[ofa-general] XmtDiscards

[Boris Shpolyansky]

Hi Bernd,

You can configure the HOQ (Head-Of-Queue-Lifetime) value programmed in
any switch in the fabric managed by OpenSM following these simple steps:

1. Stop the SM
/etc/init.d/opensmd stop

2. Run the SM manually with the "-c" option (to dump its default
configuration to a file)
opensm -c

3. Kill the SM with ^C

4. The configuration is saved in /var/cache/opensm/opensm.opts. Open the
file and look for head_of_queue_lifetime. Change the value and save the
file.

5. Restart the SM
/etc/init.d/opensmd start

P.S. You might find 'opensm -h' and 'man opensm' useful.



Hope this helps,

Boris Shpolyansky
Sr. Member of Technical Staff
Applications
Mellanox Technologies Inc.
2900 Stender Way
Santa Clara, CA 95054
Tel.: (408) 916 0014
Fax: (408) 970 3403
Cell: (408) 834 9365
www.mellanox.com


-----Original Message-----
From: general-bounces at lists.openfabrics.org
[mailto:general-bounces at lists.openfabrics.org] On Behalf Of Bernd
Schubert
Sent: Friday, April 04, 2008 3:13 PM
To: OpenIB
Subject: [ofa-general] XmtDiscards

Hello,

after I upgraded one of our clusters to opensm-3.2.1 it seems to have
gotten much better there, at least no further RcvSwRelayErrors, even
when the cluster is in idle state and so far also no SymbolErrors, which
we also have seens before.

However, after I just started a lustre stress test on 50 clients (to a
lustre storage system with 20 OSS servers and 60 OSTs), ibcheckerrors
reports about 9000 XmtDiscards within 30 minutes.

Searching for this error I find "This is a symptom of congestion and may
require tweaking either HOQ or switch lifetime values". 
Well, I have to admit I neither know what HOQ is, nor do I know how to
tweak it. I also do not have an idea to set switch lifetime values.  I
guess this isn't related to the opensm timeout option, is it?

Hmm, I just found a cisci pdf describing how to set the lifetime on
these switches, but is this also possible on Flextronics switches?


Thanks for any help,
Bernd

-- 
Bernd Schubert
Q-Leap Networks GmbH
_______________________________________________
general mailing list
general at lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general

To unsubscribe, please visit
http://openib.org/mailman/listinfo/openib-general