[openib-general] problems using the ipoib ha package of RH4 U3

[Vladimir Sokolovsky]

Hi Or,
Try to disconnect the port 1 of HCA from IB subnet and then check if ib1
became active.

Regards,
Vladimir
 

> -----Original Message-----
> From: Or Gerlitz [mailto:ogerlitz at voltaire.com] 
> Sent: Tuesday, October 24, 2006 10:36 AM
> To: Vladimir Sokolovsky
> Cc: openib-general at openib.org
> Subject: problems using the ipoib ha package of RH4 U3
> 
> Hi Vlad,
> 
> I am trying to test the ipoib ha package on OFED 1.1 over RH4 
> U3 and does not manage to get it working fine. I am running 
> the following commands:
> 
> ifup ib0
> ifup ib1
> ipoib_ha_log="/var/log/ipoib_ha.log"
> sleep 2
> ./ipoib_ha.pl -p ib0 -s ib1 --with-arping --with-multicast -v 
> > ${ipoib_ha_log} 2>&1 &
> 
> and have ib0 working fine, then i cause the ib port 
> associated with ib0 (port 1 of the hca) to get into 
> initialized state but i don't see any failover happening.
> 
> After the port state is changed, the HA script prints the following
> 
> 	Got CARRIER-ON event on ib0.
> 	Got CARRIER-ON event on ib1.
> 	Got CARRIER-ON event on ib1.
> 
> but it does seem to do any fail-over (ie does not configure 
> ib1 with the address of ib0, etc)
> 
> Below is the full output of the ha script 
> (/var/log/ipoib_ha.log) and my ifcfg config scripts.
> 
> I think there might was an issue with the version of the 
> iproute package, i see that my RH4 U3 system uses 
> iproute-2.6.9-3, is this the issue?
> 
> Or.
> 
> Date:Tue Oct 24 13:25:02 2006
> ib0:
> ======================================
> NETMASK = 255.255.255.0
> BOOTPROTO = static
> IPADDR = 192.168.3.61
> status =
> ONBOOT = yes
> HA = 0
> DEVICE = ib0
> 
> Date:Tue Oct 24 13:25:02 2006
> Bond:
> ======================================
> NETMASK = 255.255.255.0
> BOOTPROTO = static
> IPADDR = 192.168.3.61
> status =
> ONBOOT = yes
> HA = 0
> DEVICE = ib0
> Got CARRIER-ON event on ib0.
> Got CARRIER-ON event on ib1.
> Got CARRIER-ON event on ib1.
> 
> 
> 
> 
> [root at excell02 ipoibtools]# cat 
> /etc/sysconfig/network-scripts/ifcfg-ib0
> # Static settings; all values provided by this file 
> DEVICE=ib0 BOOTPROTO=static ONBOOT=yes
> IPADDR=192.168.3.61
> NETMASK=255.255.255.0
> 
> [root at excell02 ipoibtools]# cat 
> /etc/sysconfig/network-scripts/ifcfg-ib1
> # Static settings; all values provided by this file
> DEVICE=ib1
> BOOTPROTO=static
> ONBOOT=yes
> IPADDR=192.168.3.61
> NETMASK=255.255.255.0
>