Red Hat Bugzilla – Bug 826425
[NetApp 5.7 z- KVM Bug] Host Loss seen on Qlogic FC SAN Booted LUN.
Last modified: 2014-02-06 14:26:09 EST
Description of problem:
A SAN Booted RHEL 5.7z-KVM host, with its root LUN mapped to NetApp Storage array over FC, hangs during IO with fabric faults.
When the host hangs, no network service (ssh/telnet/rsh) is operational. Multiple kernel messages logged in syslog reporting about CPU lock-ups (soft lockup - CPU#10 stuck for X seconds) on different processes like quemu-kvm,multipathd,kjournald etc
Version-Release number of selected component (if applicable):
Host Kernel : 2.6.18-274.18.1.el5
HBA : Qlogic 2562
Driver : 8.03.07.03.05.07-k
Firmware : 5.03.16 (95)
Steps to Reproduce:
1. Map 40 LUNs (with 4 FC paths each, i.e, 160 SCSI devices) from controllers
and configure multipath devices on the host.
2.Create 10 LVs on the dm-multipath devices.
3.Create 4 VMs on 4 LVs. And map rest of the LVs to the VMs.
4.Start IO to the LVs and Introduce fabric faults repeatedly.
Should not hang.
It happens only on Qlogic host.
Can you be more specific about the comment "introduce fabric faults"
Also how long does the hypervisor hang?, does it recover.
Can you share the message logs.
More data is needed here please.
Is it possible to obtain kernel traces - so it could be seen, where the process are being stuck ?
This request was evaluated by Red Hat Product Management for inclusion
in a Red Hat Enterprise Linux release. Product Management has
requested further review of this request by Red Hat Engineering, for
potential inclusion in a Red Hat Enterprise Linux release for currently
deployed products. This request is not yet committed for inclusion in