Can't get webserver to run (AWS ECS)

On-Premise
1

I’m running sentry on premise and am running into an issue and finding it difficult to debug. I’ve got a container running on ECS but when I go to run the web command it looks like it hangs before the web server begins running. I’ve set the loglevel to “DEBUG” but I get no output past the following lines[0]. I think I’ve made some sort of configuration error but I am unsure where to begin since there is no clear failure that I can trace.

I’m running Elasticache, and RDS. I’ve tested opening a connection to RDS (with psycopg) and that works. I’ve not tested connecting to the redis cluster as I’m not sure how to do so. I’m also using S3 as the filestore backend. The only other real difference I can think of is that I’m using the “sentry-auth-google” plugin.

If there are any command that I can run, or configuration that I can change to get more insight into what the system is hanging on, I would appreciate the help.

[0]: $: sentry run web --loglevel DEBUG
12:56:38 [INFO] sentry.plugins.github: apps-not-configured
12:56:35 [WARNING] sentry.utils.geo: settings.GEOIP_PATH_MMDB not configured.

1 Like
2

So it looks like the sentry run web is not the only command that just hangs. If I try running sentry createuser or sentry upgrade I get the same behavior. The same two lines are printed and the application just hangs indefinitely.

3

I think we’d need more info than this to be able to help you. Those lines should not be blocking anything, especially for createuser or upgrade commands.

4

Having the exact same problem using ECS. Can’t get any meaningful logging. Any updates here?

5

My problem seemed to be database access. I fixed my database routing permissions and the service started right up. Previously it just hung without exiting and without an error message.

2 Likes