[Linux-cluster] Re: GFS on md on shared disks?

[Ken Preslan]

On Thu, Oct 07, 2004 at 02:39:24PM -0400, Ed L Cashin wrote:
> > 5)  Node A completes writing inode 23.  This consists of:
> >     A) Writing the new block to Disk 0
> >     A) Writing the new parity block to Disk 2
> > 6)  Node A completes writing inode 24.  This consists of:
> 
> That's node B if I am following you correctly.  

Yep.  And steps 5 and 6 should have an A and B step instead of two A steps.
I never was good at using the alphabet.

> ...
> It sounds like it would be slow.  Maybe not in a situation with
> reader-writer locks, where writes were infrequent, though.

Yes, it would be slow.

Reader-writer locks would help some when you're running in degraded mode
after losing a disk.  But in normal operation, you wouldn't need to
lock stripes to read from them -- so all locking requests would be for
the writer version of the lock.

-- 
Ken Preslan <kpreslan at redhat.com>