File system monitoring limited to Volume Group and/or Logical Volumes
To start with I'm very new to Zabbix, as we're still in eval mode, but liking it a lot so far. I have managed Nagios for several years at 2 locations, so the way Zabbix works threw me for a bit. I've been an Admin/Engineer since 1994, so not new at this, just new at Zabbix.
I tried searching a few times, and may have messed that up. New here, and not looking for others to do homework for me, but I've hit a wall.
if you know of the location of docs that would help just post a link, I'll be glad to dig through 'em.
What I need to do is Monitor some PowerHA System Mirror Servers, There are some file systems that are only active on the active node of the cluster, and there are only a few file systems that remain on both boxes. The problem is the name(s) change depending on the application hosted on the server.
the way I handled this in Nagios was to exclude any Logical volume that ended in _lv (as that would indicate a Shared File system). However, that can become tedious, as would listing file system names explicitly for every different application, hence the filter on Volume Group name, and/or Logical Volumes.
So my question is this, in Zabbix, which would be the most efficient method to handle this, and what is possible?
My thought was to have a base OS template that monitored only rootvg file systems (whatever the combo of LV or file systems it included), and then have templates for each application that would specify which vg's or lv's to monitor on top of that. For the VIP node name it would JUST monitor the app vgs, for the base nodes JUST the root vg.
My problem is that I haven't found a way to "exclude" or explicitly include the vgs or lvs.
So, any help, doc pointers, etc. would be greatly appreciated.
I tried searching a few times, and may have messed that up. New here, and not looking for others to do homework for me, but I've hit a wall.
if you know of the location of docs that would help just post a link, I'll be glad to dig through 'em.
What I need to do is Monitor some PowerHA System Mirror Servers, There are some file systems that are only active on the active node of the cluster, and there are only a few file systems that remain on both boxes. The problem is the name(s) change depending on the application hosted on the server.
the way I handled this in Nagios was to exclude any Logical volume that ended in _lv (as that would indicate a Shared File system). However, that can become tedious, as would listing file system names explicitly for every different application, hence the filter on Volume Group name, and/or Logical Volumes.
So my question is this, in Zabbix, which would be the most efficient method to handle this, and what is possible?
My thought was to have a base OS template that monitored only rootvg file systems (whatever the combo of LV or file systems it included), and then have templates for each application that would specify which vg's or lv's to monitor on top of that. For the VIP node name it would JUST monitor the app vgs, for the base nodes JUST the root vg.
My problem is that I haven't found a way to "exclude" or explicitly include the vgs or lvs.
So, any help, doc pointers, etc. would be greatly appreciated.
Comment