Fix OOM (Out Of Memory) Errors

[mjnaderi]

Summary

This PR fixes mattermost/mattermost#20625.

Ticket Link

None

[mvitale1989]

Thank you for your contribution @mjnaderi !
I don't have an objection to this change myself, but I don't think this fixes mattermost/mattermost#20625 , since the reporter mentions using Openshift, so they are unlikely to be using docker compose.
I would include @agnivade as a reviewer, to double check if there are conditions under which such a spike in PIDs in the Mattermost container is expected.

[agnivade]

There's no need to change the limit for postgres.

[agnivade]

More than 200 processes means that something is very wrong with the system. I'd like to get a list of the processes if possible and see if there's an actual bug from the product that needs to be fixed. Otherwise, there's a chance that we are hiding something under the rug here.

[mvitale1989]

Thanks for the input Agniva! Agreed, if more than 200 PIDs is an unreasonable number then we'd better understand what that is first before thinking about merging this change.
I'll remove my approval for the time being to reduce confusion

We first need to understand why there is a spike in PIDs

[erfantkerfan]

I had the same problem and changing this values fixed it.
Like you said, I pushed the problem under the rug by increasing the limit.

[agnivade]

Did you try listing out the processes to see what was wrong?

[erfantkerfan]

Did you try listing out the processes to see what was wrong?

No, but since then, I did not have any crashes, so I did not have the chance to