Derek, thank you for participating.
In the end, I completed the task (OVA imported).
The root of the problem is that I tried to use the import with thin provision.
But there is a known issue https://bugzilla.redhat.com/show_bug.cgi?id=1382404
Workaround: Import VM from OVA as preallocated
It's not a very good solution, but it is better than nothing
14.12.2016, 17:50, "Derek Atkins" <derek@ihtfp.com>:
> Hi Aleksey,
>
> One more question for debugging purposes: How long does the import go
> before it dies? Do you have enough time to run:
>
> ps aux | grep virt-v2v
>
> On the import host while it's running? This might help us determine
> where it's trying to store the data. It should be storing it in the
> target storage, but it's possible that it's using temp space and then
> running out.
>
> -derek
>
> aleksey.maksimov@it-kb.ru writes:
>
>> Hi Shahar.
>>
>> Look at the attached screenshot
>>
>> 14.12.2016, 12:15, "Shahar Havivi" <shavivi@redhat.com>:
>>
>> Hi,
>> I was able to import your VMs with storage domain that had 350GB free.
>> Your ova have a disk with 1GB actual size and 256 virtual size - when I
>> try to import to a storage with 50G I got the error that you had.
>>
>> if you will look at /var/log/vdsm/import/... you will see the logs of each
>> import,
>> In the one that fail I found this line:
>> qemu-img: error while writing sector 423174528: No space left on device
>>
>> virt-v2v try to convert via qemu-img the img from vmdk to qcow and
>> encountered free space issue in the storage domain.
>>
>> Please consider asking the mailing list libguestfs@redhat.com about this
>> issue or try to increase your storage domain if you want a quick fix.
>>
>> The reason that I think you where able to import via the virt-v2v command
>> is the usage of 'virt-v2v -o null' which is not writing to the disk.
>>
>> Shahar.
>>
>> On Tue, Dec 13, 2016 at 12:54 PM, <aleksey.maksimov@it-kb.ru> wrote:
>>
>> Engine - oVirt Engine Version: 4.0.5.5-1.el7.centos (CentOS 7.2)
>>
>> All Hosts:
>> OS Version:RHEL - 7 - 2.1511.el7.centos.2.10
>> OS Description:CentOS Linux 7 (Core)
>> Kernel Version:3.10.0 - 327.36.3.el7.x86_64
>> KVM Version:2.3.0 - 31.el7.16.1
>> LIBVIRT Version:libvirt-2.0.0-10.el7_3.2
>> VDSM Version:vdsm-4.18.15.3-1.el7.centos
>> SPICE Version:0.12.4 - 15.el7_2.2
>> GlusterFS Version:[N/A]
>> CEPH Version:librbd1-0.80.7-3.el7
>>
>> [root@KOM-AD01-VM31 ~]# virt-v2v -V
>> virt-v2v 1.28.1
>>
>> 13.12.2016, 13:20, "Shahar Havivi" <shavivi@redhat.com>:
>>
>> version of Engine and Host and virt-v2v (which is running on your
>> host)
>> and are you running on Fedora, Centos ext?
>>
>> On Tue, Dec 13, 2016 at 12:08 PM, <aleksey.maksimov@it-kb.ru>
>> wrote:
>>
>> 1. No. This is an unacceptable option for me
>> 2. No. This is my first experience
>> 3. Versions where? On Engine or on Host ?
>> 4. Yes.
>>
>> 13.12.2016, 13:03, "Shahar Havivi" <shavivi@redhat.com>:
>>
>> Thanks you,
>> Several questions:
>> 1. did you try to import the disk target as preallocate?
>> 2. did you try to import other ova files?
>> 3. can you please send us version of virt-v2v, vdsm and
>> engine and the os that you are running with.
>> 4. Is it possible for you to share the ova file?
>>
>> Thank you,
>> Shahar.
>>
>> On Tue, Dec 13, 2016 at 11:23 AM, <
>> aleksey.maksimov@it-kb.ru > wrote:
>>
>> Log attached
>>
>> 13.12.2016, 11:32, "Shahar Havivi" <shavivi@redhat.com
>> >:
>>
>>Thank you,
>>We need the vdsm log as well please - its the host
>>that you try to import the ova from under /var/log
>> /vdsm/vdsm.log
>>
>>Shahar.
>>
>>On Tue, Dec 13, 2016 at 9:58 AM, <
>> aleksey.maksimov@it-kb.ru > wrote:
>>
>>Log attached
>>
>>13.12.2016, 10:04, "Shahar Havivi" <
>>shavivi@redhat.com>:
>>
>>Please attach the log and will see what
>>the error is.
>>
>>On Tue, Dec 13, 2016 at 9:01 AM, <
>>aleksey.maksimov@it-kb. ru > wrote:
>>
>>Preallocated ?
>>I'm trying to import image of virtual
>>appliance "HP 3PAR Virtual Service
>>Processor"
>>This virtual machine has drive 256GB.
>>But really, this is a 40GB disk never
>>grows. I don't want to waste that much
>>disk space.
>>
>>There is no other way ?
>>
>>13.12.2016, 09:53, "Shahar Havivi" <
>>shavivi@redhat.com> :
>>
>>Ok
>>so try to import as preallocated and
>>attach the log if you have any errors.
>>
>>Shahar.
>>
>>On Tue, Dec 13, 2016 at 8:47 AM, <
>>aleksey.maksimov@ it-kb.ru > wrote:
>>
>>All my Data-Domains - FC SAN. Not
>>ISCSI.
>>
>>13.12.2016, 09:41, "Shahar Havivi" <
>>shavivi@redhat.com> :
>>
>>What is the type of the target storage
>>domain that you try to import to?(nfs
>>or iscsi)
>>if its iscsi try to change it to
>>preallocate instead of thin-provision,
>>also please attach the engine log if
>>you have more problems.
>>
>>Shahar.
>>
>>On Tue, Dec 13, 2016 at 7:08 AM, <
>>aleksey.maksimov@ it-kb.ru > wrote:
>>
>># LIBGUESTFS_BACKEND=direct virt-v2v
>>-v -x -i ova /tmp/
>>rhel_Vsp-4.4.0.GA- 53.ova -of raw -o
>>null
>>
>>....
>>fsync /dev/sda
>>guestfsd: main_loop: proc 282
>>(internal_autosync) took 0.01 seconds
>>libguestfs: trace: internal_autosync =
>>0
>>libguestfs: sending SIGTERM to process
>>3338
>>libguestfs: trace: shutdown = 0
>>libguestfs: trace: close
>>libguestfs: closing guestfs handle
>>0x2880a10 (state 0)
>>libguestfs: command: run: rm
>>libguestfs: command: run: \ -rf /tmp/
>>libguestfsrxuByo
>>[ 99.0] Checking if the guest needs
>>BIOS or UEFI to boot
>>[ 99.0] Copying disk 1/1 to /var/tmp/
>>null.lP6vGa/sda (raw)
>>target_file = /var/tmp/null.lP6vGa/sda
>>target_format = raw
>>target_estimated_ size = 13480052354
>>target_overlay = /var/tmp/
>>v2vovle81937.qcow2
>>target_overlay.ov_ source = /var/tmp/
>>ova.zJ6ks8/
>>rhel_Vsp-4.4.0.GA- 53-disk1.vmdk
>>qemu-img convert -p -n -f qcow2 -O
>>'raw' '/var/tmp/v2vovle81937.qcow2' '/
>>var/tmp/null. lP6vGa/sda'
>>(100.00/100%)
>>virtual copying rate: 95364.1 M bits/
>>sec
>>real copying rate: 811.7 M bits/sec
>>sda: estimate 13480052354 (12.6G)
>>versus actual 2339610624 (2.2G):
>>476.2%
>>[ 126.0] Creating output metadata
>>[ 126.0] Finishing off
>>
>>The output is very voluminous. Errors
>>are seen.
>>
>>13.12.2016, 07:59, "
>>aleksey.maksimov@ it-kb.ru " <
>>aleksey.maksimov@ it-kb.ru >:
>>> -bash: annotate-output: command not
>>found
>>>
>>> 12.12.2016, 21:26, "Derek Atkins" <
>>derek@ihtfp.com>:
>>>> Hi,
>>>>
>>>> What happened when you ran the
>>virt-v2v command I suggested?
>>>>
>>>> -derek
>>>>
>>>> On Mon, December 12, 2016 12:32
>>pm, aleksey.maksimov@it-kb.ru wrote:
>>>>> I mount 5TB nfs-share to my host
>>and tried to import OVA from this
>>>>> mount-point.
>>>>> Exact same error.
>>>>>
>>>>> Do the import process attempts
>>to use 256G inside the Engine virtual
>>>>> machine ???
>>>>>
>>>>> 12.12.2016, 18:01, "
>>aleksey.maksimov@ it-kb.ru "
>>>>> <aleksey.maksimov@it-kb.ru>:
>>>>>> 1.2G is real OVA image size.
>>>>>> 256G is virtual disk size.
>>>>>> Could this be a problem?
>>>>>>
>>>>>> But before I start the import,
>>in the settings I choose "Thin
>>provision"
>>>>>> (see the attached screenshot)
>>>>>>
>>>>>> 12.12.2016, 17:47, "
>>aleksey.maksimov@ it-kb.ru "
>>>>>> <aleksey.maksimov@it-kb.ru>:
>>>>>>> Hi Derek
>>>>>>>
>>>>>>> # df -H /tmp/
>>>>>>>
>>>>>>> Filesystem Size Used Avail
>>Use% Mounted on
>>>>>>> /dev/cciss/c0d0p2 53G 5.7G
>>45G 12% /
>>>>>>>
>>>>>>> But my OVA file size - 1.2G
>>>>>>>
>>>>>>> 12.12.2016, 17:26, "Derek
>>Atkins" <derek@ihtfp.com>:
>>>>>>>> Hi,
>>>>>>>>
>>>>>>>> Did you perhaps run out of
>>disk space in /tmp? I'm not sure where
>>>>>>>> the
>>>>>>>> virt-v2v does it work, but
>>the error "Conversion of VM from
>>exteral
>>>>>>>> enironment failed:
>>copy-disk stream closed unexpectedly"
>>seems to
>>>>>>>> imply an
>>>>>>>> error there.
>>>>>>>>
>>>>>>>> I had no issues myself on
>>4.0.5 (modulo setfiles taking a very
>>long
>>>>>>>> time
>>>>>>>> in a few cases).
>>>>>>>>
>>>>>>>> -derek
>>>>>>>>
>>>>>>>> On Mon, December 12, 2016
>>9:19 am, aleksey.maksimov@it-kb.ru
>>wrote:
>>>>>>>>> Hello oVirt guru`s!
>>>>>>>>>
>>>>>>>>> In the process of
>>importing from OVA file (from
>>>>>>>>> /tmp/
>>rhel_Vsp-4.4.0.GA- 53.ova in one of
>>hosts) VM first appears in
>>>>>>>>> the web
>>>>>>>>> console, then disappears.
>>>>>>>>>
>>>>>>>>> In the oVirt Engine log:
>>>>>>>>>
>>>>>>>>> ...
>>>>>>>>> 2016-12-12 16:55:20,209
>>INFO
>>>>>>>>>
>>[org.ovirt.engine. core.bll.exportimport. ImportVmFromOvaCommand]
>>>>>>>>> (default
>>>>>>>>> task-28) [6179021a] Lock
>>Acquired to object
>>>>>>>>> 'EngineLock:
>>{exclusiveLocks='[ rhel_Vsp-4.4.0.GA-53
>>=<VM_NAME,
>>>>>>>>>
>>ACTION_TYPE_ FAILED_NAME_ALREADY_USED>,
>>>>>>>>>
>>2a3d20f9-502a- 4c60-8c14-371456cd4c99
>>=<VM,
>>>>>>>>>
>>ACTION_TYPE_ FAILED_VM_IS_BEING_IMPORTED$ VmName
>>>>>>>>> rhel_Vsp-4.4.0.GA-53>]',
>>>>>>>>> sharedLocks='null'}'
>>>>>>>>> 2016-12-12 16:55:20,326
>>WARN
>>>>>>>>>
>>[org.ovirt.engine. core.dal.job. ExecutionMessageDirector]
>>>>>>>>>
>>(org.ovirt.thread. pool-6-thread-35) [
>>6179021a] The message key
>>>>>>>>> 'ImportVmFromOva' is
>>missing from 'bundles/
>>ExecutionMessages'
>>>>>>>>> ...
>>>>>>>>> 2016-12-12 16:57:34,758
>>INFO
>>>>>>>>>
>>[org.ovirt.engine. core.bll.exportimport. ConvertVmCallback]
>>>>>>>>> (DefaultQuartzScheduler4)
>>[27df05b8] Conversion of VM from
>>exteral
>>>>>>>>> enironment failed:
>>copy-disk stream closed unexpectedly
>>>>>>>>> 2016-12-12 16:57:35,761
>>ERROR
>>>>>>>>>
>>[org.ovirt.engine. core.bll.exportimport. ConvertOvaCommand]