clamav-daemon is not running on Heroku

[tomhamiltonstubber]

I deploy to Heroku in a container, using the configuration set out in tc_av/Dockerfile:

tc-virus-checker/tc_av/Dockerfile

Lines 1 to 16 in 58b3468

	FROM heroku/heroku:18.v16
	WORKDIR /root/
	ADD install.sh install.sh
	RUN bash install.sh && rm /bin/sh && ln -s /bin/bash /bin/sh
	ADD ./requirements.txt /home/app/requirements.txt
	RUN pip install -r /home/app/requirements.txt
	ADD ./app /home/root/app
	ADD run.py /home/root/run.py
	ENV PYTHONUNBUFFERED 1
	WORKDIR /home/root
	CMD ["python", "./run.py"]

Now, if I shell in and run service clamav-daemon status, I get:

chown: invalid user: 'clamav'
chown: invalid user: 'clamav'

Starting the service gives me:

~ $ service clamav-daemon start
chown: invalid user: 'clamav'
 * Starting ClamAV daemon clamd           
start-stop-daemon: user 'clamav' not found

It's my understanding that the user clamav gets created when we install clamav-daemon, and indeed, when I edited my Dockerfile to add them, I get an error saying they already exist.

Running freshclam gives me the following output:

Wed Mar 24 18:37:07 2021 -> ClamAV update process started at Wed Mar 24 18:37:07 2021
Wed Mar 24 18:37:07 2021 -> daily database available for download (remote version: 26119)
Time: 2.1s, ETA: 0.0s [=============================>] 100.57MiB/100.57MiB      
Wed Mar 24 18:37:12 2021 -> Testing database: '/var/lib/clamav/tmp.4d8a8/clamav-d121f3eabdb6bdb755e64
d41210f1c14.tmp-daily.cvd' ...
Wed Mar 24 18:38:07 2021 -> Database test passed.
Wed Mar 24 18:38:07 2021 -> daily.cvd updated (version: 26119, sigs: 3965409, f-level: 63, builder: r
aynman)
Wed Mar 24 18:38:07 2021 -> main database available for download (remote version: 59)
Time: 2.4s, ETA: 0.0s [=============================>] 112.40MiB/112.40MiB      
Wed Mar 24 18:38:13 2021 -> Testing database: '/var/lib/clamav/tmp.4d8a8/clamav-7c4a2e48008286c6b8f29
8de65ec0b05.tmp-main.cvd' ...
Wed Mar 24 18:38:21 2021 -> Database test passed.
Wed Mar 24 18:38:21 2021 -> main.cvd updated (version: 59, sigs: 4564902, f-level: 60, builder: sigmg
r)
Wed Mar 24 18:38:21 2021 -> bytecode database available for download (remote version: 333)
Time: 0.0s, ETA: 0.0s [=============================>] 286.79KiB/286.79KiB     
Wed Mar 24 18:38:21 2021 -> Testing database: '/var/lib/clamav/tmp.4d8a8/clamav-ae231a6a35a0e07fc20f3
3b0f0442a21.tmp-bytecode.cvd' ...
Wed Mar 24 18:38:23 2021 -> Database test passed.
Wed Mar 24 18:38:23 2021 -> bytecode.cvd updated (version: 333, sigs: 92, f-level: 63, builder: awill
ia2)
Wed Mar 24 18:38:23 2021 -> ^Clamd was NOT notified: Can't connect to clamd through /var/run/clamav/c
lamd.ctl: No such file or directory

Any help would be appreciated. I have a feeling this is due to how the users are created with Docker, but I'm not sure.

[tomhamiltonstubber]

For the record, I moved away from Docker and deployed with the following buildpacks:

https://buildpack-registry.s3.amazonaws.com/buildpacks/heroku-community/apt.tgz
https://github.com/riskmethods/heroku-buildpack-clamav.git

I then discovered that ClamAV needs 1.2GB of RAM and, seeing this would cost $250 pm, moved to EC2.