alertad.conf settings not taken into account (same settings working if put inside the alerta.conf file)

[AndreaLVR]

Hi,
I am facing a problem with some settings in the alertad.conf file. I state that I'm using Alerta inside a docker container, but I doubt that is the source of the problem.
What I am trying to do is simply preventing Alerta from automatically deleting the alerts after a certain period of time.
Below are the relevant environment variables and the contents of /app/alertad.conf, /app/alerta.conf and /web/config.json.

• environment variables:

ALERTA_SVR_CONF_FILE=/app/alertad.conf
ALERTA_CONF_FILE=/app/alerta.conf

• /app/alertad.conf:

DEFAULT_EXPIRED_DELETE_HRS = 0
DEFAULT_INFO_DELETE_HRS = 0

• /app/alerta.conf:

[DEFAULT]
endpoint = http://localhost:8080/api

• /web/config.json:

{"endpoint": "http://localhost:8080/api"}

Despite the settings in the /app/alertad.conf file, the alerts are cleared shortly after their insertion, but the interesting thing is that if I copy the exact content of /app/alertad.conf into /app/alerta.conf then everything works as expected and the alerts are no longer deleted.

Is there something that I am wrong or that I misunderstood? Thank you in advance.

[satterly]

You don't say what version you are running.

[satterly]

This issue is blocked waiting on more information. The description or subsequent comments do not provide enough information to triage, investigate or resolve the issue. Please review the description against the issue template and ensure all relevant information is included. If you do not know what is expected you can ask on Slack.

[CWollinger]

Hi,

I have the same problem. I don't want to delete closed alarms.
For this I set DEFAULT_EXPIRED_DELETE_HRS = 0 in the alertad.conf - with no effect.

I can only find closed alarms not older than 2 hours.
Using alerta 8.5.0.

Regards
CWollinger

[CWollinger]

Found a old issue #1493

Will try the parameter DELETE_EXPIRED_AFTER = 0 and come back with a comment.

[AndreaLVR]

Sorry, I am using this version:

[CWollinger]

The parameter DELETE_EXPIRED_AFTER = 0 or DELETE_EXPIRED_AFTER = 240 dosn't work, too

[CWollinger]

I found the solution in the docker-alerta repo/commit: alerta/docker-alerta@9aa906c

Supervisord for the housekeeping is running in the docker container. The housekeeping reference the var HK_EXPIRED_DELETE_HRS & HK_INFO_DELETE_HRS which is set to 2 and 12 hours in the Dockerfile.

To change this you have to set/override the var in your docker environment e.g. docker-compose file.

The other variables in alertad.conf are not suitable/used for docker.
Maybe a lack of documentation here.

@AndreaLVR hope this will help you

[AndreaLVR]

Thank you so much, @CWollinger.

[hugoShaka]

Hello @satterly

I believe this is a regression introduced by #1508.

Passing DEFAULT_EXPIRED_DELETE_HRS and DEFAULT_INFO_DELETE_HRS in alertad.conf is ignored since 8.6.0. I was running with 1 week of retention and upgrading from 8.4.1 to 8.6.0 triggered an aggressive housekeeping flushing almost all of the alerts.

In the following piece of code the backward compatibility is kept only for environment variables. get_config() is never called on DEFAULT_EXPIRED_DELETE_HRS, thus alertad.conf is ignored for those two parameters.

alerta/alerta/utils/config.py

Lines 79 to 84 in 5059351

	delete_expired_hrs = (
	os.environ.get('DEFAULT_EXPIRED_DELETE_HRS', None)
	or os.environ.get('HK_EXPIRED_DELETE_HRS', None)
	)
	delete_expired = delete_expired_hrs * 60 * 60 if delete_expired_hrs else None
	config['DELETE_EXPIRED_AFTER'] = get_config('DELETE_EXPIRED_AFTER', default=delete_expired, type=int, config=config)