[Linux-cluster] DRBD with GFS applicable for this scenario?

[Zaeem Arshad]

On Thu, Jan 28, 2010 at 3:44 PM, Gordan Bobic <gordan at bobich.net> wrote:

You have made interesting observations. Let me see if I can answer these.

> Hmm... What sort of ping time do you get? I presume you have established
> that it is on the sensible side.

It's 7 ms to be precise. I was assuming the distance to be around 35km
but it seems less.

>
> In terms of performance you will need to make sure that machines tend to
> access only their own sub-paths on the file system (e.g. spool/1 and
> spool/2, and server 1 doesn't touch spool/2 until server 2 goes down).
> Otherwise the performance is going to be attrocious since file locks will
> end up bouncing between the machines. These normally live in cache on a
> conventional file system so if they have to start getting exchanged at most
> accesses you are looking at a latency degradation from ~ 50ns down to some
> milliseconds. If your connectivity is VERY good, if it's 35km I would be
> surprised if your latencies are better than 10ms, which you'll feel even on
> the disk latency, let along memory latency - we are talking 200,000x slower
> in the best case scenario.
>

I think this is how DRBD works. It will usually read/write to the
local device first and in case of local device failure will it switch
over to the other block over IP. I'd be surprised if it doesn't do
that.


> So you propose to have a quorum disk on site 2? OK, that works. The problem
> is that fencing works by one server fencing another, not itself. So you'll
> still need a reliable OOB fencing mechanism such as the one I described.
>
>

Yeah. I can probably use the following heuristics to make sure I fall
back to primary

Heuristics:
Check if QD is accessible at SAN2           5
yahoo check                                           3
POPXYZ check                                       2
Volume X visible on SAN1 by Mail1           4

By using these heuristics and associated scores, I am able to decide
which server has quorum. I am also looking into the OOB fencing idea
too. One question that pops in my mind is that while I will have only
2 nodes in the DRBD volume, will more than 2 cluster members be able
to access the GFS volume on top of the DRBD volume?


Regards

--
Zaeem