Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Logfilesize Ballooning #253

Closed
ich777 opened this issue Jan 30, 2021 · 16 comments
Closed

Logfilesize Ballooning #253

ich777 opened this issue Jan 30, 2021 · 16 comments
Assignees
Labels
Bug Something isn't working High Priority This issue or pull request is high priority and requires immediate attention Question Further information is requested

Comments

@ich777
Copy link

ich777 commented Jan 30, 2021

Currently I'm experiencing a huge problem with MultiAdmin.

The server writes a huge log file after 1 or 2 minutes and it's getting bigger and bigger...

grafik

I run MultiAdmin with Mono in a Debian Container for Unraid, please look at the original post here:
https://forums.unraid.net/topic/79530-support-ich777-gameserver-dockers/?do=findComment&comment=941528

This is a completely fresh install with no mods or anything installed, even tried to disable logging but that was completely ignored... this was my fault now it's creating no log file.
grafik

After writing this message the log has grown again:
grafik

EDIT: Here is the output of the log before it balloons:
grafik

EDIT2: I implemented a quick and dirty fix that disables the logfile for now.

@ButterscotchV ButterscotchV added Bug Something isn't working Question Further information is requested labels Feb 5, 2021
@ButterscotchV
Copy link
Collaborator

I'm not exactly sure what's going on here? I assume it has to do with how Unraid reads log files, I might look into this but this would be extra work to support it specifically.

@ich777
Copy link
Author

ich777 commented Feb 5, 2021

Why should this be Unraid related?
I forgot to say that I run this in a Docker container that I've created for SCP/Multiadmin: Click

It's based on Debian Buster, I don't think it's realted to Unraid because I've never had a problem with the container, the user that reported the problem also told me that this was never a problem and it began about one or two weeks ago.
I will test it later if this happens also on my Debian Buster machine.

Has anything changed in the past month or so in terms of the log?

EDIT: Forgot to say, thank you for the response @Dankrushen!

@ButterscotchV
Copy link
Collaborator

I changed the logging slightly, now instead of re-opening a handle for each log individually, it keeps one handle open the whole time and flushes after each message. That should result in essentially the same thing.

I'm assuming it's related to how you're reading the file that is causing the issue?

@ButterscotchV ButterscotchV self-assigned this Feb 5, 2021
@ich777
Copy link
Author

ich777 commented Feb 5, 2021

I don't read the file, the first screenshot is from the output of the console.

I think it outputs 'unexpected EOF' because it's getting this big (also the last screenshot is from what the log looks like if you open up the file right after the start of the container and nothing more is displayed, after that it begins to balloon).

This is really weired, have to try it on a native Debian Machine...

@ButterscotchV
Copy link
Collaborator

Where is the error coming from, though? That doesn't look like a MultiAdmin error to me unless I'm not remembering correctly... I'll look and see if that's anywhere in MultiAdmin, but I don't think it is.

@ButterscotchV
Copy link
Collaborator

It looks like it's coming from Docker itself moby/moby#35370

I'm not quite sure what Docker is expecting, if you have any idea, then I would be open to hearing it, but for now I'll look into it a bit myself...

@ButterscotchV ButterscotchV added the High Priority This issue or pull request is high priority and requires immediate attention label Feb 5, 2021
@ich777
Copy link
Author

ich777 commented Feb 5, 2021

@Dankrushen tried it now an a native Debian Machine with the same result the logfile is ballooning again and I can't even type something in the console and is completely unresponsive...

Here is a screenshot from the file when I open it, right after the start of the Server:
1

Here is a screenshot from the size after about 1 minute of runtime:
2

And here is a screenshot from the console, again, completely unresponsive, CTRL+C works after about 30 seconds:
3

@ich777
Copy link
Author

ich777 commented Feb 5, 2021

The server runs but it balloons the logfilesize completely and about after 10 minutes it's about 10GB in size.

@ich777
Copy link
Author

ich777 commented Feb 5, 2021

Can it be that this is also related to #254 ?

@ButterscotchV
Copy link
Collaborator

You're running an outdated version of MultiAdmin, I'm only taking issues for MultiAdmin v3.3.1.0 and higher

@ich777
Copy link
Author

ich777 commented Feb 5, 2021

@Dankrushen these are marked as 'pre releases' and my container grabs only always the 'latest' release from your repo but I could update too to see if it solves the issue...

@ButterscotchV
Copy link
Collaborator

They are indeed pre-relases since they are somewhat unstable (as can be seen from the influx of issues). I'm guessing that's most likely your problem. MultiAdmin versions under MultiAdmin v3.3.1.0 won't be able to run the latest version of the game.

@ich777
Copy link
Author

ich777 commented Feb 5, 2021

Yes, this solves the issue why aren't the new releases marked as latest?

Should I change the container script to pull always the latest version regardless if it's the latest or a pre release?

One thing is that the server is indeed running with a older version but the logfile is ballooning.

@ButterscotchV
Copy link
Collaborator

This should hopefully be an uncommon occurrance, since it was the game that changed the message protocol in a way that breaks MultiAdmin. There will be a new latest release once it's stable, for now it's simply a pre-relase.

One thing is that the server is indeed running with a older version but the logfile is ballooning.

What I mean is that the older version of MultiAdmin isn't able to properly read the server's messages and would allocate huge amounts of memory because the number of bytes it expects is incorrect, therefore crashing the console and potentially bloating the log files with corrupt data.

@ich777
Copy link
Author

ich777 commented Feb 5, 2021

Thank you for you assistance @Dankrushen and clarification!
For now I changed the container to always pull the latest available release regardless if it's a pre release or release version.

@ich777 ich777 closed this as completed Feb 5, 2021
@ButterscotchV
Copy link
Collaborator

All good, I should've double checked that you were on the right version.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Bug Something isn't working High Priority This issue or pull request is high priority and requires immediate attention Question Further information is requested
Projects
None yet
Development

No branches or pull requests

2 participants