You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@cloudstack.apache.org by Nguyen Anh Tu <ng...@gmail.com> on 2013/05/23 09:56:44 UTC
[Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Hi fork,
I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot and get
this error:
2013-05-23 09:30:40,150 DEBUG [storage.snapshot.SnapshotSchedulerImpl]
(SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for schedule id:
2 at 2013-05-23 02:27:00 GMT
2013-05-23 09:30:40,190 WARN [storage.snapshot.SnapshotSchedulerImpl]
(SnapshotPollTask:null) Scheduling snapshot failed due to
java.lang.NullPointerException
2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
(SnapshotPollTask:null) Was unable to find lock for the key
snapshot_schedule2 and thread id 21407801
I found two same issues on jira. How can we fix it or I miss some new code?
https://issues.apache.org/jira/browse/CLOUDSTACK-2523
https://issues.apache.org/jira/browse/CLOUDSTACK-1353
--
N.g.U.y.e.N.A.n.H.t.U
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Nguyen Anh Tu <ng...@gmail.com>.
Thanks
2013/5/24 Min Chen <mi...@citrix.com>
> FYI. This issue has been fixed in object_store branch.
>
> Thanks
> -min
>
> On 5/23/13 10:16 AM, "Min Chen" <mi...@citrix.com> wrote:
>
> >From the bug comments from Fang and Animesh, snapshot is not officially
> >supported for KVM. Maybe they can shed more light on this?
> >
> >Thanks
> >-min
> >
> >On 5/23/13 10:04 AM, "Chip Childers" <ch...@sungard.com> wrote:
> >
> >>On Thu, May 23, 2013 at 05:03:04PM +0000, Min Chen wrote:
> >>> Sorry, my bad. I misunderstood Deepti's comment and thought that
> >>>recurring
> >>> snapshot is also broken in 4.1.0 for Xenserver. Now I looked closer at
> >>>the
> >>> comment, it refers to master branch. In that case, this should not
> >>>block
> >>> 4.1.0 release then.
> >>>
> >>> -min
> >>
> >>What about Ove's comment below about it being broken in 4.1?
> >>
> >>>
> >>>
> >>> On 5/23/13 9:57 AM, "Ove Ewerlid" <Ov...@oracle.com> wrote:
> >>>
> >>> >On 05/23/2013 06:30 PM, Min Chen wrote:
> >>> >> We also encountered this issue when we ran test on object_store
> >>>branch
> >>> >> before sending merge request, then we realize that this has been
> >>>broken
> >>> >>in
> >>> >> master and consider to fix that after merging object_store branch to
> >>> >> master. But from looking through comments in
> >>> >> https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached
> >>>here,
> >>> >> now I am feeling that recurring snapshot is even broken in 4.1.0
> >>>branch
> >>> >> (see Deepti's comment in the bug). If that is the case, this may
> >>>need to
> >>> >> be considered as blocker for 4.1.
> >>> >
> >>> >It is broken in 4.1 for KVM. Scripts that work with 4.0 fails with 4.1
> >>> >(we see a Null pointer exception when the scheduler kicks in to take
> >>>the
> >>> >snapshot. Manual snapshots work. reproducible in basic local storage
> >>> >setup).
> >>> >
> >>> >Is the scheduled volume snapshot feature officially supported for KVM?
> >>> >
> >>> >/Ove
> >>> >
> >>> >> Thanks
> >>> >> -min
> >>> >>
> >>> >>
> >>> >>
> >>> >> On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
> >>> >>
> >>> >>> Hi fork,
> >>> >>>
> >>> >>> I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot
> >>>and
> >>> >>>get
> >>> >>> this error:
> >>> >>>
> >>> >>> 2013-05-23 09:30:40,150 DEBUG
> >>>[storage.snapshot.SnapshotSchedulerImpl]
> >>> >>> (SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for
> >>>schedule
> >>> >>> id:
> >>> >>> 2 at 2013-05-23 02:27:00 GMT
> >>> >>> 2013-05-23 09:30:40,190 WARN
> >>>[storage.snapshot.SnapshotSchedulerImpl]
> >>> >>> (SnapshotPollTask:null) Scheduling snapshot failed due to
> >>> >>> java.lang.NullPointerException
> >>> >>> 2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
> >>> >>> (SnapshotPollTask:null) Was unable to find lock for the key
> >>> >>> snapshot_schedule2 and thread id 21407801
> >>> >>>
> >>> >>>
> >>> >>> I found two same issues on jira. How can we fix it or I miss some
> >>>new
> >>> >>> code?
> >>> >>>
> >>> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-2523
> >>> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353
> >>> >>>
> >>> >>>
> >>> >>> --
> >>> >>>
> >>> >>> N.g.U.y.e.N.A.n.H.t.U
> >>> >>
> >>> >
> >>> >
> >>> >--
> >>> >Ove Everlid
> >>> >System Administrator / Architect / SDN & Linux hacker
> >>> >Mobile: +46706662363
> >>> >Office: +4618656913 (note EMEA Time Zone)
> >>>
> >>>
> >
>
>
--
N.g.U.y.e.N.A.n.H.t.U
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Min Chen <mi...@citrix.com>.
FYI. This issue has been fixed in object_store branch.
Thanks
-min
On 5/23/13 10:16 AM, "Min Chen" <mi...@citrix.com> wrote:
>From the bug comments from Fang and Animesh, snapshot is not officially
>supported for KVM. Maybe they can shed more light on this?
>
>Thanks
>-min
>
>On 5/23/13 10:04 AM, "Chip Childers" <ch...@sungard.com> wrote:
>
>>On Thu, May 23, 2013 at 05:03:04PM +0000, Min Chen wrote:
>>> Sorry, my bad. I misunderstood Deepti's comment and thought that
>>>recurring
>>> snapshot is also broken in 4.1.0 for Xenserver. Now I looked closer at
>>>the
>>> comment, it refers to master branch. In that case, this should not
>>>block
>>> 4.1.0 release then.
>>>
>>> -min
>>
>>What about Ove's comment below about it being broken in 4.1?
>>
>>>
>>>
>>> On 5/23/13 9:57 AM, "Ove Ewerlid" <Ov...@oracle.com> wrote:
>>>
>>> >On 05/23/2013 06:30 PM, Min Chen wrote:
>>> >> We also encountered this issue when we ran test on object_store
>>>branch
>>> >> before sending merge request, then we realize that this has been
>>>broken
>>> >>in
>>> >> master and consider to fix that after merging object_store branch to
>>> >> master. But from looking through comments in
>>> >> https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached
>>>here,
>>> >> now I am feeling that recurring snapshot is even broken in 4.1.0
>>>branch
>>> >> (see Deepti's comment in the bug). If that is the case, this may
>>>need to
>>> >> be considered as blocker for 4.1.
>>> >
>>> >It is broken in 4.1 for KVM. Scripts that work with 4.0 fails with 4.1
>>> >(we see a Null pointer exception when the scheduler kicks in to take
>>>the
>>> >snapshot. Manual snapshots work. reproducible in basic local storage
>>> >setup).
>>> >
>>> >Is the scheduled volume snapshot feature officially supported for KVM?
>>> >
>>> >/Ove
>>> >
>>> >> Thanks
>>> >> -min
>>> >>
>>> >>
>>> >>
>>> >> On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
>>> >>
>>> >>> Hi fork,
>>> >>>
>>> >>> I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot
>>>and
>>> >>>get
>>> >>> this error:
>>> >>>
>>> >>> 2013-05-23 09:30:40,150 DEBUG
>>>[storage.snapshot.SnapshotSchedulerImpl]
>>> >>> (SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for
>>>schedule
>>> >>> id:
>>> >>> 2 at 2013-05-23 02:27:00 GMT
>>> >>> 2013-05-23 09:30:40,190 WARN
>>>[storage.snapshot.SnapshotSchedulerImpl]
>>> >>> (SnapshotPollTask:null) Scheduling snapshot failed due to
>>> >>> java.lang.NullPointerException
>>> >>> 2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
>>> >>> (SnapshotPollTask:null) Was unable to find lock for the key
>>> >>> snapshot_schedule2 and thread id 21407801
>>> >>>
>>> >>>
>>> >>> I found two same issues on jira. How can we fix it or I miss some
>>>new
>>> >>> code?
>>> >>>
>>> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-2523
>>> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353
>>> >>>
>>> >>>
>>> >>> --
>>> >>>
>>> >>> N.g.U.y.e.N.A.n.H.t.U
>>> >>
>>> >
>>> >
>>> >--
>>> >Ove Everlid
>>> >System Administrator / Architect / SDN & Linux hacker
>>> >Mobile: +46706662363
>>> >Office: +4618656913 (note EMEA Time Zone)
>>>
>>>
>
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Min Chen <mi...@citrix.com>.
>From the bug comments from Fang and Animesh, snapshot is not officially
supported for KVM. Maybe they can shed more light on this?
Thanks
-min
On 5/23/13 10:04 AM, "Chip Childers" <ch...@sungard.com> wrote:
>On Thu, May 23, 2013 at 05:03:04PM +0000, Min Chen wrote:
>> Sorry, my bad. I misunderstood Deepti's comment and thought that
>>recurring
>> snapshot is also broken in 4.1.0 for Xenserver. Now I looked closer at
>>the
>> comment, it refers to master branch. In that case, this should not block
>> 4.1.0 release then.
>>
>> -min
>
>What about Ove's comment below about it being broken in 4.1?
>
>>
>>
>> On 5/23/13 9:57 AM, "Ove Ewerlid" <Ov...@oracle.com> wrote:
>>
>> >On 05/23/2013 06:30 PM, Min Chen wrote:
>> >> We also encountered this issue when we ran test on object_store
>>branch
>> >> before sending merge request, then we realize that this has been
>>broken
>> >>in
>> >> master and consider to fix that after merging object_store branch to
>> >> master. But from looking through comments in
>> >> https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached
>>here,
>> >> now I am feeling that recurring snapshot is even broken in 4.1.0
>>branch
>> >> (see Deepti's comment in the bug). If that is the case, this may
>>need to
>> >> be considered as blocker for 4.1.
>> >
>> >It is broken in 4.1 for KVM. Scripts that work with 4.0 fails with 4.1
>> >(we see a Null pointer exception when the scheduler kicks in to take
>>the
>> >snapshot. Manual snapshots work. reproducible in basic local storage
>> >setup).
>> >
>> >Is the scheduled volume snapshot feature officially supported for KVM?
>> >
>> >/Ove
>> >
>> >> Thanks
>> >> -min
>> >>
>> >>
>> >>
>> >> On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
>> >>
>> >>> Hi fork,
>> >>>
>> >>> I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot
>>and
>> >>>get
>> >>> this error:
>> >>>
>> >>> 2013-05-23 09:30:40,150 DEBUG
>>[storage.snapshot.SnapshotSchedulerImpl]
>> >>> (SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for
>>schedule
>> >>> id:
>> >>> 2 at 2013-05-23 02:27:00 GMT
>> >>> 2013-05-23 09:30:40,190 WARN
>>[storage.snapshot.SnapshotSchedulerImpl]
>> >>> (SnapshotPollTask:null) Scheduling snapshot failed due to
>> >>> java.lang.NullPointerException
>> >>> 2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
>> >>> (SnapshotPollTask:null) Was unable to find lock for the key
>> >>> snapshot_schedule2 and thread id 21407801
>> >>>
>> >>>
>> >>> I found two same issues on jira. How can we fix it or I miss some
>>new
>> >>> code?
>> >>>
>> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-2523
>> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353
>> >>>
>> >>>
>> >>> --
>> >>>
>> >>> N.g.U.y.e.N.A.n.H.t.U
>> >>
>> >
>> >
>> >--
>> >Ove Everlid
>> >System Administrator / Architect / SDN & Linux hacker
>> >Mobile: +46706662363
>> >Office: +4618656913 (note EMEA Time Zone)
>>
>>
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Chip Childers <ch...@sungard.com>.
On Thu, May 23, 2013 at 05:03:04PM +0000, Min Chen wrote:
> Sorry, my bad. I misunderstood Deepti's comment and thought that recurring
> snapshot is also broken in 4.1.0 for Xenserver. Now I looked closer at the
> comment, it refers to master branch. In that case, this should not block
> 4.1.0 release then.
>
> -min
What about Ove's comment below about it being broken in 4.1?
>
>
> On 5/23/13 9:57 AM, "Ove Ewerlid" <Ov...@oracle.com> wrote:
>
> >On 05/23/2013 06:30 PM, Min Chen wrote:
> >> We also encountered this issue when we ran test on object_store branch
> >> before sending merge request, then we realize that this has been broken
> >>in
> >> master and consider to fix that after merging object_store branch to
> >> master. But from looking through comments in
> >> https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached here,
> >> now I am feeling that recurring snapshot is even broken in 4.1.0 branch
> >> (see Deepti's comment in the bug). If that is the case, this may need to
> >> be considered as blocker for 4.1.
> >
> >It is broken in 4.1 for KVM. Scripts that work with 4.0 fails with 4.1
> >(we see a Null pointer exception when the scheduler kicks in to take the
> >snapshot. Manual snapshots work. reproducible in basic local storage
> >setup).
> >
> >Is the scheduled volume snapshot feature officially supported for KVM?
> >
> >/Ove
> >
> >> Thanks
> >> -min
> >>
> >>
> >>
> >> On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
> >>
> >>> Hi fork,
> >>>
> >>> I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot and
> >>>get
> >>> this error:
> >>>
> >>> 2013-05-23 09:30:40,150 DEBUG [storage.snapshot.SnapshotSchedulerImpl]
> >>> (SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for schedule
> >>> id:
> >>> 2 at 2013-05-23 02:27:00 GMT
> >>> 2013-05-23 09:30:40,190 WARN [storage.snapshot.SnapshotSchedulerImpl]
> >>> (SnapshotPollTask:null) Scheduling snapshot failed due to
> >>> java.lang.NullPointerException
> >>> 2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
> >>> (SnapshotPollTask:null) Was unable to find lock for the key
> >>> snapshot_schedule2 and thread id 21407801
> >>>
> >>>
> >>> I found two same issues on jira. How can we fix it or I miss some new
> >>> code?
> >>>
> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-2523
> >>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353
> >>>
> >>>
> >>> --
> >>>
> >>> N.g.U.y.e.N.A.n.H.t.U
> >>
> >
> >
> >--
> >Ove Everlid
> >System Administrator / Architect / SDN & Linux hacker
> >Mobile: +46706662363
> >Office: +4618656913 (note EMEA Time Zone)
>
>
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Min Chen <mi...@citrix.com>.
Sorry, my bad. I misunderstood Deepti's comment and thought that recurring
snapshot is also broken in 4.1.0 for Xenserver. Now I looked closer at the
comment, it refers to master branch. In that case, this should not block
4.1.0 release then.
-min
On 5/23/13 9:57 AM, "Ove Ewerlid" <Ov...@oracle.com> wrote:
>On 05/23/2013 06:30 PM, Min Chen wrote:
>> We also encountered this issue when we ran test on object_store branch
>> before sending merge request, then we realize that this has been broken
>>in
>> master and consider to fix that after merging object_store branch to
>> master. But from looking through comments in
>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached here,
>> now I am feeling that recurring snapshot is even broken in 4.1.0 branch
>> (see Deepti's comment in the bug). If that is the case, this may need to
>> be considered as blocker for 4.1.
>
>It is broken in 4.1 for KVM. Scripts that work with 4.0 fails with 4.1
>(we see a Null pointer exception when the scheduler kicks in to take the
>snapshot. Manual snapshots work. reproducible in basic local storage
>setup).
>
>Is the scheduled volume snapshot feature officially supported for KVM?
>
>/Ove
>
>> Thanks
>> -min
>>
>>
>>
>> On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
>>
>>> Hi fork,
>>>
>>> I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot and
>>>get
>>> this error:
>>>
>>> 2013-05-23 09:30:40,150 DEBUG [storage.snapshot.SnapshotSchedulerImpl]
>>> (SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for schedule
>>> id:
>>> 2 at 2013-05-23 02:27:00 GMT
>>> 2013-05-23 09:30:40,190 WARN [storage.snapshot.SnapshotSchedulerImpl]
>>> (SnapshotPollTask:null) Scheduling snapshot failed due to
>>> java.lang.NullPointerException
>>> 2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
>>> (SnapshotPollTask:null) Was unable to find lock for the key
>>> snapshot_schedule2 and thread id 21407801
>>>
>>>
>>> I found two same issues on jira. How can we fix it or I miss some new
>>> code?
>>>
>>> https://issues.apache.org/jira/browse/CLOUDSTACK-2523
>>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353
>>>
>>>
>>> --
>>>
>>> N.g.U.y.e.N.A.n.H.t.U
>>
>
>
>--
>Ove Everlid
>System Administrator / Architect / SDN & Linux hacker
>Mobile: +46706662363
>Office: +4618656913 (note EMEA Time Zone)
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Ove Ewerlid <Ov...@oracle.com>.
On 05/23/2013 06:30 PM, Min Chen wrote:
> We also encountered this issue when we ran test on object_store branch
> before sending merge request, then we realize that this has been broken in
> master and consider to fix that after merging object_store branch to
> master. But from looking through comments in
> https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached here,
> now I am feeling that recurring snapshot is even broken in 4.1.0 branch
> (see Deepti's comment in the bug). If that is the case, this may need to
> be considered as blocker for 4.1.
It is broken in 4.1 for KVM. Scripts that work with 4.0 fails with 4.1
(we see a Null pointer exception when the scheduler kicks in to take the
snapshot. Manual snapshots work. reproducible in basic local storage setup).
Is the scheduled volume snapshot feature officially supported for KVM?
/Ove
> Thanks
> -min
>
>
>
> On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
>
>> Hi fork,
>>
>> I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot and get
>> this error:
>>
>> 2013-05-23 09:30:40,150 DEBUG [storage.snapshot.SnapshotSchedulerImpl]
>> (SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for schedule
>> id:
>> 2 at 2013-05-23 02:27:00 GMT
>> 2013-05-23 09:30:40,190 WARN [storage.snapshot.SnapshotSchedulerImpl]
>> (SnapshotPollTask:null) Scheduling snapshot failed due to
>> java.lang.NullPointerException
>> 2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
>> (SnapshotPollTask:null) Was unable to find lock for the key
>> snapshot_schedule2 and thread id 21407801
>>
>>
>> I found two same issues on jira. How can we fix it or I miss some new
>> code?
>>
>> https://issues.apache.org/jira/browse/CLOUDSTACK-2523
>> https://issues.apache.org/jira/browse/CLOUDSTACK-1353
>>
>>
>> --
>>
>> N.g.U.y.e.N.A.n.H.t.U
>
--
Ove Everlid
System Administrator / Architect / SDN & Linux hacker
Mobile: +46706662363
Office: +4618656913 (note EMEA Time Zone)
Re: [Urgent] Scheduling snapshot failed on CS 4.2.0-snapshot
Posted by Min Chen <mi...@citrix.com>.
We also encountered this issue when we ran test on object_store branch
before sending merge request, then we realize that this has been broken in
master and consider to fix that after merging object_store branch to
master. But from looking through comments in
https://issues.apache.org/jira/browse/CLOUDSTACK-1353 you attached here,
now I am feeling that recurring snapshot is even broken in 4.1.0 branch
(see Deepti's comment in the bug). If that is the case, this may need to
be considered as blocker for 4.1.
Thanks
-min
On 5/23/13 12:56 AM, "Nguyen Anh Tu" <ng...@gmail.com> wrote:
>Hi fork,
>
>I'm using CS 4.2.0-snapshot + Xen XCP. I set a recurring snapshot and get
>this error:
>
>2013-05-23 09:30:40,150 DEBUG [storage.snapshot.SnapshotSchedulerImpl]
>(SnapshotPollTask:null) Scheduling 1 snapshot for volume 7 for schedule
>id:
>2 at 2013-05-23 02:27:00 GMT
>2013-05-23 09:30:40,190 WARN [storage.snapshot.SnapshotSchedulerImpl]
>(SnapshotPollTask:null) Scheduling snapshot failed due to
>java.lang.NullPointerException
>2013-05-23 09:30:40,191 WARN [utils.db.Merovingian2]
>(SnapshotPollTask:null) Was unable to find lock for the key
>snapshot_schedule2 and thread id 21407801
>
>
>I found two same issues on jira. How can we fix it or I miss some new
>code?
>
>https://issues.apache.org/jira/browse/CLOUDSTACK-2523
>https://issues.apache.org/jira/browse/CLOUDSTACK-1353
>
>
>--
>
>N.g.U.y.e.N.A.n.H.t.U