Bug 1013443

xhan ,Could you try whether ide can trigger it ?

Created attachment 804921 [details]
memory_dump_bugcheck

QE can reproduce this issue with build 72,win7-64 guest bsod with D1 code.

package info:
    Red Hat Enterprise Linux Server release 6.4 (Santiago)
    qemu-kvm-rhev-0.12.1.2-2.355.el6_4.9.x86_64
    kernel-2.6.32-358.21.1.el6.x86_64
    seabios-0.6.1.2-28.el6.x86_64
    vgabios-0.6b-3.7.el6.noarch
    virtio-win-prewhql-72

QE retest with win7-32,win7-64,win2k8-64,win2k8-32,win2k8R2,only win7-64 hit this issue,cannot reproduce this issue on other guests.

QE retest with win8-32,win8-64,cannot reproduce this issue as well.

QE retest with win2012, cannot reproduce this issue.

since image from https://bugzilla.redhat.com/attachment.cgi?id=804907 and  
http://fileshare.englab.nay.redhat.com/pub/section2/coredump/BZ1013443-MEMMORY.DMP are two different crashes, could you please be more specific on
which BSOD you see? Could you please update the crasdump file with the most recent one? Does the command line from bug description match crashdump file?
Does the problem happen buring unattended install only ot it also reproducible
during manual install?

Thanks,
Vadim.

(In reply to Mike Cao from comment #3)
> xhan ,Could you try whether ide can trigger it ?

change the drive to ide, 

    -device ide-drive,bus=ide.0,unit=0,drive=virtio-scsi0-id0,id=ide0-0-0 \
    -drive file='win7-64-ide.qcow2',if=none,id=virtio-scsi0-id0,media=disk,cache=none,snapshot=off,format=qcow2,aio=native \

This issue could not be hit.

Summary 

on win7 hosts 
scsi system disk + scsi cdrom on the same virtio-scsi-pci  ----> BSOD
scsi system disk + ide cdrom   - ---> works fine
IDE disk    + scsi cdrom        ----> works fine

(In reply to Mike Cao from comment #20)
> Summary 
> 
> on win7 hosts 
Sorry I mean with win7 -64 bit guest

> scsi system disk + scsi cdrom on the same virtio-scsi-pci  ----> BSOD
> scsi system disk + ide cdrom   - ---> works fine
> IDE disk    + scsi cdrom        ----> works fine

Tried scsi system disk + scsi cdrom on different virtio-scsi-pci ---> works fine 

Note I only tested one time for all the results in this comment and comment #21

Tried scsi system disk + scsi cdrom on same virtio-scsi-pci with -cpu family=0xe --> BSOD

(In reply to Vadim Rozenfeld from comment #18)
> since image from https://bugzilla.redhat.com/attachment.cgi?id=804907 and  
> http://fileshare.englab.nay.redhat.com/pub/section2/coredump/BZ1013443-
> MEMMORY.DMP are two different crashes, could you please be more specific on
> which BSOD you see? 

As BSOD most occurs during guest installation ,so there is no dump file for most of times(0x0000001e) . While it is happened after we finished installation ,then we upload the dumps .So we trigger the BSOD with different crashes with the same steps 

>Could you please update the crasdump file with the most
> recent one? Does the command line from bug description match crashdump file?
> Does the problem happen buring unattended install only ot it also
> reproducible
> during manual install?

it can be reproduced on manual install
From my testing ,it only happened when we using scsi-cdrom+scsi-hd on the same virtio-scsi-pci

Mike





> 
> Thanks,
> Vadim.

(In reply to Mike Cao from comment #24)
> (In reply to Vadim Rozenfeld from comment #18)
> > since image from https://bugzilla.redhat.com/attachment.cgi?id=804907 and  
> > http://fileshare.englab.nay.redhat.com/pub/section2/coredump/BZ1013443-
> > MEMMORY.DMP are two different crashes, could you please be more specific on
> > which BSOD you see? 
> 
> As BSOD most occurs during guest installation ,so there is no dump file for
> most of times(0x0000001e) . While it is happened after we finished
> installation ,then we upload the dumps .So we trigger the BSOD with
> different crashes with the same steps 
> 
> >Could you please update the crasdump file with the most
> > recent one? Does the command line from bug description match crashdump file?
> > Does the problem happen buring unattended install only ot it also
> > reproducible
> > during manual install?
> 
> it can be reproduced on manual install
> From my testing ,it only happened when we using scsi-cdrom+scsi-hd on the
> same virtio-scsi-pci
> 
> Mike
> 
> 
> 
> 
> 
> > 
> > Thanks,
> > Vadim.

Hi Mike,
Many thanks for detailed analysis of the problem.

Ronen, let's move this problem to 6.6 because the 
appropriate  problem dissecting, fix and testing can 
take some time. Especially because the problem doesn't 
happen on all platforms.

Thanks,
Vadim.

I am sorry Vadim , win2k8R2 hit the exactly same issue .What I reported on comment #15 is using IDE cdrom

(In reply to Mike Cao from comment #26)
> I am sorry Vadim , win2k8R2 hit the exactly same issue .What I reported on
> comment #15 is using IDE cdrom

Thanks Mike,
Now it seems less weird than before, 
Could you pleas try reproducing the problem on 32-bit
platforms?

Best regards,
Vadim.

(In reply to Vadim Rozenfeld from comment #27)
> (In reply to Mike Cao from comment #26)
> > I am sorry Vadim , win2k8R2 hit the exactly same issue .What I reported on
> > comment #15 is using IDE cdrom
> 
> Thanks Mike,
> Now it seems less weird than before, 
> Could you pleas try reproducing the problem on 32-bit
> platforms?
Following the results is about starting VM with scsi cdrom and scsi hd at the same virtio-scsi-pci

win7-32,win2k8-32,win2k8-64  can not reproduce 
win7-64,win2k8R2 ---> BSOD

QE will try win8 and win2012 tomorrow 


> 
> Best regards,
> Vadim.

(In reply to Mike Cao from comment #28)
> (In reply to Vadim Rozenfeld from comment #27)
> > (In reply to Mike Cao from comment #26)
> > > I am sorry Vadim , win2k8R2 hit the exactly same issue .What I reported on
> > > comment #15 is using IDE cdrom
> > 
> > Thanks Mike,
> > Now it seems less weird than before, 
> > Could you pleas try reproducing the problem on 32-bit
> > platforms?
> Following the results is about starting VM with scsi cdrom and scsi hd at
> the same virtio-scsi-pci
> 
> win7-32,win2k8-32,win2k8-64  can not reproduce 
> win7-64,win2k8R2 ---> BSOD
> 
> QE will try win8 and win2012 tomorrow 
> 
> 
> > 
> > Best regards,
> > Vadim.

win8-64 and win2012 can reproduce this issue as well;
win8-32 guest can be installed without any error.

retest win8-64 with scsi system disk + scsi cdrom on different virtio-scsi-pci,guest can be installed correctly.

*** Bug 996951 has been marked as a duplicate of this bug. ***

Reproduced this issue on virtio-win-prewhql-72;
Verified this issue on virtio-win-prewhql-73 with win7-64 guest.

steps same as comment #0

Actual Results:
on virtio-win-prewhql-72,guest bsod with D1 code during installation;
on virtio-win-prewhql-73,guest can finish installation correctly,no BSOD.

Based on above ,this issue has been fixed already .

Move status to VERIFIED according to comment #32

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2014-1592.html