Bug 705019

Summary: 2.6.38.5 and greater doesn't boot with mvsas driver
Product: [Fedora] Fedora Reporter: wcronen
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 15CC: andyzhu35, dap78, gansalmon, itamar, jonathan, kernel-maint, madhu.chinakonda, root
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Linux   
Whiteboard:
Fixed In Version: kernel-2.6.38.8-35.fc15 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-07-12 05:26:00 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
full dmesg after modprobe mvsas
none
Proposed fix none

Description wcronen 2011-05-16 12:37:47 UTC 
Description of problem:

Since kernel 2.6.38.5 system hangs at boot time at the debug console. The problem seems to be the sas-driver "mvsas" which is used to boot my system- disks. With kernel 2.6.38.4 and before I works great.

Version-Release number of selected component (if applicable):
kernel 2.6.38.5 and greater with mvsas driver

How reproducible:
Boot system-disks which use the mvsas-driver

Steps to Reproduce:
1.
2.
3.
  
Actual results:
Here are the messages at which the boot process stops:
....
scsi8: mvsas
ata9:00:failed to IDENTIFY (I/O error, err_mask: 0x40)
ata9:00 disabled
scs_alloc_sdev: Allocation failure during SCSI scanning, some SCSI devices might not be configured.
...
dracut Warning: No root device "block:..." found. 

Expected results:
System boots without hang up

Additional info:
Comment 1 Chuck Ebbert 2011-05-18 19:53:42 UTC 
Can you capture all of the boot messages when it fails?
Comment 2 Roland Pallai 2011-06-06 16:21:03 UTC 
I can confirm it. I've 2 Supermicro SASLP-MV8 cards:

01:00.0 RAID bus controller: Marvell Technology Group Ltd. MV64460/64461/64462 System Controller, Revision B (rev 01)
02:00.0 RAID bus controller: Marvell Technology Group Ltd. MV64460/64461/64462 System Controller, Revision B (rev 01)

Full dmesg attached of "modprobe mvsas"
Comment 3 Roland Pallai 2011-06-06 16:21:48 UTC 
Created attachment 503274 [details]
full dmesg after modprobe mvsas
Comment 4 wcronen 2011-06-17 20:28:58 UTC 
Sorry for the late reply.

As the system doesn't boot it was not possible to capture all the messages.

I replaced the controller with an intel raid controller because it's a productive system and it must run without such nasty problems.
Comment 5 Chuck Ebbert 2011-06-22 11:39:23 UTC 
It looks like this may be the same bug that was reported here:

https://patchwork.kernel.org/patch/887662/

The commit that caused this upstream bug was added to 2.6.38.5, and the error messages are exactly the same as reported there. This is still unfixed in 3.0-rcX but the patch in that link should be merged soon.
Comment 6 Chuck Ebbert 2011-06-22 11:42:23 UTC 
Created attachment 505977 [details]
Proposed fix
Comment 7 Chuck Ebbert 2011-06-22 11:43:29 UTC 
*** Bug 708812 has been marked as a duplicate of this bug. ***
Comment 8 Chuck Ebbert 2011-06-23 09:03:57 UTC 
*** Bug 709920 has been marked as a duplicate of this bug. ***
Comment 9 Chuck Ebbert 2011-06-24 19:27:23 UTC 
Should be fixed in 2.6.38.8-35
Comment 10 Fedora Update System 2011-07-07 01:00:41 UTC 
kernel-2.6.38.8-35.fc15 has been submitted as an update for Fedora 15.
https://admin.fedoraproject.org/updates/kernel-2.6.38.8-35.fc15
Comment 11 Fedora Update System 2011-07-08 17:59:04 UTC 
Package kernel-2.6.38.8-35.fc15:
* should fix your issue,
* was pushed to the Fedora 15 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing kernel-2.6.38.8-35.fc15'
as soon as you are able to, then reboot.
Please go to the following url:
https://admin.fedoraproject.org/updates/kernel-2.6.38.8-35.fc15
then log in and leave karma (feedback).
Comment 12 Fedora Update System 2011-07-12 05:25:05 UTC 
kernel-2.6.38.8-35.fc15 has been pushed to the Fedora 15 stable repository.  If problems still persist, please make note of it in this bug report.
Comment 13 andyzhu35 2011-07-13 20:04:23 UTC 
(In reply to comment #12)
> kernel-2.6.38.8-35.fc15 has been pushed to the Fedora 15 stable repository.  If
> problems still persist, please make note of it in this bug report.

Thanks for mvsas bug fix.  I still have problem with kernel-2.6.38.8-35.fc15.x86_64:

My workstation has 2 types of connection: IDE that is meant for optical drives and SATA through Marvell.  Since updated to FC15, the disk drive was no longer accessible through SATA connection; I took advice on the manufacturer's discussion forum: connect the hard disk to IDE port.  FC15 runs fine although slow in I/O (it also fixes the hibernation problem in FC14 through SATA). Now with new kernel for mvsas fix; I gave it a try.  First yum upgrade to the new kernel so that uname -r has 2.6.38.8-35.fc15.x86_64.  Then I reconnected the disk to SATA and it dropped to dracut shell.  blkid returns empty. dmesg has this last line: no root device "block: /dev/disk/by-uuid/${UUID}" found.

So I back into IDE connection and made a fresh new initramfs there and still has same problem when getting back to sata.

On the other hand, I think even if we had a working mvsas in update repository, when we install from DVD we won't be able to direct install FC15 since the installation image has disfunct mvsas driver.  Am I right? please correct me.

Thank you.