[DRBD-user] Parallel resource startup, scalability questions

Christian Balzer chibi at gol.com
Tue Jul 2 10:08:30 CEST 2013

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


not purely a DRBD issue, but it sort of touches most of the bases, so here

I'm looking at deploying a small (as things go these days) cluster (2
machines) for about 30 KVM guests, with DRBD providing the storage.

My main concern here is how long it would take to fail-over (restart) all
the guests if a node goes down. From what I gathered none of the things
listed below do anything in terms of parallelism when it comes to starting
up resources, even if the HW (I/O system) could handle it.

I've briefly looked (installed and tested) Proxmox, but the 2.6.32 kernel
really puts me off (for various reasons I'd prefer at least a 3.4 kernel).
While other people have asked that question before, I couldn't find any
answer if Proxmox would work with another kernel, provided of course that
one doesn't need OpenVZ support. DRBD is "supported", but that really
means setting it up manually and w/o any real integration into the
clustering they use.

I stared at the ganeti docs for 3 days and while it seems to be a more
complete system the fact that the stable release doesn't support DRBD 8.4
is a show stopper of sorts. Also this beast is very much in need of a GUI.

Lastly I could go with Pacemaker, as I've done in the past for much
simpler clusters, but I really wonder how long starting up those resources
will take. If I forgo live-migration I guess I could just do one DRBD
backing resource for all the LVMs. But still, firing up 30 guests in
sequence will take considerable time, likely more than I would consider
really a "HA failover" level of quality.

Any inputs and thoughts welcome.

Christian Balzer        Network/Systems Engineer                
chibi at gol.com   	Global OnLine Japan/Fusion Communications

More information about the drbd-user mailing list