[ofa-general] What causes "SRP abort called" error?

[David Dillow]

On Tue, 2008-01-08 at 22:45 -0600, John Valdes wrote:
> In our case, I'm pretty sure it is heavy load.  Well, I didn't see
> what was going on at the time this started, but the targets (LUNs)
> were already mounted, and we've been seeing heavy load on the DDN
> recently. 

Search the list archives for some patches I sent to limit the queue
length under IB/srp and also to respect the targets credit limit. If you
don't find them, I can forward them to you when I'm back in the office.

As a quick workaround, one each host, you can add the
"max_cmd_per_lun=..." parameter to your add-target string to avoid
overrunning the array. Set it to 
floor(30 / <# of LUNs active on this IB port>)

If you still have problems, try lowering it in steps of 1.

This is a workaround for the initiator overrunning the credit limit,
which is known to cause problems on some arrays. Perhaps it will help
you.

Dave