(null) attacks

Support forum for Nagios Core, Nagios Plugins, NCPA, NRPE, NSCA, NDOUtils and more. Engage with the community of users including those using the open source solutions.
grimsniffer
Posts: 9
Joined: Thu Aug 29, 2013 2:44 am

Re: (null) attacks

Post by grimsniffer »

In order:

1. I very much doubt this is a perl issue. 100% of my plugins are perl based, but when I run the same plugins exactly (copy-paste, adjusting for resource/host variables, from the configuration) from the command line while this is happening, all is well.
2. I just checked - no segfaults while this happened.
3. I would definitely consider upgrading if I was absolutely sure this would solve the problem. Otherwise, seeing as this is a production environment, I'd rather not tinker with it too much potentially needlessly.

Regarding IO - there actually could be something there. I set up a monitor for this on the localhost (silly me, I monitor this everywhere else, but not on good ol' 127.0.0.1)... I'll wait for this to happen again, correlate and update here. I remember tinkering with moving some temp data into the RAM and having some trouble with it (mostly GUI trouble, pages not rendering properly, etc.) - but if this is indeed the problem, I'll delve deeper into it.

Thanks for the suggestions! I promise to keep you all posted.
sreinhardt
-fno-stack-protector
Posts: 4366
Joined: Mon Nov 19, 2012 12:10 pm

Re: (null) attacks

Post by sreinhardt »

the only other thing, off the top of my head, would be per user process\thread limits, especially since you can run these via root or another user when you log in. Let's just check the current limits when you have a chance.

Code: Select all

ulimit -a
Nagios-Plugins maintainer exclusively, unless you have other C language bugs with open-source nagios projects, then I am happy to help! Please pm or use other communication to alert me to issues as I no longer track the forum.
grimsniffer
Posts: 9
Joined: Thu Aug 29, 2013 2:44 am

Re: (null) attacks

Post by grimsniffer »

Okay - I think I found the problem.
I say "think", because this is one of those "it makes sense, but I don't have hard-proof yet" things.

I was looking at the kernel thread limit, which is indeed ~140k - however, I did not look at the per-user pending signals limit, which is about 70K - which (as you can see in the graph on the first page of this thread) is (more or less) the number of threads at which I crash.
I doubled that number to 140K, and I sincerely hope it'll do the trick.

Thanks for all the help, you've got a great forum here!
slansing
Posts: 7698
Joined: Mon Apr 23, 2012 4:28 pm
Location: Travelling through time and space...

Re: (null) attacks

Post by slansing »

Great! Let us know what the results are!!
grimsniffer
Posts: 9
Joined: Thu Aug 29, 2013 2:44 am

Re: (null) attacks

Post by grimsniffer »

The better part of a year has passed and there was no occurrence, so I can safely say this solved my problem.

In case someone happens upon this thread, please see this page for the jist of it:
http://www.unixarena.com/2013/12/how-to ... edhat.html
tmcdonald
Posts: 9117
Joined: Mon Sep 23, 2013 8:40 am

Re: (null) attacks

Post by tmcdonald »

Thanks for coming back and letting us know! I remember this thread from my first few weeks here, so it's cool to see it's working out alright now.

I'd like to put this thread to rest, so if you need anything else just let us know.
Former Nagios employee
Locked