Re: [ovirt-users] [Qemu-block] qcow2 images corruption

[Nicolas Ecarnot]

https://framadrop.org/r/Lvvr392QZo#/wOeYUUlHQAtkUw1E+x2YdqTqq21Pbic6OPBIH0TjZE=

Le 14/02/2018 à 00:01, John Snow a écrit :



On 02/13/2018 04:41 AM, Kevin Wolf wrote:

Am 07.02.2018 um 18:06 hat Nicolas Ecarnot geschrieben:

TL; DR : qcow2 images keep getting corrupted. Any workaround?


Not without knowing the cause.

The first thing to make sure is that the image isn't touched by a second
process while QEMU is running a VM. The classic one is using 'qemu-img
snapshot' on the image of a running VM, which is instant corruption (and
newer QEMU versions have locking in place to prevent this), but we have
seen more absurd cases of things outside QEMU tampering with the image
when we were investigating previous corruption reports.

This covers the majority of all reports, we haven't had a real
corruption caused by a QEMU bug in ages.


After having found (https://access.redhat.com/solutions/1173623) the right
logical volume hosting the qcow2 image, I can run qemu-img check on it.
- On 80% of my VMs, I find no errors.
- On 15% of them, I find Leaked cluster errors that I can correct using
"qemu-img check -r all"
- On 5% of them, I find Leaked clusters errors and further fatal errors,
which can not be corrected with qemu-img.
In rare cases, qemu-img can correct them, but destroys large parts of the
image (becomes unusable), and on other cases it can not correct them at all.


It would be good if you could make the 'qemu-img check' output available
somewhere.

It would be even better if we could have a look at the respective image.
I seem to remember that John (CCed) had a few scripts to analyse
corrupted qcow2 images, maybe we would be able to see something there.



Hi! I did write a pretty simplistic tool for trying to tell the shape of
a corruption at a glance. It seems to work pretty similarly to the other
tool you already found, but it won't hurt anything to run it:

https://github.com/jnsnow/qcheck

(Actually, that other tool looks like it has an awful lot of options.
I'll have to check it out.)

It can print a really upsetting amount of data (especially for very
corrupt images), but in the default case, the simple setting should do
the trick just fine.

You could always put the output from this tool in a pastebin too; it
might help me visualize the problem a bit more -- I find seeing the
exact offsets and locations of where all the various tables and things
to be pretty helpful.

You can also always use the "deluge" option and compress it if you want,
just don't let it print to your terminal:

jsnow@probe (dev) ~/s/qcheck> ./qcheck -xd
/home/bos/jsnow/src/qemu/bin/git/install_test_f26.qcow2 > deluge.log;
and ls -sh deluge.log
4.3M deluge.log

but it compresses down very well:

jsnow@probe (dev) ~/s/qcheck> 7z a -t7z -m0=ppmd deluge.ppmd.7z deluge.log
jsnow@probe (dev) ~/s/qcheck> ls -s deluge.ppmd.7z
316 deluge.ppmd.7z

So I suppose if you want to send along:
(1) The basic output without any flags, in a pastebin
(2) The zipped deluge output, just in case

and I will try my hand at guessing what went wrong.


(Also, maybe my tool will totally choke for your image, who knows. It
hasn't received an overwhelming amount of testing apart from when I go
to use it personally and inevitably wind up displeased with how it
handles certain situations, so ...)


What I read similar to my case is :
- usage of qcow2
- heavy disk I/O
- using the virtio-blk driver

In the proxmox thread, they tend to say that using virtio-scsi is the
solution. Having asked this question to oVirt experts
(https://lists.ovirt.org/pipermail/users/2018-February/086753.html) but it's
not clear the driver is to blame.


This seems very unlikely. The corruption you're seeing is in the qcow2
metadata, not only in the guest data. If anything, virtio-scsi exercises
more qcow2 code paths than virtio-blk, so any potential bug that affects
virtio-blk should also affect virtio-scsi, but not the other way around.


I agree with the answer Yaniv Kaul gave to me, saying I have to properly
report the issue, so I'm longing to know which peculiar information I can
give you now.


To be honest, debugging corruption after the fact is pretty hard. We'd
need the 'qemu-img check' output and ideally the image to do anything,
but I can't promise that anything would come out of this.

Best would be a reproducer, or at least some operation that you can link
to the appearance of the corruption. Then we could take a more targeted
look at the respective code.


As you can imagine, all this setup is in production, and for most of the
VMs, I can not "play" with them. Moreover, we launched a campaign of nightly
stopping every VM, qemu-img check them one by one, then boot.
So it might take some time before I find another corrupted image.
(which I'll preciously store for debug)

Other informations : We very rarely do snapshots, but I'm close to imagine
that automated migrations of VMs could trigger similar behaviors on qcow2
images.


To my knowledge, oVirt only uses 

Re: [ovirt-users] [Qemu-block] qcow2 images corruption

[John Snow]

On 02/13/2018 04:41 AM, Kevin Wolf wrote:
> Am 07.02.2018 um 18:06 hat Nicolas Ecarnot geschrieben:
>> TL; DR : qcow2 images keep getting corrupted. Any workaround?
> 
> Not without knowing the cause.
> 
> The first thing to make sure is that the image isn't touched by a second
> process while QEMU is running a VM. The classic one is using 'qemu-img
> snapshot' on the image of a running VM, which is instant corruption (and
> newer QEMU versions have locking in place to prevent this), but we have
> seen more absurd cases of things outside QEMU tampering with the image
> when we were investigating previous corruption reports.
> 
> This covers the majority of all reports, we haven't had a real
> corruption caused by a QEMU bug in ages.
> 
>> After having found (https://access.redhat.com/solutions/1173623) the right
>> logical volume hosting the qcow2 image, I can run qemu-img check on it.
>> - On 80% of my VMs, I find no errors.
>> - On 15% of them, I find Leaked cluster errors that I can correct using
>> "qemu-img check -r all"
>> - On 5% of them, I find Leaked clusters errors and further fatal errors,
>> which can not be corrected with qemu-img.
>> In rare cases, qemu-img can correct them, but destroys large parts of the
>> image (becomes unusable), and on other cases it can not correct them at all.
> 
> It would be good if you could make the 'qemu-img check' output available
> somewhere.
> 
> It would be even better if we could have a look at the respective image.
> I seem to remember that John (CCed) had a few scripts to analyse
> corrupted qcow2 images, maybe we would be able to see something there.
> 

Hi! I did write a pretty simplistic tool for trying to tell the shape of
a corruption at a glance. It seems to work pretty similarly to the other
tool you already found, but it won't hurt anything to run it:

https://github.com/jnsnow/qcheck

(Actually, that other tool looks like it has an awful lot of options.
I'll have to check it out.)

It can print a really upsetting amount of data (especially for very
corrupt images), but in the default case, the simple setting should do
the trick just fine.

You could always put the output from this tool in a pastebin too; it
might help me visualize the problem a bit more -- I find seeing the
exact offsets and locations of where all the various tables and things
to be pretty helpful.

You can also always use the "deluge" option and compress it if you want,
just don't let it print to your terminal:

jsnow@probe (dev) ~/s/qcheck> ./qcheck -xd
/home/bos/jsnow/src/qemu/bin/git/install_test_f26.qcow2 > deluge.log;
and ls -sh deluge.log
4.3M deluge.log

but it compresses down very well:

jsnow@probe (dev) ~/s/qcheck> 7z a -t7z -m0=ppmd deluge.ppmd.7z deluge.log
jsnow@probe (dev) ~/s/qcheck> ls -s deluge.ppmd.7z
316 deluge.ppmd.7z

So I suppose if you want to send along:
(1) The basic output without any flags, in a pastebin
(2) The zipped deluge output, just in case

and I will try my hand at guessing what went wrong.


(Also, maybe my tool will totally choke for your image, who knows. It
hasn't received an overwhelming amount of testing apart from when I go
to use it personally and inevitably wind up displeased with how it
handles certain situations, so ...)

>> What I read similar to my case is :
>> - usage of qcow2
>> - heavy disk I/O
>> - using the virtio-blk driver
>>
>> In the proxmox thread, they tend to say that using virtio-scsi is the
>> solution. Having asked this question to oVirt experts
>> (https://lists.ovirt.org/pipermail/users/2018-February/086753.html) but it's
>> not clear the driver is to blame.
> 
> This seems very unlikely. The corruption you're seeing is in the qcow2
> metadata, not only in the guest data. If anything, virtio-scsi exercises
> more qcow2 code paths than virtio-blk, so any potential bug that affects
> virtio-blk should also affect virtio-scsi, but not the other way around.
> 
>> I agree with the answer Yaniv Kaul gave to me, saying I have to properly
>> report the issue, so I'm longing to know which peculiar information I can
>> give you now.
> 
> To be honest, debugging corruption after the fact is pretty hard. We'd
> need the 'qemu-img check' output and ideally the image to do anything,
> but I can't promise that anything would come out of this.
> 
> Best would be a reproducer, or at least some operation that you can link
> to the appearance of the corruption. Then we could take a more targeted
> look at the respective code.
> 
>> As you can imagine, all this setup is in production, and for most of the
>> VMs, I can not "play" with them. Moreover, we launched a campaign of nightly
>> stopping every VM, qemu-img check them one by one, then boot.
>> So it might take some time before I find another corrupted image.
>> (which I'll preciously store for debug)
>>
>> Other informations : We very rarely do snapshots, but I'm close to imagine
>> that automated migrations of VMs could trigger similar behaviors on qcow2
>> images.
> 
> To my 

Re: [ovirt-users] [Qemu-block] qcow2 images corruption

[Nicolas Ecarnot]

Le 13/02/2018 à 16:26, Nicolas Ecarnot a écrit :
>> It would be good if you could make the 'qemu-img check' output available
>> somewhere.
>

I found this :
https://github.com/ShijunDeng/qcow2-dump

and the transcript (beautiful colors when viewed with "more") is attached :


--
Nicolas ECARNOT
Le script a débuté sur mar. 13 févr. 2018 17:31:05 CET
]0;root@serv-hv-adm13:/home[?1034hroot@serv-hv-adm13:/home#
 /root/qcow2-dump -m check serv-term-adm4-corr.qcow2.img

File: serv-term-adm4-corr.qcow2.img


magic: 0x514649fb
version: 2
backing_file_offset: 0x0
backing_file_size: 0
fs_type: xfs
virtual_size: 64424509440 / 61440M / 60G
disk_size: 36507222016 / 34816M / 34G
seek_end: 36507222016 [0x88000] / 34816M / 34G
cluster_bits: 16
cluster_size: 65536
crypt_method: 0
csize_shift: 54
csize_mask: 255
cluster_offset_mask: 0x3f
l1_table_offset: 0x76a46
l1_size: 120
l1_vm_state_index: 120
l2_size: 8192
refcount_order: 4
refcount_bits: 16
refcount_block_bits: 15
refcount_block_size: 32768
refcount_table_offset: 0x1
refcount_table_clusters: 1
snapshots_offset: 0x0
nb_snapshots: 0
incompatible_features: 
compatible_features: 
autoclear_features: 



Active Snapshot:

L1 Table:   [offset: 0x76a46, len: 120]

Result:
L1 Table:   unaligned: 0, invalid: 0, unused: 53, 
used: 67
L2 Table:   unaligned: 0, invalid: 0, unused: 20304, 
used: 528560



Refcount Table:

Refcount Table: [offset: 0x1, len: 8192]

Result:
Refcount Table: unaligned: 0, invalid: 0, unused: 
8175, used: 17
Refcount:   error: 4342, leak: 0, unused: 28426, 
used: 524288



COPIED OFLAG:


Result:
L1 Table ERROR OFLAG_COPIED: 1
L2 Table ERROR OFLAG_COPIED: 4323
Active L2 COPIED: 528560 [34639708160 / 33035M / 32G]



Active Cluster:


Result:
Active Cluster: reuse: 17



Summary:
preallocation:  off
Active Cluster: reuse: 17
Refcount Table: unaligned: 0, invalid: 0, unused: 
8175, used: 17
Refcount:   error: 4342, leak: 0, 
rebuild: 4325, unused: 28426, used: 524288
L1 Table:   unaligned: 0, invalid: 0, unused: 53, 
used: 67
oflag copied: 1
L2 Table:   unaligned: 0, invalid: 0, unused: 
20304, used: 528560
oflag copied: 4323


### qcow2 image has refcount errors!   (=_=#)###
###and qcow2 image has copied errors!  (o_0)?###
###  Sadly: refcount error cause active cluster reused! Orz  ###
### Please backup this image and contact the author! ###



]0;root@serv-hv-adm13:/homeroot@serv-hv-adm13:/home#
 exit

Script terminé sur mar. 13 févr. 2018 17:31:13 CET
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] [Qemu-block] qcow2 images corruption

[Nicolas Ecarnot]

Hello Kevin,

Le 13/02/2018 à 10:41, Kevin Wolf a écrit :

Am 07.02.2018 um 18:06 hat Nicolas Ecarnot geschrieben:

TL; DR : qcow2 images keep getting corrupted. Any workaround?


Not without knowing the cause.


Actually, my main concern is mostly about finding the cause rather than 
correcting my corrupted VMs.


Another way to say it : I prefer to help oVirt than help myself.


The first thing to make sure is that the image isn't touched by a second
process while QEMU is running a VM.


Indeed, I read some BZ about this issue : they were raised by a user who 
ran some qemu-img commands on a "mounted" image, thus leading to some 
corruption.
In my case, I'm not playing with this, and the corrupted VMs were only 
touched by classical oVirt actions.



The classic one is using 'qemu-img
snapshot' on the image of a running VM, which is instant corruption (and
newer QEMU versions have locking in place to prevent this), but we have
seen more absurd cases of things outside QEMU tampering with the image
when we were investigating previous corruption reports.

This covers the majority of all reports, we haven't had a real
corruption caused by a QEMU bug in ages.


May I ask after what QEMU version this kind of locking has been added.
As I wrote, our oVirt setup is 3.6 so not recent.




After having found (https://access.redhat.com/solutions/1173623) the right
logical volume hosting the qcow2 image, I can run qemu-img check on it.
- On 80% of my VMs, I find no errors.
- On 15% of them, I find Leaked cluster errors that I can correct using
"qemu-img check -r all"
- On 5% of them, I find Leaked clusters errors and further fatal errors,
which can not be corrected with qemu-img.
In rare cases, qemu-img can correct them, but destroys large parts of the
image (becomes unusable), and on other cases it can not correct them at all.


It would be good if you could make the 'qemu-img check' output available
somewhere.


See attachment.



It would be even better if we could have a look at the respective image.
I seem to remember that John (CCed) had a few scripts to analyse
corrupted qcow2 images, maybe we would be able to see something there.


I just exported it like this :
qemu-img convert /dev/the_correct_path /home/blablah.qcow2.img

The resulting file is 32G and I need an idea to transfer this img to you.




What I read similar to my case is :
- usage of qcow2
- heavy disk I/O
- using the virtio-blk driver

In the proxmox thread, they tend to say that using virtio-scsi is the
solution. Having asked this question to oVirt experts
(https://lists.ovirt.org/pipermail/users/2018-February/086753.html) but it's
not clear the driver is to blame.


This seems very unlikely. The corruption you're seeing is in the qcow2
metadata, not only in the guest data.


Are you saying:
- the corruption is in the metadata and in the guest data
OR
- the corruption is only in the metadata
?


If anything, virtio-scsi exercises
more qcow2 code paths than virtio-blk, so any potential bug that affects
virtio-blk should also affect virtio-scsi, but not the other way around.


I get that.




I agree with the answer Yaniv Kaul gave to me, saying I have to properly
report the issue, so I'm longing to know which peculiar information I can
give you now.


To be honest, debugging corruption after the fact is pretty hard. We'd
need the 'qemu-img check' output


Done.


and ideally the image to do anything,


I remember some Redhat people once gave me a temporary access to put 
heavy file on some dedicated server. Is it still possible?



but I can't promise that anything would come out of this.

Best would be a reproducer, or at least some operation that you can link
to the appearance of the corruption. Then we could take a more targeted
look at the respective code.


Sure.
Alas I find no obvious pattern leading to corruption :
From the guest side, it appeared with windows 2003, 2008, 2012, linux 
centOS 6 and 7. It appeared with virtio-blk; and I changed some VMs to 
used virtio-scsi but it's too soon to see appearance of corruption in 
that case.
As I said, I'm using snapshots VERY rarely, and our versions are too old 
so we do them the cold way only (VM shutdown). So very safely.
The "weirdest" thing we do is to migrate VMs : you see how conservative 
we are!



As you can imagine, all this setup is in production, and for most of the
VMs, I can not "play" with them. Moreover, we launched a campaign of nightly
stopping every VM, qemu-img check them one by one, then boot.
So it might take some time before I find another corrupted image.
(which I'll preciously store for debug)

Other informations : We very rarely do snapshots, but I'm close to imagine
that automated migrations of VMs could trigger similar behaviors on qcow2
images.


To my knowledge, oVirt only uses external snapshots and creates them
with QMP. This should be perfectly safe because from the perspective of
the qcow2 image being snapshotted, it just means that it gets no new
write 

Re: [ovirt-users] [Qemu-block] qcow2 images corruption

[Kevin Wolf]

Am 07.02.2018 um 18:06 hat Nicolas Ecarnot geschrieben:
> TL; DR : qcow2 images keep getting corrupted. Any workaround?

Not without knowing the cause.

The first thing to make sure is that the image isn't touched by a second
process while QEMU is running a VM. The classic one is using 'qemu-img
snapshot' on the image of a running VM, which is instant corruption (and
newer QEMU versions have locking in place to prevent this), but we have
seen more absurd cases of things outside QEMU tampering with the image
when we were investigating previous corruption reports.

This covers the majority of all reports, we haven't had a real
corruption caused by a QEMU bug in ages.

> After having found (https://access.redhat.com/solutions/1173623) the right
> logical volume hosting the qcow2 image, I can run qemu-img check on it.
> - On 80% of my VMs, I find no errors.
> - On 15% of them, I find Leaked cluster errors that I can correct using
> "qemu-img check -r all"
> - On 5% of them, I find Leaked clusters errors and further fatal errors,
> which can not be corrected with qemu-img.
> In rare cases, qemu-img can correct them, but destroys large parts of the
> image (becomes unusable), and on other cases it can not correct them at all.

It would be good if you could make the 'qemu-img check' output available
somewhere.

It would be even better if we could have a look at the respective image.
I seem to remember that John (CCed) had a few scripts to analyse
corrupted qcow2 images, maybe we would be able to see something there.

> What I read similar to my case is :
> - usage of qcow2
> - heavy disk I/O
> - using the virtio-blk driver
> 
> In the proxmox thread, they tend to say that using virtio-scsi is the
> solution. Having asked this question to oVirt experts
> (https://lists.ovirt.org/pipermail/users/2018-February/086753.html) but it's
> not clear the driver is to blame.

This seems very unlikely. The corruption you're seeing is in the qcow2
metadata, not only in the guest data. If anything, virtio-scsi exercises
more qcow2 code paths than virtio-blk, so any potential bug that affects
virtio-blk should also affect virtio-scsi, but not the other way around.

> I agree with the answer Yaniv Kaul gave to me, saying I have to properly
> report the issue, so I'm longing to know which peculiar information I can
> give you now.

To be honest, debugging corruption after the fact is pretty hard. We'd
need the 'qemu-img check' output and ideally the image to do anything,
but I can't promise that anything would come out of this.

Best would be a reproducer, or at least some operation that you can link
to the appearance of the corruption. Then we could take a more targeted
look at the respective code.

> As you can imagine, all this setup is in production, and for most of the
> VMs, I can not "play" with them. Moreover, we launched a campaign of nightly
> stopping every VM, qemu-img check them one by one, then boot.
> So it might take some time before I find another corrupted image.
> (which I'll preciously store for debug)
> 
> Other informations : We very rarely do snapshots, but I'm close to imagine
> that automated migrations of VMs could trigger similar behaviors on qcow2
> images.

To my knowledge, oVirt only uses external snapshots and creates them
with QMP. This should be perfectly safe because from the perspective of
the qcow2 image being snapshotted, it just means that it gets no new
write requests.

Migration is something more involved, and if you could relate the
problem to migration, that would certainly be something to look into. In
that case, it would be important to know more about the setup, e.g. is
it migration with shared or non-shared storage?

> Last point about the versions we use : yes that's old, yes we're planning to
> upgrade, but we don't know when.

That would be helpful, too. Nothing is more frustrating that debugging a
bug in an old version only to find that it's already fixed in the
current version (well, except maybe debugging and finding nothing).

Kevin
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users