[CentOS] lvm monitoring

Mon Dec 7 18:25:34 UTC 2009
Karanbir Singh <mail-lists at karan.org>


At the moment, I am monitoring my lvm resources on a machine with the 
usual 'pvs;vgs;lvs' output and comparing with whatever is a known good 
state. But this has a few problems:

1- I need to setup a known good state per machine, which can look quite 
different from each other : specially when machines are getting dynamic 
vm builds and teardowns dozens of times per day, or when disk resource 
allocations are spread over more than one machine.

2- Its hard to work out what is really going on with bits that are not 
mounted ( like snapshots ).

3- its hard to tie down to physical media ( I screen scrape at the 
moment, which is dirty and exceptionally prone to nose! )

I am sure there are plenty of people who are monitoring lvm setups on 
their machines, just wondering what it is that you guys are doing.

- KB