Ethan Galstad wrote:
> It looks like the error is occurring in the pthread_mutex_unlock()
> function, which is strange. Checking Google resulted in a couple of hits
> that make it sound like a problem in NetBSD's pthread implementation.
It's not an error in NetBSD's pthread implementation, it's an error in
Nagios. But our pthread implementation is a lot stricter than Linux's,
which makes errors apparant that (usually) go unnoticed on Linux systems,
where they are silently ignored.
> Does the error still occur if you set "PTHREAD_DIAGASSERT='A'" before
> starting Nagios up? Here's one article that describes how doing so
> fixed a similar error with gftp under NetBSD:
That's only a workaround, it makes libpthread ignore errors (like under
Linux) instead of aborting on them (NetBSD's default). Yes, Nagios runs
when setting that flag, but that doesn't solve the underlying problem.
Geert
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]