[linux-lvm] Linux LVM snmp MIB.

[linux-lvm] Linux LVM snmp MIB.

[Ceri Storey]

Eyup.

My company is looking at doing per logical volume I/O statistics 
gathering via SNMP for our hosting platform, and was wondering if there 
was any pre-defined MIB for this purpose?

I've found 
https://www.redhat.com/archives/linux-lvm/2001-November/msg00140.html 
which references 
http://people.debian.org/~wakkerma/iana/oid.html#asgn-debian.project , 
but that doesn't detail how the sub-tree is laid out. Has anyone defined 
an "official" MIB for this, or are people just using ad-hoc structures 
for this?

Cheers.
-- 
ceri.storey@dsvr.co.uk (Ceri Storey)
Systems Administrator

[linux-lvm] Re: Linux LVM snmp MIB.

[Russell Coker]

On Monday 25 July 2005 19:38, Ceri Storey <ceri.storey@dsvr.co.uk> wrote:
> My company is looking at doing per logical volume I/O statistics
> gathering via SNMP for our hosting platform, and was wondering if there
> was any pre-defined MIB for this purpose?
>
> I've found
> https://www.redhat.com/archives/linux-lvm/2001-November/msg00140.html
> which references
> http://people.debian.org/~wakkerma/iana/oid.html#asgn-debian.project ,

It's been ages since I looked at that.  Nothing that I devised became 
official.  It would be best if someone else took over that, I'm sure that 
Wichert would be happy to delegate it to someone else instead of me.

Any volunteers?  Ceri, maybe you should do this.

-- 
http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/    Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page

[linux-lvm] Re: Linux LVM snmp MIB.

[Ceri Storey]

Russell Coker wrote:
> On Monday 25 July 2005 19:38, Ceri Storey <ceri.storey@dsvr.co.uk> wrote:

> It's been ages since I looked at that.  Nothing that I devised became 
> official.  It would be best if someone else took over that, I'm sure that 
> Wichert would be happy to delegate it to someone else instead of me.
> 
> Any volunteers?  Ceri, maybe you should do this.

Fair enough; although I'd welcome feedback. Also, I only really plan to
do monitoring of logical volumes for the moment.

Here's a preliminary suggestion (LVM standing for 1.3.6.1.4.1.9586.100.1)
NAME
LVM.physicalVolume			LVM.1
LVM.volumeGroup				LVM.2
LVM.volumeGroup.groupName		LVM.2.1
LVM.volumeGroup.extentSize		LVM.2.2		The size of each extent in bytes
LVM.volumeGroup.extentCount		LVM.2.3
LVM.volumeGroup.extentFree		LVM.2.4
LVM.volumeGroup.writable		LVM.2.5
LVM.volumeGroup.resizeable		LVM.2.6
LVM.volumeGroup.exported		LVM.2.7
LVM.volumeGroup.partial			LVM.2.8
LVM.logicalVolume			LVM.3		A table of devices
LVM.logicalVolume.volumeName		LVM.3.1		
LVM.logicalVolume.volumeGroup		LVM.3.2		
LVM.logicalVolume.extentCount		LVM.3.3		the number of extents in this volume
LVM.logicalVolume.reads			LVM.3.4		the number of extents in this volume
LVM.logicalVolume.writes		LVM.3.5		the number of extents in this volume

Comments? Suggestions?
-- 
ceri.storey@dsvr.co.uk (Ceri Storey)
Systems Administrator

[linux-lvm] Re: Linux LVM snmp MIB.

[Russell Coker]

On Wednesday 27 July 2005 20:26, Ceri Storey <ceri.storey@dsvr.co.uk> wrote:
> Here's a preliminary suggestion (LVM standing for 1.3.6.1.4.1.9586.100.1)
> NAME
> LVM.physicalVolume			LVM.1
> LVM.volumeGroup				LVM.2
> LVM.volumeGroup.groupName		LVM.2.1
> LVM.volumeGroup.extentSize		LVM.2.2		The size of each extent in bytes
> LVM.volumeGroup.extentCount		LVM.2.3
> LVM.volumeGroup.extentFree		LVM.2.4
> LVM.volumeGroup.writable		LVM.2.5
> LVM.volumeGroup.resizeable		LVM.2.6
> LVM.volumeGroup.exported		LVM.2.7
> LVM.volumeGroup.partial			LVM.2.8
> LVM.logicalVolume			LVM.3		A table of devices
> LVM.logicalVolume.volumeName		LVM.3.1
> LVM.logicalVolume.volumeGroup		LVM.3.2
> LVM.logicalVolume.extentCount		LVM.3.3		the number of extents in this
> volume LVM.logicalVolume.reads			LVM.3.4		the number of extents in this
> volume LVM.logicalVolume.writes		LVM.3.5		the number of extents in this
> volume
>
> Comments? Suggestions?

I think it would be nicer to have the LV entries under the VG trees.  That's 
the way things are done conceptually.

Also as PVs are assigned to a VG then maybe we should have the PV entries for 
assigned PVs under the VG entries as well and just have a separate tree for 
unassigned PVs.

But your design will work.

-- 
http://www.coker.com.au/selinux/   My NSA Security Enhanced Linux packages
http://www.coker.com.au/bonnie++/  Bonnie++ hard drive benchmark
http://www.coker.com.au/postal/    Postal SMTP/POP benchmark
http://www.coker.com.au/~russell/  My home page