Fedora 11 Alpha on KVM-84

Fedora 11 Alpha on KVM-84

[Mark Bidewell]

I tried installing Fedora 11 Alpha in KVM-84 however Anaconda hard
crashes when installing packages.  Installing F10 and updating to
Rawhide in KVM yields guest kernel errors.  I had similar problems in
the F9 to F10 step.  I was wondering if anyone else had seen this or
what I could do to debug it?

Thanks.

-- 
Mark Bidewell
http://www.linkedin.com/in/markbidewell

Re: Fedora 11 Alpha on KVM-84

[Fabian Deutsch]

Hey,

Am Samstag, den 21.02.2009, 14:53 -0500 schrieb Mark Bidewell:
> I tried installing Fedora 11 Alpha in KVM-84 however Anaconda hard
> crashes when installing packages.  Installing F10 and updating to
> Rawhide in KVM yields guest kernel errors.  I had similar problems in
> the F9 to F10 step.  I was wondering if anyone else had seen this or
> what I could do to debug it?

Yes. I recognized this problem.
You might have a look at :

https://bugzilla.redhat.com/show_bug.cgi?id=476433

and

https://bugzilla.redhat.com/show_bug.cgi?id=484721

I've also tried those setups with kvm master.

- fabiand

Re: Fedora 11 Alpha on KVM-84

[Marcelo Tosatti]

On Sat, Feb 21, 2009 at 08:59:49PM +0100, Fabian Deutsch wrote:
> Hey,
> 
> Am Samstag, den 21.02.2009, 14:53 -0500 schrieb Mark Bidewell:
> > I tried installing Fedora 11 Alpha in KVM-84 however Anaconda hard
> > crashes when installing packages.  Installing F10 and updating to
> > Rawhide in KVM yields guest kernel errors.  I had similar problems in
> > the F9 to F10 step.  I was wondering if anyone else had seen this or
> > what I could do to debug it?
> 
> Yes. I recognized this problem.
> You might have a look at :
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=476433
> 
> and
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=484721
> 
> I've also tried those setups with kvm master.

This errors are probably caused by a paravirt mmu bug which is fixed in
linux-2.6.git.

Mark, is the fix part of recent FC11 kernels?

Re: Fedora 11 Alpha on KVM-84

[Mark McLoughlin]

On Sun, 2009-02-22 at 16:35 -0300, Marcelo Tosatti wrote:
> On Sat, Feb 21, 2009 at 08:59:49PM +0100, Fabian Deutsch wrote:
> > Hey,
> > 
> > Am Samstag, den 21.02.2009, 14:53 -0500 schrieb Mark Bidewell:
> > > I tried installing Fedora 11 Alpha in KVM-84 however Anaconda hard
> > > crashes when installing packages.

Mark, can you confirm that this is the same as either of the issues
reported by Fabian? Was this a LiveCD install or ... ?

> > > Installing F10 and updating to Rawhide in KVM yields guest kernel 
> > > errors.  

Could you post the errors you see?

> > > I had similar problems in
> > > the F9 to F10 step.  I was wondering if anyone else had seen this or
> > > what I could do to debug it?

The best way to debug such issues is to use a serial console - if you're
using libvirt, that should already be available to the guest and you
just need to add "console=ttyS0" to the kernel command line.

> > Yes. I recognized this problem.
> > You might have a look at :
> > 
> > https://bugzilla.redhat.com/show_bug.cgi?id=476433

Your report has an identical anaconda stack trace to a previous
bare-metal F10 bug ... I don't think it's KVM specific.

> > and
> > 
> > https://bugzilla.redhat.com/show_bug.cgi?id=484721

We need some more detail on this one, I think - e.g. where does anaconda
hang? Can you get an strace of anaconda?

> > I've also tried those setups with kvm master.
> 
> This errors are probably caused by a paravirt mmu bug which is fixed in
> linux-2.6.git.

Hmm, I'm not sure why you think that ... the pvmmu issue had a very
specific kernel oops signature, right?

  https://bugzilla.redhat.com/480822

> Mark, is the fix part of recent FC11 kernels?

If you mean the flush_lazy_mmu_mode() fix, then yep it first appeared in
v2.6.29-rc5-git1:

  http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=4f06b0436b

Cheers,
Mark.

Fwd: Fedora 11 Alpha on KVM-84

[Mark Bidewell]

---------- Forwarded message ----------
From: Mark McLoughlin <markmc@redhat.com>
Date: Tue, Feb 24, 2009 at 9:36 AM
Subject: Re: Fedora 11 Alpha on KVM-84
To: Mark Bidewell <mark.bidewell@alumni.clemson.edu>


On Tue, 2009-02-24 at 09:14 -0500, Mark Bidewell wrote:
> BUG: unable to handle kernel paging request at ffff88002d1d9f58
> IP: [<ffffffff8100fc7a>] copy_thread+0x42/0x1bf
> PGD 202063 PUD 206063 PMD 15c3067 PTE 2d1d9160
> Oops: 0002 [#1] SMP DEBUG_PAGEALLOC
> last sysfs file: /sys/devices/pci0000:00/0000:00:01.1/host0/target0:0:0/0:0:0:0/block/sda/uevent
> CPU 0
> Modules linked in: ipv6 xts lrw gf128mul sha256_generic cbc dm_crypt dm_round_robin dm_multipath btrfs zlib_deflate libcrc32c xfs exportfs jfs reiserfs gfs2 msdos linear raid10 raid456 async_xor async_memcpy async_tx xor raid1 raid0 virtio_pci virtio_ring virtio 8139cp 8139too mii iscsi_ibft iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ext2 ext4 jbd2 crc16 squashfs pcspkr edd floppy nfs lockd nfs_acl auth_rpcgss sunrpc vfat fat cramfs
> Pid: 1543, comm: udevd Not tainted 2.6.29-0.66.rc3.fc11.x86_64 #1
> RIP: 0010:[<ffffffff8100fc7a>]  [<ffffffff8100fc7a>] copy_thread+0x42/0x1bf
> RSP: 0018:ffff88002d0b5d80  EFLAGS: 00010286
> RAX: ffff88002d1da000 RBX: ffff880033000000 RCX: 000000000000002a
> RDX: 00007fff25d6c900 RSI: ffff88002d0b5f58 RDI: ffff88002d1d9f58
> RBP: ffff88002d0b5db0 R08: ffff880033000000 R09: ffff88002d0b5f58
> R10: ffff880033002b80 R11: 800000001f230045 R12: ffff88002d1d9f58
> R13: ffff880033002390 R14: 0000000001200011 R15: 0000000001200011
> FS:  00007f031dd6e790(0000) GS:ffffffff81934000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: ffff88002d1d9f58 CR3: 000000002f959000 CR4: 00000000000006e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process udevd (pid: 1543, threadinfo ffff88002d0b4000, task ffff880033002390)
> Stack:
>  ffff880037ca44a8 ffff880033dd50c0 0000000000000000 ffff880033000000
>  0000000000000000 0000000000000000 ffff88002d0b5e60 ffffffff81049396
>  0000000000000000 ffff88002d0b5e18 ffffffff8102ad8c ffff880033002b80
> Call Trace:
>  [<ffffffff81049396>] copy_process+0xc68/0x136d
>  [<ffffffff8102ad8c>] ? pvclock_clocksource_read+0x42/0x7e
>  [<ffffffff81049bfb>] do_fork+0x160/0x31f
>  [<ffffffff813833cb>] ? _spin_unlock+0x26/0x2a
>  [<ffffffff810dcf08>] ? fd_install+0x56/0x5f
>  [<ffffffff8101130a>] ? sysret_check+0x46/0x81
>  [<ffffffff8100f62d>] sys_clone+0x23/0x25
>  [<ffffffff810117c3>] stub_clone+0x13/0x20
>  [<ffffffff810112ba>] ? system_call_fastpath+0x16/0x1b
> Code: 89 c3 48 83 ec 10 49 8b 40 08 65 4c 8b 2c 25 00 00 00 00 48 83 fa ff 4c 8d a0 58 1f 00 00 4c 89 e7 49 0f 44 d4 48 05 00 20 00 00 <f3> a5 49 c7 44 24 50 00 00 00 00 49 89 94 24 98 00 00 00 49 89
> RIP  [<ffffffff8100fc7a>] copy_thread+0x42/0x1bf
>  RSP <ffff88002d0b5d80>
> CR2: ffff88002d1d9f58

Post this on kvm@vger for Marcelo to look at, but my guess is that it's
the pvmmu issue he fixed already. If so, the F11 beta should work fine.

Cheers,
Mark.


Just forwarding this as requested.

-- 
Mark Bidewell
http://www.linkedin.com/in/markbidewell

Re: Fwd: Fedora 11 Alpha on KVM-84

[Marcelo Tosatti]

On Tue, Feb 24, 2009 at 09:57:19AM -0500, Mark Bidewell wrote:
> ---------- Forwarded message ----------
> From: Mark McLoughlin <markmc@redhat.com>
> Date: Tue, Feb 24, 2009 at 9:36 AM
> Subject: Re: Fedora 11 Alpha on KVM-84
> To: Mark Bidewell <mark.bidewell@alumni.clemson.edu>
> 
> 
> On Tue, 2009-02-24 at 09:14 -0500, Mark Bidewell wrote:
> > BUG: unable to handle kernel paging request at ffff88002d1d9f58
> > IP: [<ffffffff8100fc7a>] copy_thread+0x42/0x1bf
> > PGD 202063 PUD 206063 PMD 15c3067 PTE 2d1d9160
> > Oops: 0002 [#1] SMP DEBUG_PAGEALLOC
> > last sysfs file: /sys/devices/pci0000:00/0000:00:01.1/host0/target0:0:0/0:0:0:0/block/sda/uevent
> > CPU 0
> > Modules linked in: ipv6 xts lrw gf128mul sha256_generic cbc dm_crypt dm_round_robin dm_multipath btrfs zlib_deflate libcrc32c xfs exportfs jfs reiserfs gfs2 msdos linear raid10 raid456 async_xor async_memcpy async_tx xor raid1 raid0 virtio_pci virtio_ring virtio 8139cp 8139too mii iscsi_ibft iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ext2 ext4 jbd2 crc16 squashfs pcspkr edd floppy nfs lockd nfs_acl auth_rpcgss sunrpc vfat fat cramfs
> > Pid: 1543, comm: udevd Not tainted 2.6.29-0.66.rc3.fc11.x86_64 #1
> > RIP: 0010:[<ffffffff8100fc7a>]  [<ffffffff8100fc7a>] copy_thread+0x42/0x1bf
> > RSP: 0018:ffff88002d0b5d80  EFLAGS: 00010286
> > RAX: ffff88002d1da000 RBX: ffff880033000000 RCX: 000000000000002a
> > RDX: 00007fff25d6c900 RSI: ffff88002d0b5f58 RDI: ffff88002d1d9f58
> > RBP: ffff88002d0b5db0 R08: ffff880033000000 R09: ffff88002d0b5f58
> > R10: ffff880033002b80 R11: 800000001f230045 R12: ffff88002d1d9f58
> > R13: ffff880033002390 R14: 0000000001200011 R15: 0000000001200011
> > FS:  00007f031dd6e790(0000) GS:ffffffff81934000(0000) knlGS:0000000000000000
> > CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> > CR2: ffff88002d1d9f58 CR3: 000000002f959000 CR4: 00000000000006e0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> > Process udevd (pid: 1543, threadinfo ffff88002d0b4000, task ffff880033002390)
> > Stack:
> >  ffff880037ca44a8 ffff880033dd50c0 0000000000000000 ffff880033000000
> >  0000000000000000 0000000000000000 ffff88002d0b5e60 ffffffff81049396
> >  0000000000000000 ffff88002d0b5e18 ffffffff8102ad8c ffff880033002b80
> > Call Trace:
> >  [<ffffffff81049396>] copy_process+0xc68/0x136d
> >  [<ffffffff8102ad8c>] ? pvclock_clocksource_read+0x42/0x7e
> >  [<ffffffff81049bfb>] do_fork+0x160/0x31f
> >  [<ffffffff813833cb>] ? _spin_unlock+0x26/0x2a
> >  [<ffffffff810dcf08>] ? fd_install+0x56/0x5f
> >  [<ffffffff8101130a>] ? sysret_check+0x46/0x81
> >  [<ffffffff8100f62d>] sys_clone+0x23/0x25
> >  [<ffffffff810117c3>] stub_clone+0x13/0x20
> >  [<ffffffff810112ba>] ? system_call_fastpath+0x16/0x1b
> > Code: 89 c3 48 83 ec 10 49 8b 40 08 65 4c 8b 2c 25 00 00 00 00 48 83 fa ff 4c 8d a0 58 1f 00 00 4c 89 e7 49 0f 44 d4 48 05 00 20 00 00 <f3> a5 49 c7 44 24 50 00 00 00 00 49 89 94 24 98 00 00 00 49 89
> > RIP  [<ffffffff8100fc7a>] copy_thread+0x42/0x1bf
> >  RSP <ffff88002d0b5d80>
> > CR2: ffff88002d1d9f58
> 
> Post this on kvm@vger for Marcelo to look at, but my guess is that it's
> the pvmmu issue he fixed already. If so, the F11 beta should work fine.

That could be the pvmmu bug. Please try F11 beta.