On 2008-09-22T15:07:37, Xinwei Hu <[EMAIL PROTECTED]> wrote:
> Hi all,
> As you still remember, there's a discussion about sfex in the list a
> while ago. The key idea is to manage the owner ship of shared disk a
> RA, and prevent the nodes to write the disk parallely on split-brain.
>
> I made a little changes to sfex then. Making it's a daemon and
> mlocking itself to prevent from swapping out. And also attached an RA
> to work with scsi-2 reservation, which works in a very similar way
> with sfex.
Hi Xinwei,
thanks for the code.
(I would suggest submission in form of a patch though, or even better an
hg tree to pull from. No need to send binaries to the list ...)
Why not use sg_persist, which is already installed?
SCSI-2 reservations are also not very scalable; I think we need SCSI-3
to handle multiple paths to the device,or possibly multiple hosts. I
think this should support clone groups as well (which could share a
key).
Same goes for sfex, it lacks provision for clones; something which sbd
does deliver.
There's also a bunch of bugs in sfex lib - memory allocation, handling
of direct IO, use of fprintf instead of cl_log, non-portable types in
data written to disk, unimplemented functions ... Does sfex still
provide anything which sbd can't do?
Regards,
Lars
--
Teamlead Kernel, SuSE Labs, Research and Development
SUSE LINUX Products GmbH, GF: Markus Rex, HRB 16746 (AG Nürnberg)
"Experience is the name everyone gives to their mistakes." -- Oscar Wilde
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
