Cisco Switch check issue

[BanditBBS]

but remember, it doesn't always happen from the gui.

This is the only graphing issue I have. In the event log, this host is the only thing with that error.

[BanditBBS]

Also, why is it looking in /var/lib/mrtg/ instead of /usr/local/nagios/share/perfdata/. In the /var/lib/mrtg, this host is the only item with files.

What am I not understanding here? I am 100% sure this is at least part of my issue as these bandwidth checks are done from two workers(primary nagios server and another worker). What is this check doing that is different than all other checks that return performance data?

EDIT: if I copy the contents of /var/lib/mrtg to my other worker, will that cause any issues with the data? I now see the wizard uses a check that is actually deling specifically with mrtg.

[abrist]

Bandwidth checks are done by mrtg. We have a script on a cron that copies (after some processing) the data from the mrtg files to the rrds in perfdata. Moving the /var/lib/mrtg folder may allow for the mrtg checks to work remotely, but will still stop the perfdata rrds from populating. I am not sure the best course of action here. I will talk to the big bad Scott!

[BanditBBS]

Bandwidth checks are done by mrtg. We have a script on a cron that copies (after some processing) the data from the mrtg files to the rrds in perfdata. Moving the /var/lib/mrtg folder may allow for the mrtg checks to work remotely, but will still stop the perfdata rrds from populating. I am not sure the best course of action here. I will talk to the big bad Scott!

Andy, that sounds dangerously like a "No"

wow...I've started typing a sentence 5 times and deleted it, as I am no sure what to ask/say....hmmmmm

Why on earth is BW graphing handled completely different than all other graphing? is there a logical reason for that?

[BanditBBS]

To continue on my previous post.....I have a solution in mind already if I am not happy with your response

basically, I can use the plugin check_nwc_health for the checks instead of use the wizard. It'll be a little more painful to setup, but as of right now, we are not planning on monitoring every last port, just the uplinks.

[abrist]

I am still thinking and consulting with coworkers over this. Bump this post tomorrow so we see it and I will strive to get you an answer longer than a two letter word (not four letters either, I promise!).

[BanditBBS]

Andy, I'm bumping this as requested. Actually, a day later since I was out yesterday!

[abrist]

To use the wizard, you would probably have to create a netmount so that the worker could update the /var/lib/mrtg rrds. Alternatively, you could look at using one of the other port status/bandwidth monitoring plugins. What was your idea?

[BanditBBS]

To use the wizard, you would probably have to create a netmount so that the worker could update the /var/lib/mrtg rrds. Alternatively, you could look at using one of the other port status/bandwidth monitoring plugins. What was your idea?

That was my idea, using the check_nwc_health plugin instead. It'll be a little more manual than using the wizard, but since you guys wanted to treat bw monitoring special, I guess I gotta deal with it since I do distributed checks.(Partial joking )

I could also make all BW checks just force to run from the nagios server and only distribute the other checks...I'll discuss with my network team and see which they prefer. The check_nwc_health plugin really is a darn good plugin. I use it for quite a bit of my network checks. They also wrote the check_logfiles plugin that I consider the best as well.

I guess we can consider this one closed, as I have options now, just need to decide on best course.

EDIT: Going with the other plugin and the more manual method. Yay, more work for me!