XI Post Upgrade (5.2.2) Monitoring Engine (nagios) Part II

[larrydhelms]

See my prior post XI https://support.nagios.com/forum/viewto ... 87#p161687 for background...

I did a LOT of debugging/testing. It turns out... that if you set result_worker=(# != 1) in the mod_gearman_neb.conf... the nagios process crashes with the - Caught SIGSEGV, shutting down... message previously pointed out...

So... seems that there's a BUG somewhere.

[ssax]

Thank you for posting this, our developer is aware of this but we're not sure why having result_worker greater than 1 is causing the nagios process to segfault but we do see that it is likely instructions inside the mod_gearman module that is causing it. Unfortunately, mod_gearman is a 3rd party piece of software and we don't really have any control over their code/product. We have been in contact with the mod_gearman developers and they said they haven't see it before. We'll do what we can.

[larrydhelms]

Does the problem exist with Nagios 2014 (aka Version 4) as well or is this limited to Version 5.x?

Is there a work-around (other than - NOT using mod_gearman; cause we have a distributed environment (aka only certain workers can perform certain task due to network connectivity/permissions)?

Is there an ETF (estimated time to fix)?

[ssax]

The problem does exist in 2014, the workaround is to set result_worker to 1 as you found out, we're not sure why it works for some people and not others or why it just starts happening:

result_workers

Number of result worker threads. Usually one is enough. You may increase the value if your result queue is not processed fast enough.

Taken from here:

https://labs.consol.de/nagios/mod-gearman/index.html

I'm unable to give an ETA/ETF, the developers are aware of it though.