This message was deleted.

Hmm not something super trivial I’m aware of. But I think you can use opal’s own health-check at port 7000 (

/

or

/healthcheck

or

/healthy

) - this has the same meaning and doesn’t require setting

OPAL_OPA_HEALTH_CHECK_POLICY_ENABLED

(When off, you’ll have few logs less because OPAL won’t keep writing its status to OPA). You still gonna have logs for each health probe, but these would be Gunicorn logs. Maybe you can play with

gunicorn_conf.py

to hide them - as detailed here

Interesting - that does reduce the amount of data logged, but there's still a non-gunicorn log line:

2023-06-20T14:08:10.419438+0000 | 7 | opal_client.policy_store.opa_client     | INFO  | OPA client health: True (policy: True, data: True)
2023-06-20T14:08:10.419639+0000 | 7 | uvicorn.protocols.http.httptools_impl   | INFO  | 10.244.0.5:43432 - "GET /healthy HTTP/1.1" 200

This has a similar issue in that I don't want to filter out all

opal_client.policy_store.opa_client

logs

Oh - the first line we can possibly change to DEBUG (or only print it when status changes). The second one might be filtered out by filtering gunicorn logs (my previous msg). Or you can override the container’s command line and add

… | grep -v "health"

? 😆