[DRBD-user] DRBD sync rate way low

[Jesse A Dubin]

We finally got it back up.  The resync was a bit of a red herring.  You hit
it right on the nose when you asked about the file system mounting.  When
Our Primary rebooted, it came back up before the Secondary was able to
finish taking control of all the resources and without a properly mounted
file system, so they were fighting back and forth while
resyncing....nothing?  I stopped the Secondary, made sure the FS mounted on
the primary and rebooted it.  The Primary came up fine and has been serving
for a day now.  I just booted up the Secondary, and its now syncing at a
more respectable ~450K/sec for each of the 10 nodes.

I've been told that a STONITH solution may be the answer to our problem
when this happens.

As for your configuration questions, I was told:
"load-only is used because heartbeat handles the determination of which
node
(origin or indigo) is to be primary on the DRBD volume.

Protocol B is used because the DRBD HOW-TO uses protocol B.  Using C will
reduce
performance somewhat."

I'm open to suggestion.

Thank you very much, Lars (and anyone else who may have spent some
brainpower consider it), for working on puzzling this out with me.

Jesse
----





Lars Ellenberg <Lars.Ellenberg at linbit.com>@lists.linbit.com on 04/12/2004
12:12:21 PM

Sent by:    drbd-user-admin at lists.linbit.com


To:    drbd-user at nudlaug.linbit.com
cc:     (bcc: Jesse A Dubin/jdubin1/LSU)

Subject:    Re: [DRBD-user] DRBD sync rate way low

/ 2004-04-12 08:43:27 -0500
\ Jesse A Dubin:
> > > why my data is missing and my server is down....
> >
> > I still don't understand what you mean with "my data is missing"...
> > Please give more detail about
> > what you are doing, and
> > what you expect to happen, and
> > where "real world" does not meet your expectations.
> >
>
> Well, before the reboot, I had data in my node dirs.  After the reboot,
> this information is gone, as well as the whole directory structure.  If
it
> were just a DRBD issue, as you said, I would still have data on the
primary
> node and I would still be serving pages.

The filesystem did not complain upon mount,
but the data is gone?

Doh.
Not DRBDs fault I hope :-/

> I expected that upon the shutdown of the primary, there would be a
failover
> and the Secondary would promote itself (it started to, but I think the
> Primary came back up before the Secondary could take over all but one
node)
> and the Primary, upon its reboot, would become Secondary.

What cluster manager?
heartbeat? homegrown?
Did it complain about something?

> I'm running DRBD 0.6.10 on SuSE 8.2.  drbd.conf (minus comments) looks


> resource www {
>   protocol = B

You should use protocol C. Always.
Unless you know a good reason not to.

>   load-only

Now, why did you use this one?

rest looks ok.

> This is mostly identical for each of our nodes.

Uhm, do you mean nodes as in "computer boxes" or as in "DRBD resources" ?

Do your DRBD resources all share the same device (vgr5) ?


             Lars Ellenberg
_______________________________________________
drbd-user mailing list
drbd-user at lists.linbit.com
http://lists.linbit.com/mailman/listinfo/drbd-user