg about docker-clamav HOT 20 CLOSED

Was having the same problem. Looks like the config files (and most files in the alpine/edge directory) have windows line endings which are causing the issue. I've pulled the code, ran dos2unix on the project and then could run the image. However it then fails after because /usr/bin/freshclam -d -c 6 runs before clamd so it exits with the error

^Clamd was NOT notified: Can't connect to clamd through /tmp/clamd.sock: No such file or directory

I "fixed" this by changing the freshclam run to /usr/bin/freshclam -d -c 6 &

from docker-clamav.

Having the exact same issue.

from docker-clamav.

Think upstream was not up-to-date. Should now be fixed.
I'll check it in the evening in detail.

from docker-clamav.

I will check this week. A bit surprising, but I can (sadly) confirm it on my machines as well. @mko-x can you check the files itself. I am pretty sure I had just copied them.
@nootle and @JandreCoetzee do you still have the old image available? can you check a docker run -p 3310:3310 <old image id> cat /etc/clamav/freshclam.conf. Just wondering if the pushed image has another content compared to the sources

from docker-clamav.

I recovered the state and pushed a stable image back to docker hub.

@WhiteBahamut please check that out and tell us what you found (:

from docker-clamav.

from docker-clamav.

Thanks for investigation. Can you check https://github.com/WhiteBahamut/docker-clamav/tree/fix/win-eol
Changed EOL and the background task.

from docker-clamav.

Checked out your branch, reran dos2unix and there were no changes which is good! Ran docker build/run but the run is failing because bootstrap.sh is not executable? Not looked into where your branch is based off but your Dockerfile seems to be missing RUN chmod u+x bootstrap.sh check.sh somehow...

from docker-clamav.

Experienced this issue yesterday but everything is working now 👍

from docker-clamav.

can anyone confirm that the alpine image built with https://github.com/WhiteBahamut/docker-clamav/tree/fix/win-eol runs?
I merged the lastes master into it, changed eol and tried /usr/bin/freshclam -d -c 6 &, yet I get ^Clamd was NOT notified: Can't connect to clamd through /tmp/clamd.sock: No such file or directory
To mae it run I had to apply quite some more changes

• change .sock dir back to default /run/clamav/
• change storage dir /store -> /var/lib/clamav (same as debian incl chown)
• run clamd -F to keep the container alive

But health check fails...

from docker-clamav.

I get that error but it looks like all is well afterwards. The container continues running and I'm able to send files over (tested with an insecure file too).

Pulled the Alpine image from Docker Hub about 4 hours ago - apologies if I'm running the wrong thing!

Output:

[bootstrap] Initial clam DB download.
Tue Oct 13 16:56:20 2020 -> ClamAV update process started at Tue Oct 13 16:56:20 2020
Tue Oct 13 16:56:20 2020 -> daily database available for download (remote version: 25956)
Tue Oct 13 16:56:36 2020 -> Testing database: '/store/tmp.6fc5d/clamav-9549e492c998bf79966512c2ec51538c.tmp-daily.cvd' ...
Tue Oct 13 16:56:49 2020 -> Database test passed.
Tue Oct 13 16:56:49 2020 -> daily.cvd updated (version: 25956, sigs: 4330724, f-level: 63, builder: raynman)
Tue Oct 13 16:56:49 2020 -> main database available for download (remote version: 59)
Tue Oct 13 16:57:05 2020 -> Testing database: '/store/tmp.6fc5d/clamav-86ff6739e45c5cfe4ed3fbeb9023e4e8.tmp-main.cvd' ...
Tue Oct 13 16:57:11 2020 -> Database test passed.
Tue Oct 13 16:57:11 2020 -> main.cvd updated (version: 59, sigs: 4564902, f-level: 60, builder: sigmgr)
Tue Oct 13 16:57:11 2020 -> bytecode database available for download (remote version: 331)
Tue Oct 13 16:57:11 2020 -> Testing database: '/store/tmp.6fc5d/clamav-5af110111df551c38496f508e0ffb2a9.tmp-bytecode.cvd' ...
Tue Oct 13 16:57:11 2020 -> Database test passed.
Tue Oct 13 16:57:11 2020 -> bytecode.cvd updated (version: 331, sigs: 94, f-level: 63, builder: anvilleg)
Tue Oct 13 16:57:11 2020 -> safebrowsing database available for download (remote version: 49191)
Tue Oct 13 16:57:16 2020 -> Testing database: '/store/tmp.6fc5d/clamav-4342bcf0d3e103165a0c556bc8723623.tmp-safebrowsing.cvd' ...
Tue Oct 13 16:57:19 2020 -> Database test passed.
Tue Oct 13 16:57:19 2020 -> safebrowsing.cvd updated (version: 49191, sigs: 2213119, f-level: 63, builder: google)
Tue Oct 13 16:57:19 2020 -> ^Clamd was NOT notified: Can't connect to clamd through /tmp/clamd.sock: No such file or directory
[bootstrap] Schedule freshclam DB updater.
[bootstrap] Run clamav daemon
Tue Oct 13 16:57:47 2020 -> Limits: Global time limit set to 120000 milliseconds.
Tue Oct 13 16:57:47 2020 -> Limits: Global size limit set to 314572800 bytes.
... more limit stuff ...
Tue Oct 13 16:57:47 2020 -> Archive support enabled.
Tue Oct 13 16:57:47 2020 -> AlertExceedsMax heuristic detection disabled.
Tue Oct 13 16:57:47 2020 -> Heuristic alerts enabled.
... more things enabled ...
Tue Oct 13 16:57:47 2020 -> Self checking every 600 seconds.
Tue Oct 13 16:57:47 2020 -> Set stacksize to 1048576
Tue Oct 13 16:58:05 2020 -> instream(172.21.0.5@49568): Win.Test.EICAR_HDB-1 FOUND    <--- test file
Tue Oct 13 17:07:52 2020 -> SelfCheck: Database status OK.

from docker-clamav.

The repo owner rolled dockerhub back to a working version. With my local latest version I get the same result you posted. Can you described how you tester the file? Was it via TCP/3310?
Will check more tomorrow and update my branch.

from docker-clamav.

Ahh gotcha - I'm probably using the rolled-back image 🤦‍♂️

Yeah, file send is TCP/3310 via docker-compose and python-clamd.

from docker-clamav.

from docker-clamav.

Does anyone have a kubernetes.yaml for this? I've been messing around but not sure if I've got it quite right
…
On Tue, 13 Oct 2020, 18:44 Ahmed Al-Jawahiry, @.***> wrote: Ahh gotcha - I'm probably using the rolled-back image 🤦‍♂️ Yeah, file send is TCP/3310 via docker-compose and python-clamd...

You may open a new issue for your question (:

from docker-clamav.

can anyone confirm that the alpine image built with https://github.com/WhiteBahamut/docker-clamav/tree/fix/win-eol runs?
I merged the lastes master into it, changed eol and tried /usr/bin/freshclam -d -c 6 &, yet I get ^Clamd was NOT notified: Can't connect to clamd through /tmp/clamd.sock: No such file or directory
To mae it run I had to apply quite some more changes

* change .sock dir back to default /run/clamav/

* change storage dir /store -> /var/lib/clamav (same as debian incl chown)

* run `clamd -F` to keep the container alive

But health check fails...

It would be interesting why these changes are necessary and healtcheck is still failing.

from docker-clamav.

Does anyone have a kubernetes.yaml for this? I've been messing around but not sure if I've got it quite right

Please open new issue. have parts of it already for my cluster and debian images seems to run. Once this one is solved I can work on my k8s yamls ;-)

@mko-x I dont understand it either. I just tried to get as close as possible to the debian dockerfiles and and read some clamav docs. Without detailed understanding clamd -F seems quite clean to keep the container alive.

from docker-clamav.

Would be creat if you can check the PR and test it out. "works on my machine".
For me it turned out the config file was not loaded correctly. Once I added -c /etc/clamav/clam.conf it worked. TCP open, running in foreground, ...

from docker-clamav.

Update on this: pushed the changes to my dockerhub and tested with my nextcloud. :alpine works :)

from docker-clamav.

Fixed by #77

from docker-clamav.