[Linux-PowerEdge] qla2xxx hangs during reboot

Baird, Josh jbaird at follett.com
Sat Feb 16 10:27:59 CST 2013


I have a RHEL 5.9 R710 with a QLE8152 in it.  The card is operational and is using FCoE for several multipath devices.  No errors in /var/log/messages.  The problem is that it hangs during reboots (5-10minutes).  The console displays the following:

Unmounting file systems
qla2xxx 0000:05:00.3 scsi(4:1:0): Abort command issued -- 1 497 2002.
qla2xxx 0000:05:00.3 scsi(4:1:0): Abort command issued -- 1 497 2002.
qla2xxx 0000:05:00.3 scsi(4:1:0): DEVICE RESET ISSUED.
qla2xxx 0000:05:00.3 qlx2xxx_eh_device_reset: device reset failed
qla2xxx 0000:05:00.3 scsi(4:1:0): LOOP RESET ISSUED.

Then, a lengthy call trace scrolls across the screen.  After, this is displayed:

qla2xxx 0000:05:00.3 qlx2xxx_eh_device_reset: reset succeeded.
Rport-4:0-1: blocked FC remote port time out:  saving binding 
qla2xxx 0000:05:00.3 scsi(4:1:0): ADAPTER RESET ISSUED.
qla2xxx 0000:05:00.3 Performing ISP error recovery - ha = fff81102e4e84f8.
qla2xxx 0000:05:00.3 LOOP UP detected (10 Gbps).
qla2xxx 0000:05:00.3 qla2xxx_eh_host_reset: reset succeeded.

Then, this entire 5+ minute process repeats for the second mpath device.  I have tried the qla2xxx drivers that ship with RHEL5.9, as well the newest available from qLogic:

$ modinfo qla2xxx
filename:       /lib/modules/2.6.18-348.1.1.el5/weak-updates/qlgc-qla2xxx/qla2xxx.ko
version:        8.04.00.06.5.6-k
license:        GPL
description:    QLogic Fibre Channel HBA Driver
author:         QLogic Corporation
srcversion:     975E6DDF55788BFC9645012 

Does anyone know what may be causing these problems during a reboot?

Thanks,

Josh



More information about the Linux-PowerEdge mailing list