[one-users] No system datastore with enough capacity form the VM
Carlos Martín Sánchez
cmartin at opennebula.org
Mon Dec 16 07:53:48 PST 2013
Hi,
On Mon, Dec 16, 2013 at 1:21 PM, Mario Giammarco <mgiammarco at gmail.com>
wrote:
> Thank you for your explanation.
> Infact I have already found it by trial and error.
> I usually read documentation (and I have read the big opennebula
> documentation).
> But, as I have already said in an other post, probably because I am new to
> cloud virtualization systems (I am used to standard virtualization like
> vmware), even reading documentation I cannot understand easily several
> concepts:
>
> - difference among persistent/non persistent/volatile;
> - why I need to create a template and I cannot create virtual machines
> directly;
>
> So I wonder if there is a tutorial on using cloud systems or something
> similar.
>
> Thanks again,
> Mario
>
The gist of it is that a disk using a non-persistent image is a copy that
can be modified during the VM lifecycle, and these changes will be
destroyed with the VM.
If the disk is using a persistent image, OpenNebula will save the changes
back to the datastore when the VM is shut down.
You can decide to save the changes made to a disk using a non-persistent
image. In this case, instead of overwriting the original image, OpenNebula
will create a new one.
Volatile disks are not created from any Image, a clean disk is created in
the system DS when the VM is deployed. These disks are destroyed with the
VM, and there is no way to save the changes back to an image datastore [3].
Regards
[1] http://opennebula.org/documentation:rel4.4:template#disks_section
[2]
http://opennebula.org/documentation:rel4.4:img_guide#how_to_use_images_in_virtual_machines
[3]
http://opennebula.org/documentation:rel4.4:img_guide#making_images_persistent
--
Carlos Martín, MSc
Project Engineer
OpenNebula - Flexible Enterprise Cloud Made Simple
www.OpenNebula.org | cmartin at opennebula.org |
@OpenNebula<http://twitter.com/opennebula><cmartin at opennebula.org>
On Mon, Dec 16, 2013 at 1:21 PM, Mario Giammarco <mgiammarco at gmail.com>wrote:
> Thank you for your explanation.
> Infact I have already found it by trial and error.
> I usually read documentation (and I have read the big opennebula
> documentation).
> But, as I have already said in an other post, probably because I am new to
> cloud virtualization systems (I am used to standard virtualization like
> vmware), even reading documentation I cannot understand easily several
> concepts:
>
> - difference among persistent/non persistent/volatile;
> - why I need to create a template and I cannot create virtual machines
> directly;
>
> So I wonder if there is a tutorial on using cloud systems or something
> similar.
>
> Thanks again,
> Mario
>
>
> 2013/12/16 Carlos Martín Sánchez <cmartin at opennebula.org>
>
>> Hi,
>>
>> On Fri, Dec 13, 2013 at 7:04 PM, Mario Giammarco <mgiammarco at gmail.com>
>> wrote:
>>
>> I have done other tests.
>>> If I use a premade image from marketplace all works.
>>> The problem is that I have tried to install an so from scratch (windows)
>>> and I have created a volatile disk.
>>> Probably it is my fault but I still do not understand how to put all my
>>> vm in ceph filesystem.
>>>
>>
>> That's the problem. Volatile disks are created in the system datastore.
>> To install a new OS you need to create an empty datablock persistent image
>> [1], because volatile disks are always disposed when the VM is deleted.
>>
>> Regards
>>
>> [1]
>> http://opennebula.org/documentation:rel4.4:img_guide#making_images_persistent
>> [2] http://opennebula.org/documentation:rel4.4:template#disks_section
>>
>> --
>> Carlos Martín, MSc
>> Project Engineer
>> OpenNebula - Flexible Enterprise Cloud Made Simple
>> www.OpenNebula.org | cmartin at opennebula.org | @OpenNebula<http://twitter.com/opennebula><cmartin at opennebula.org>
>>
>>
>> On Fri, Dec 13, 2013 at 7:04 PM, Mario Giammarco <mgiammarco at gmail.com>wrote:
>>
>>> I have done other tests.
>>> If I use a premade image from marketplace all works.
>>> The problem is that I have tried to install an so from scratch (windows)
>>> and I have created a volatile disk.
>>> Probably it is my fault but I still do not understand how to put all my
>>> vm in ceph filesystem.
>>>
>>>
>>> 2013/12/12 Carlos Martín Sánchez <cmartin at opennebula.org>
>>>
>>>> So the scheduler thinks that the VM needs 100 GB in the system DS
>>>> instead of the ceph image DS.
>>>> Could you please paste the output of onedatastore show 100 & 104 ?
>>>>
>>>> Regards
>>>> --
>>>> Carlos Martín, MSc
>>>> Project Engineer
>>>> OpenNebula - Flexible Enterprise Cloud Made Simple
>>>> www.OpenNebula.org | cmartin at opennebula.org | @OpenNebula<http://twitter.com/opennebula><cmartin at opennebula.org>
>>>>
>>>>
>>>> On Wed, Dec 11, 2013 at 8:53 PM, Mario Giammarco <mgiammarco at gmail.com>wrote:
>>>>
>>>>> Here are some parts of the log, please note that I tried two things:
>>>>>
>>>>> - clean some requirements of the vm (datastore rank stripping);
>>>>> - adding another system datastore (I have one shared and one ssh);
>>>>> - please also note that system datastore with id 0 was never used
>>>>> because it was not part of the cluster or "public";
>>>>> - I have also deleted and recreated VM;
>>>>>
>>>>> Wed Dec 11 10:02:06 2013 [SCHED][I]: Init Scheduler Log system
>>>>> Wed Dec 11 10:02:06 2013 [SCHED][I]: Starting Scheduler Daemon
>>>>> ----------------------------------------
>>>>> Scheduler Configuration File
>>>>> ----------------------------------------
>>>>> DEFAULT_DS_SCHED=POLICY=1
>>>>> DEFAULT_SCHED=POLICY=1
>>>>> HYPERVISOR_MEM=0.1
>>>>> LIVE_RESCHEDS=0
>>>>> LOG=DEBUG_LEVEL=3,SYSTEM=file
>>>>> MAX_DISPATCH=30
>>>>> MAX_HOST=1
>>>>> MAX_VM=5000
>>>>> ONED_PORT=2633
>>>>> SCHED_INTERVAL=30
>>>>> ----------------------------------------
>>>>> Wed Dec 11 10:02:06 2013 [SCHED][I]: Starting scheduler loop...
>>>>> Wed Dec 11 10:02:06 2013 [SCHED][I]: Scheduler loop started.
>>>>> Wed Dec 11 10:08:36 2013 [VM][D]: Pending/rescheduling VM and capacity
>>>>> requirements:
>>>>> VM CPU Memory System DS Image DS
>>>>> ------------------------------------------------------------
>>>>> 4 100 524288 0 DS 100: 0
>>>>> Wed Dec 11 10:08:36 2013 [HOST][D]: Discovered Hosts (enabled):
>>>>> 0
>>>>> Wed Dec 11 10:08:36 2013 [SCHED][D]: VM 4: Datastore 0 filtered out.
>>>>> It does not fulfill SCHED_DS_REQUIREMENTS.
>>>>> Wed Dec 11 10:08:36 2013 [SCHED][I]: Scheduling Results:
>>>>> Virtual Machine: 4
>>>>>
>>>>> PRI ID - HOSTS
>>>>> ------------------------
>>>>> 0 0
>>>>>
>>>>> PRI ID - DATASTORES
>>>>> ------------------------
>>>>> 1 104
>>>>>
>>>>> Wed Dec 11 10:08:36 2013 [VM][I]: Dispatching VM 4 to host 0 and
>>>>> datastore 104
>>>>> Wed Dec 11 10:10:35 2013 [VM][D]: Pending/rescheduling VM and capacity
>>>>> requirements:
>>>>> VM CPU Memory System DS Image DS
>>>>> ------------------------------------------------------------
>>>>> 5 100 524288 0 DS 100: 0
>>>>> Wed Dec 11 10:10:35 2013 [HOST][D]: Discovered Hosts (enabled):
>>>>> 0
>>>>> Wed Dec 11 10:10:35 2013 [SCHED][D]: VM 5: Datastore 0 filtered out.
>>>>> It does not fulfill SCHED_DS_REQUIREMENTS.
>>>>> Wed Dec 11 10:10:35 2013 [SCHED][I]: Scheduling Results:
>>>>> Virtual Machine: 5
>>>>>
>>>>> PRI ID - HOSTS
>>>>> ------------------------
>>>>> 0 0
>>>>>
>>>>> PRI ID - DATASTORES
>>>>> ------------------------
>>>>> 1 104
>>>>>
>>>>>
>>>>> Wed Dec 11 10:10:35 2013 [VM][I]: Dispatching VM 5 to host 0 and
>>>>> datastore 104
>>>>> Wed Dec 11 11:24:20 2013 [VM][D]: Pending/rescheduling VM and capacity
>>>>> requirements:
>>>>> VM CPU Memory System DS Image DS
>>>>> ------------------------------------------------------------
>>>>> 6 100 1048576 102400 DS 100: 0
>>>>> Wed Dec 11 11:24:20 2013 [HOST][D]: Discovered Hosts (enabled):
>>>>> 0
>>>>> Wed Dec 11 11:24:20 2013 [SCHED][D]: VM 6: Datastore 0 filtered out.
>>>>> It does not fulfill SCHED_DS_REQUIREMENTS.
>>>>> Wed Dec 11 11:24:20 2013 [SCHED][D]: VM 6: Datastore 104 filtered out.
>>>>> Not enough capacity.
>>>>> Wed Dec 11 11:24:20 2013 [SCHED][I]: Scheduling Results:
>>>>>
>>>>> Wed Dec 11 11:24:50 2013 [VM][D]: Pending/rescheduling VM and capacity
>>>>> requirements:
>>>>> VM CPU Memory System DS Image DS
>>>>> ------------------------------------------------------------
>>>>> 6 100 1048576 102400 DS 100: 0
>>>>> Wed Dec 11 11:24:50 2013 [HOST][D]: Discovered Hosts (enabled):
>>>>>
>>>>> ...........
>>>>> ...........
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ed Dec 11 20:51:20 2013 [SCHED][I]: Scheduling
>>>>> Results:
>>>>>
>>>>> Virtual Machine:
>>>>> 7
>>>>>
>>>>>
>>>>> PRI ID - HOSTS
>>>>> ------------------------
>>>>> 0 0
>>>>>
>>>>> PRI ID - DATASTORES
>>>>> ------------------------
>>>>> 0 105
>>>>>
>>>>>
>>>>> Wed Dec 11 20:51:20 2013 [SCHED][D]: VM 7: Local Datastore 105 in Host
>>>>> 0 filtered out. Not enough capacity.
>>>>> Wed Dec 11 20:51:20 2013 [SCHED][I]: VM 7: No suitable System DS found
>>>>> for Host: 0. Filtering out host.
>>>>> Wed Dec 11 20:51:50 2013 [VM][D]: Pending/rescheduling VM and capacity
>>>>> requirements:
>>>>> VM CPU Memory System DS Image DS
>>>>> ------------------------------------------------------------
>>>>> 7 100 1048576 102400 DS 100: 0
>>>>> Wed Dec 11 20:51:50 2013 [HOST][D]: Discovered Hosts (enabled):
>>>>> 0
>>>>> Wed Dec 11 20:51:50 2013 [SCHED][D]: VM 7: Datastore 0 filtered out.
>>>>> It does not fulfill SCHED_DS_REQUIREMENTS.
>>>>> Wed Dec 11 20:51:50 2013 [SCHED][D]: VM 7: Datastore 104 filtered out.
>>>>> Not enough capacity.
>>>>> Wed Dec 11 20:51:50 2013 [SCHED][I]: Scheduling Results:
>>>>> Virtual Machine: 7
>>>>>
>>>>> PRI ID - HOSTS
>>>>> ------------------------
>>>>> 0 0
>>>>>
>>>>> PRI ID - DATASTORES
>>>>> ------------------------
>>>>> 0
>>>>> 105
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Wed Dec 11 20:51:50 2013 [SCHED][D]: VM 7: Local Datastore 105 in Host
>>>>> 0 filtered out. Not enough capacity.
>>>>> Wed Dec 11 20:51:50 2013 [SCHED][I]: VM 7: No suitable System DS found
>>>>> for Host: 0. Filtering out host.
>>>>>
>>>>>
>>>>>
>>>>> 2013/12/11 Carlos Martín Sánchez <cmartin at opennebula.org>
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> Could you paste the contents of the /var/log/one/sched.log file?
>>>>>> It should have the amount of space that the VM needs for the system
>>>>>> and image datastores.
>>>>>>
>>>>>> Regards.
>>>>>>
>>>>>> --
>>>>>> Carlos Martín, MSc
>>>>>> Project Engineer
>>>>>> OpenNebula - Flexible Enterprise Cloud Made Simple
>>>>>> www.OpenNebula.org | cmartin at opennebula.org | @OpenNebula<http://twitter.com/opennebula><cmartin at opennebula.org>
>>>>>>
>>>>>>
>>>>>> On Wed, Dec 11, 2013 at 12:31 PM, Mario Giammarco <
>>>>>> mgiammarco at gmail.com> wrote:
>>>>>>
>>>>>>> Hello,
>>>>>>> I have created a 100gb image on ceph.
>>>>>>> My system datastore have few gb free.
>>>>>>> But image is on ceph so it should not check system datastore free
>>>>>>> space!
>>>>>>>
>>>>>>> I will never have a lot of space on system datastore!
>>>>>>>
>>>>>>> Thanks,
>>>>>>> Mario
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Users mailing list
>>>>>>> Users at lists.opennebula.org
>>>>>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20131216/b34d3abf/attachment-0002.htm>
More information about the Users
mailing list