[Sheepdog] convert raw sheepdog store file into an image?

icez network icez at icez.net
Thu Dec 15 22:31:53 CET 2011


when i shutdown the whole cluster and run only on server 172.16.0.12
it said the following when try 'collie vdi list'

Failed to read object 80cdba8600000000 Waiting for other nodes to join cluster
Failed to read inode header
Failed to read object 80cdbc3900000000 Waiting for other nodes to join cluster
Failed to read inode header
Failed to read object 80d0d92800000000 Waiting for other nodes to join cluster
Failed to read inode header
Failed to read object 80d0dc8e00000000 Waiting for other nodes to join cluster
Failed to read inode header
Failed to read object 80e5729a00000000 Waiting for other nodes to join cluster
Failed to read inode header
Failed to read object 80fa0ac800000000 Waiting for other nodes to join cluster
Failed to read inode header
Failed to read object 80fe898600000000 Waiting for other nodes to join cluster
Failed to read inode header


$ collie cluster info
Cluster status: Waiting for other nodes to join cluster


On Fri, Dec 16, 2011 at 04:27, icez network <icez at icez.net> wrote:
> i've tested with another machine.
> when i run collie cluster format again, all data is loss. so it
> doesn't probably the way i want to
>
>
> ' collie cluster info'  on 172.16.0.11 is the same result as .12
>
> Cluster status: running
>
> Cluster created at Thu Jan  1 07:00:00 1970
>
> Epoch Time           Version
> 2011-12-15 19:52:57     21 [172.16.0.11:7000, 172.16.0.12:7000]
> 2011-12-15 19:46:26     20 [172.16.0.12:7000]
> 2011-12-15 19:42:13     19 [172.16.0.11:7000, 172.16.0.12:7000]
> 2011-12-15 19:41:03     18 [172.16.0.12:7000]
> 2011-12-15 18:38:35     17 [172.16.0.11:7000, 172.16.0.12:7000]
> 2011-12-15 18:35:09     16 [172.16.0.12:7000]
> 1970-01-01 07:00:00     15 [172.16.0.11:7000, 172.16.0.12:7000]
> 1970-01-01 07:00:00     14 [172.16.0.12:7000]
>
>
> On Fri, Dec 16, 2011 at 04:14, icez network <icez at icez.net> wrote:
>> is it safe to run cluster format again ? will it make my data lost?
>>
>> On Fri, Dec 16, 2011 at 04:10, MORITA Kazutaka
>> <morita.kazutaka at lab.ntt.co.jp> wrote:
>>> At Fri, 16 Dec 2011 03:40:28 +0700,
>>> icez network wrote:
>>>>
>>>> i didn't specify -H, how can i add it?
>>>
>>>  $ collie cluster format --copies=2 -H
>>>
>>> See also 'collie cluster format -h'.
>>>
>>>>
>>>>
>>>> Cluster status: running
>>>>
>>>> Cluster created at Thu Jan  1 07:00:00 1970
>>>>
>>>> Epoch Time           Version
>>>> 2011-12-15 19:52:57     21 [172.16.0.11:7000, 172.16.0.12:7000]
>>>> 2011-12-15 19:46:26     20 [172.16.0.12:7000]
>>>> 2011-12-15 19:42:13     19 [172.16.0.11:7000, 172.16.0.12:7000]
>>>> 2011-12-15 19:41:03     18 [172.16.0.12:7000]
>>>> 2011-12-15 18:38:35     17 [172.16.0.11:7000, 172.16.0.12:7000]
>>>> 2011-12-15 18:35:09     16 [172.16.0.12:7000]
>>>> 1970-01-01 07:00:00     15 [172.16.0.11:7000, 172.16.0.12:7000]
>>>> 1970-01-01 07:00:00     14 [172.16.0.12:7000]
>>>
>>> It seems that Sheepdog is running correctly.
>>>
>>> Did you run 'collie cluster info' on the node 172.16.0.12?  Can you
>>> try it on 172.16.0.11, too?  And what is the output of 'collie vdi list'?
>>>
>>> I guess you Sheepdog will work if you stop both sheeps and restart the
>>> sheep only on 172.16.0.12.
>>>
>>> Thanks,
>>>
>>> Kazutaka
>>>
>>>>
>>>>
>>>> On Fri, Dec 16, 2011 at 03:23, MORITA Kazutaka
>>>> <morita.kazutaka at lab.ntt.co.jp> wrote:
>>>> >
>>>> > At Thu, 15 Dec 2011 22:33:38 +0700,
>>>> > icez network wrote:
>>>> > > my cluster is running with 2 nodes and formatted with --copies=2
>>>> > > so the file seems to equally match on both server.
>>>> >
>>>> > Did you specify a '-H' option when formmating?  Sheepdog stops working
>>>> > when the number of node is less than the data redundancy by default.
>>>> >
>>>> > >
>>>> > > all files still exists
>>>> > >
>>>> > > collie vdi list : can list all the vdi
>>>> > > but when i try to start the vm or do anything with vdi it said vdi not
>>>> > > found.
>>>> >
>>>> > Can you give us the output of 'collie cluster info'?
>>>> >
>>>> > Thanks,
>>>> >
>>>> > Kazutaka
>>>> >
>>>> > >
>>>> > >
>>>> > >
>>>> > > On Thu, Dec 15, 2011 at 21:38, icez network <worldicez at gmail.com> wrote:
>>>> > >
>>>> > > > currently i have a problem with my sheepdog cluster.
>>>> > > > one of my sheepdog node failed and cannot be recover back.
>>>> > > > but the raw sheepdog store file still exists.
>>>> > > > how can i restore them back or get data from these file?
>>>> > > >
>>>> > > > thank
>>>> > > >
>>>> > > >
>>>> > > > --
>>>> > > > Personal hosting by icez network
>>>> > > > http://www.thzhost.com
>>>> > > >
>>>> > >
>>>> > >
>>>> > >
>>>> > > --
>>>> > > Personal hosting by icez network
>>>> > > http://www.thzhost.com
>>>>
>>>> --
>>>> Personal hosting by icez network
>>>> http://www.thzhost.com
>>>> --
>>>> sheepdog mailing list
>>>> sheepdog at lists.wpkg.org
>>>> http://lists.wpkg.org/mailman/listinfo/sheepdog
>>
>>
>>
>> --
>> Personal hosting by icez network
>> http://www.thzhost.com
>
>
>
> --
> Personal hosting by icez network
> http://www.thzhost.com



-- 
Personal hosting by icez network
http://www.thzhost.com



More information about the sheepdog mailing list