Bug 495881

Summary: [NetApp 4.8 bug]Kernel panics on RHEL4.8 GASnapshot3 after reloading QLogic FC driver when LUNs are mapped
Product: Red Hat Enterprise Linux 4 Reporter: Naveen Reddy <naveenr>
Component: kernelAssignee: Marcus Barrow <mbarrow>
Status: CLOSED WORKSFORME QA Contact: Red Hat Kernel QE team <kernel-qe>
Severity: urgent Docs Contact:
Priority: low    
Version: 4.8CC: andriusb, benl, coughlan, naveenr, qlogic-redhat-ext, xdl-redhat-bugzilla
Target Milestone: rcKeywords: Regression
Target Release: 4.8   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-04-19 17:52:55 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 450896, 450897    
Attachments:
Description Flags
Attaching the kernel panic message none

Description Naveen Reddy 2009-04-15 12:43:40 UTC 
Description of problem:
On a RHEL4.8 GASnapshot3 host with QLogic FC HBA, kernel panics when the QLogic FC HBA driver is reloaded after mapping few LUNs from NetApp Storage Controller.


Version-Release number of selected component (if applicable):
OS - RHEL4.8 GASnapshot3
kernel  - 2.6.9-86.ELsmp
QLogic FC HBA driver version:    v.8.02.09.01.04.08-d

How reproducible:
Always

Steps to Reproduce:
1. Install RHEL4.8 GASnapshot3 on a host with QLogic FC HBA 
2. Map LUNs from a NetApp storage controller
3. Reload the qla2400 module (as it is 4GB FC HBA)
 # modprobe -r qla2400
 # modprobe -v qla2400
4. kernel panics after reloading the QLogic FC HBA driver

Actual results:
kernel panics after reloading the QLogic FC HBA driver when LUNs are mapped to the host.(Attached the log collected during kernel panic).

Expected results:
kernel should not panic.

Additional Info:
This issue was not seen in RHEL4.8 GASnapshot2 or earlier(Previous Releases).
Comment 1 Naveen Reddy 2009-04-15 12:49:00 UTC 
Created attachment 339681 [details]
Attaching the kernel panic message

Attaching the kernel panic message
Comment 2 Tom Coughlan 2009-04-17 14:34:43 UTC 
There is a fix in snap 4, kernel -87, that may be relevant. The fix removes some qla sysfs values that cause hangs when they are read (like at boot time and module load). Can you try snap 4, or just switch to the -87 kernel?
Comment 3 Marcus Barrow 2009-04-17 14:54:24 UTC 
I concur with Tom C. Please try the -88 kernel. You could also apply the patch in BZ  476704. The patch removed the sysfs entries which were causing a locking problem. These entries were unsupported and when other fixes were made to the driver they became broken. These entries are read in an automated fashion and are not required ( board nvram contents ).
Comment 4 Naveen Reddy 2009-04-19 13:53:00 UTC 
This issue is not seen in snap4(kernel v2.6.9-87).