[DRBD-user] Your thoughts on a DRBD -> LVM2 (not CLVM) stack.

Håkan drbdhberg at wayfinder.com
Fri Feb 16 17:41:27 CET 2007

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


> Linux
> Xen
> CLVM
> /dev/drbd0
> /dev/sda
> Linux

I do it the same way.

> So my question is this: why do I need CLVM? Can I just use LVM2? Although the domUs all run on top of their own LV they do belong to the same volume group (which of course sits on top of drbd because it is the drbd device that is pvcreated) - is this why CLVM is needed? For the locking of volume group meta data?

I guess you could use LVM2 if you create all the LV's you'll ever need,
with only one node running and then afterwards start the other node.
However if you create LV's as the need come for new domU's, you're in a
bit different situation. Perhaps it's possible to create the LV's on one
machine and then always do pvscan or vgscan on the other(s). They have
to be made aware somehow that the LVM metadata has changed, and clvm
takes care of that for me. clvm also enables me to create new lv's on
any node at any time, without thinking three times before.
If you try the LVM2 approach, please report your findings!

br,
/Håkan


> Jim
> 




More information about the drbd-user mailing list