[ewg] RE: bugs 541 and 465: slow IPoIB CM HA failover and eventual failing IPoIB HA

[Amit Krig]

Thanks for the update, Yohad will reproduce this failure in our labs
 
Amit

________________________________

From: Scott Weitzenkamp (sweitzen) [mailto:sweitzen at cisco.com] 
Sent: Wednesday, May 02, 2007 5:08 AM
To: ewg at lists.openfabrics.org; Scott Weitzenkamp (sweitzen); Tziporet
Koren; Amit Krig; Michael S. Tsirkin; Roland Dreier (rdreier); Moni
Shoua; Moni Levy
Cc: openib
Subject: bugs 541 and 465: slow IPoIB CM HA failover and eventual
failing IPoIB HA


With IPoIB HA (both ipoibtools and ib-bonding), I am seeing slow IPoIB
CM HA failover, and eventually IPoIB stops working after enough
failovers.  I am running netperf -D traffic between two IPoIB HA hosts,
while flipping the 4 host IB ports one at a time (port 1 down, sleep,
port 1 up, sleep, ..., port 4 down, sleep, port 4 up, sleep) in a loop.
 
This is a very easy test to set up.  Can Mellanox and Voltaire please
try to reproduce the problem?
 
I think this problem must be fixed for OFED 1.2 rc3.
 
Scott Weitzenkamp
SQA and Release Manager
Server Virtualization Business Unit
Cisco Systems
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/ewg/attachments/20070502/a6f4167b/attachment.html>