[DRBD-user] Confused about no-disk-drain

[Jake Smith]

----- Original Message -----
> From: "Lars Ellenberg" <lars.ellenberg at linbit.com>
> To: drbd-user at lists.linbit.com
> Sent: Thursday, February 2, 2012 10:37:31 AM
> Subject: Re: [DRBD-user] Confused about no-disk-drain
> 
> On Thu, Feb 02, 2012 at 10:27:05AM -0500, Jake Smith wrote:
> > Lars,
> > 
> > I didn't want to hijack a different thread so I started a new one.
> >  I
> > want to understand better why you said in another thread today not
> > to
> > use no-disk-drain (quote):
> > 
> > > Will people please STOP using no-disk-drain.  On most hardware,
> > > it does
> > > not provide measurable performance gain, but may risk data
> > > integrity
> > > because of potential violation of write-after-write dependencies!
> > 
> > I also found you said in another older thread:
> > 
> > >Do not use "no-disk-drain", unless you are absolutely sure that
> > >nothing
> > >in the IO stack below DRBD could possibly reorder writes.
> > 
> > Unfortunately I'm not that knowledgeable on storage so this may be
> > a ignorant question but...
> > 
> > In the man page for drbd.conf (version 8.3.7) it states using none
> > is
> > the best performance choice if using a storage device with BBWC.
> 
> Which it is. But as you are not the only one confused about this,
> we felt the need to clarify and improve the man page, see there:
> 
> http://git.drbd.org/gitweb.cgi?p=drbd-8.3.git;a=commitdiff;h=f952dcfe
> 
> .-- Commit message:
> | doc: clearly document no-disk-drain as DONT USE
> | 
> | When used carelessly, because it is documented as the method to get
> | "best performance" (which it is, but only if used on a suitable IO
> | stack), it can theoretically cause write-reordering (potential data
> | corruption), and will disturb the drbd protocol causing spurious
> | disconnect/reconnect cycles, like:
> | 
> | block drbd0: BAD! BarrierAck #822515458 received, expected
> | #822515457!
> | block drbd0: peer( Secondary -> Unknown ) conn( Connected ->
> | ProtocolError ) pdsk( UpToDate -> DUnknown )
> | 
> | There are very few cases where using that option is "legal",
> | and even then the performance increase is marginal.
> | 
> | Document as DONT USE.
> '--
> 
> Actual diff (and new man page text) see url above.

Well I may not understand *why* much better but that's OK because I can follow directions so Do Not Use works for me! :-)

Will both 8.3 and 8.4 users guides be updated to reflect these changes at some point?

Thanks!

Jake

> 
> --
> : Lars Ellenberg
> : LINBIT | Your Way to High Availability
> : DRBD/HA support and consulting http://www.linbit.com
> _______________________________________________
> drbd-user mailing list
> drbd-user at lists.linbit.com
> http://lists.linbit.com/mailman/listinfo/drbd-user
> 
>