Setting SHELLHUB_LOG_LEVEL is not working

[DougBR]

We are having a problem with our VM running ShellHub as the logs are taking too much space. So we would like to set the log level to ERROR (or error, as it seems to be right so), but it does not work.

We overwrote the level using the advised:
SHELLHUB_LOG_LEVEL=error in .env.override

we ran make start after deleting the containers. The environment variables do include SHELLHUB_LOG_LEVEL=true, but it still logs all [info] logs.

[gustavosbarreto]

@DougBR Which version of ShellHub are you using?

[DougBR]

v0.10.9

@gustavosbarreto thanks for replying.

[gustavosbarreto]

v0.10.9

@gustavosbarreto thanks for replying.

@DougBR This version is no longer supported. Could you please upgrade to the latest version? The issue has already been fixed.

If you’re unable to upgrade ShellHub right now, you can configure your Docker engine to limit the size of log files as a workaround.

[DougBR]

Thanks, we will update it.

[DougBR]

We updated it, but when using SHELLHUB_ENV=production, it seems it is pulling directly from DockerHub and doesn't consider the code release I am using (v0.20.1) and rather pulls the latest version there (from 2020). Is it possible that this is happening or is it pulling from somewhere else?

I still got the info logs when using the v0.20.1 with the production tag value.

[henrybarreto]

Hi, @DougBR.

We updated it, but when using SHELLHUB_ENV=production, it seems it is pulling directly from DockerHub...

When SHELLHUB_ENV=production we intentionally pull images from Docker Hub using the SHELLHUB_VERSION value. That’s currently by design.

I still got the info logs when using the v0.20.1 with the production tag value.

If you switched only SHELLHUB_ENV to production but didn’t change logging settings, the containers will keep the log level defined in the .env file. To show only errors, modify:

SHELLHUB_LOG_LEVEL=error

Note: the SHELLHUB_LOG_LEVEL change does not affect the gateway and ui containers for now.

[DougBR]

Here is my .env.override:

SHELLHUB_SSH_PORT=[ourport]
SHELLHUB_ENV=production
SHELLHUB_DOMAIN=[ourdomain]
SHELLHUB_VERSION=v0.20.1
SHELLHUB_AUTO_SSL=true
SHELLHUB_REDIRECT_TO_HTTPS=true
SHELLHUB_LOG_LEVEL=error

Here are some logs from the API container:

{"bytes_in":"0","bytes_out":"178","host":"api:8080","id":"","latency":"510115","latency_human":"510.115656ms","level":"info","method":"GET","msg":"","remote_ip":"172.18.0.8","status":200,"time":"2025-10-21T13:32:19Z","uri":"/api/info","user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.6 Safari/605.1.15"}
{"bytes_in":"0","bytes_out":"178","host":"api:8080","id":"","latency":"28600","latency_human":"28.60044ms","level":"info","method":"GET","msg":"","remote_ip":"172.18.0.8","status":200,"time":"2025-10-21T13:32:19Z","uri":"/api/info","user_agent":"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.6 Safari/605.1.15"}
{"bytes_in":"0","bytes_out":"0","host":"api:8080","id":"","latency":"5","latency_human":"5.89µs","level":"info","method":"GET","msg":"","remote_ip":"172.18.0.7","status":200,"time":"2025-10-21T13:32:24Z","uri":"/api/healthcheck","user_agent":"curl/8.14.1"}
{"bytes_in":"0","bytes_out":"0","host":"api:8080","id":"","latency":"6","latency_human":"6.98µs","level":"info","method":"GET","msg":"","remote_ip":"172.18.0.7","status":200,"time":"2025-10-21T13:32:54Z","uri":"/api/healthcheck","user_agent":"curl/8.14.1"}

I expected not to get logs with the level "info" for this version and the log level set to error.

[DougBR]

Now I used the SHELLHUB_VERSION=latest and it worked for api. It seems the issue was still not fixed for the stable release v0.20.1