Re: IBM RDAC (Multipath SAN failover) driver released (GPLed) Was: SAN w/ multiple paths on IBM bladecenter

[Ed Wilts <ewilts ewilts org>]

On Mon, Sep 27, 2004 at 11:31:57AM -0500, D Canfield wrote:
> Also, it would appear to me that problem 4) is that there is no upgrade 
> path for users who installed multipath using the qla2300 support in 
> RHEL3.  I assume that upgrading these boxes to RHEL4 will require 
> re-partitioning all of those disks to md or lvm partitions.  Given that 
> I get about one email a week from someone trying to setup mutli-path 
> booting on a bladecenter, I think this is going to be a bigger issue 
> than the Red Hat kernel team believes.

RHEL 3 U3 now has the new multipathing Qlogic driver.  This allows for
multipathing at the driver level and I would expect it to solve many of
the problems that we've been seeing in the past.

I'm running the new driver on RHEL 3 and have not seen any issues yet.
I don't boot off the SAN and I installed the HP-supplied version of the
Qlogic rev 7 driver, but so far, so good.  The OS really sees only 1
path to the drive - here's the extract from dmesg showing what's
happening:
SCSI device sda: 314572800 512-byte hdwr sectors (161061 MB)
 sda: sda1
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0,  type 12
Attached scsi generic sg2 at scsi0, channel 0, id 1, lun 0,  type 12
Attached scsi generic sg3 at scsi1, channel 0, id 0, lun 0,  type 12
Attached scsi generic sg4 at scsi1, channel 0, id 1, lun 0,  type 12

I mount only /dev/sda1 even though there are 4 paths hiding back there -
2 of which are active and 2 of which are standby (a config that mdadm's
multipathing didn't like at all!).  I'm connecting to HP EVA storage.

> I agree that the "new" solution sounds like a great solution in theory, 
> but my reading suggests the same issues as yours.  The killer being no 
> way to multi-path /boot, and therefore RHEL4 may be less 
> enterprise-ready than RHEL3.

Doubtful.  I would expect that if you build the rev 7 Qlogic driver into
the kernel, booting off the SAN should "just work".

> I'm going to try the beta of RHEL4 on a blade this week and see if it's 
> even remotely functional on a multi-pathed SAN, but I suspect the 
> complete silence we always get on the issue of multipathing /boot hints 
> at the answer to the question of whether that will work...  :-)

I'd be interested in seeing the results of your testing.  Please post
here.

        .../Ed

-- 
Ed Wilts, RHCE
Mounds View, MN, USA
mailto:ewilts ewilts org
Member #1, Red Hat Community Ambassador Program