[IBAL]path records pkey issue. was: [ofw] what's up with the OFA Windows project?

[Hal Rosenstock]

On Thu, 2008-02-14 at 11:58 -0500, chas williams - CONTRACTOR wrote:
> In message <C07C40DB2364324799506DE8FF12F8D85960A9 at EPEXCH1.qlogic.org>,"Alex Es
> trin" writes:
> >I believe if SM returns path records with different pkeys it should mean
> >host and a storage are both members of both partitions, and obtained
> >paths are valid for this host.
> >Then the consumer of path records (SRP initiator in our case) could
> >validate what path has higher priority to use.
> >
> >With proposed patch all IOCs on the fabric will be visible only by hosts
> >having default pkey in it's pkey table.
> >For example, if IOC configured to have pkey1 only, then host on the same
> >partition with pkey1 won't ever find this IOC, since it will be looking
> >for it in default partition only.
> 
> i also dont think the sm query returns both paths on both partitions.
> just the path that is the 'best' match.  if it return both paths i
> wouldnt have my problem.  i would think picking the best partition to
> use should be done at the application layer, not the discovery layer?
> 
> it is true that the storage is a member of both partitions.  but i dont
> know of any storage targets that respond to a pkey other than 0xffff.
> the linux ofed stack just ignores this problem completely and hardcodes
> 0xffff when it creates a path between the client and the storage.
> 
> 
> i was under the impression that you need to be a member of the default
> partition in all cases?

Yes, but perhaps only as a limited rather than full member.

> _______________________________________________
> ofw mailing list
> ofw at lists.openfabrics.org
> http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw