[one-users] glusterfs datastore

Wed May 7 08:29:49 PDT 2014

Am 07.05.14 16:56, schrieb Thomas Stein:
> Am 07.05.14 14:41, schrieb Javier Fontan:
>> GlusterFS integration needs the datastores tm type shared and be
>> mounted in all hosts. The symlinks should also be done in the nodes.
>> In case of GlusterFS both datastores (system and images) point to the
>> same gluster volume.
> 
> Yeah. In gerneral it works now. Now KVM/Qemu seems to lack gluster support.
> 
> kvm -drive file=glus
> ter://gluster01:24007/one-volume1/62/disk.0,if=none,id=drive-ide0-0-0,format=raw,cache=none:
> could not open disk image gluster://
> gluster01:24007/one-volume1/62/disk.0: No such file or directory
> 
> It's Debian7. Someone has an idea how to solve this?

Do i need to build qemu myself or does someone know of a debian
repository with glusterfs enabled qemu package?

cheers
t.

> best regards
> t.
> 
>> On Wed, May 7, 2014 at 1:34 PM, Thomas Stein <himbeere at meine-oma.de> wrote:
>>> Am 07.05.14 13:20, schrieb Javier Fontan:
>>>> The error tells the drivers can not find the file
>>>> /var/lib/one//datastores/146/efe78e0d61224a0b459e6cc423b789a5 in the
>>>> remote node (one2). When using shared TM drivers the datastore should
>>>> be swared between all nodes (including frontend) in the very same
>>>> location. That is, /var/lib/one//datastores/146 must be present in all
>>>> nodes with the shared storage mounted there.
>>>
>>> Thanks for your patience Javier. I sure i get this running soon.
>>>
>>> Problem now is when the system datastore is "shared" i have the same
>>> problem. With "ssh" the image is copied to the node. Shall i set the
>>> system datastore set back to "shared" and create symlinks one the nodes?
>>>
>>> Question is. Do all datastore need to be one the mounted gluster volume?
>>>
>>> best regards
>>> Thomas
>>>
>>>
>>>> On Wed, May 7, 2014 at 1:00 PM, Thomas Stein <himbeere at meine-oma.de> wrote:
>>>>> Am 07.05.14 12:35, schrieb Javier Fontan:
>>>>>> There is no datastore driver called "shared", that is for TM drivers.
>>>>>> For glusterfs the correct datastore driver is fs. CHange it and that
>>>>>> should solve the problem you're having.
>>>>>
>>>>> Thanks for the answer. I think the docu has to be adjusted.
>>>>>
>>>>> https://github.com/OpenNebula/docs/blob/master/source/administration/storage/gluster_ds.rst
>>>>>
>>>>> The first error is gone but now i get:
>>>>>
>>>>> Wed May 7 12:58:13 2014 [DiM][I]: New VM state is ACTIVE.
>>>>> Wed May 7 12:58:13 2014 [LCM][I]: New VM state is PROLOG.
>>>>> Wed May 7 12:58:14 2014 [TM][I]: Command execution fail:
>>>>> /var/lib/one/remotes/tm/shared/clone
>>>>> opennebula-frontend:/var/lib/one//datastores/146/efe78e0d61224a0b459e6cc423b789a5
>>>>> one2:/var/lib/one/datastores/143/55/disk.0 55 146
>>>>> Wed May 7 12:58:14 2014 [TM][I]: clone: Cloning
>>>>> /var/lib/one/datastores/146/efe78e0d61224a0b459e6cc423b789a5 in
>>>>> one2:/var/lib/one/datastores/143/55/disk.0
>>>>> Wed May 7 12:58:14 2014 [TM][E]: clone: Command "cd
>>>>> /var/lib/one/datastores/143/55; cp
>>>>> /var/lib/one/datastores/146/efe78e0d61224a0b459e6cc423b789a5
>>>>> /var/lib/one/datastores/143/55/disk.0" failed: cp: cannot stat
>>>>> `/var/lib/one/datastores/146/efe78e0d61224a0b459e6cc423b789a5': No such
>>>>> file or directory
>>>>> Wed May 7 12:58:14 2014 [TM][E]: Error copying
>>>>> opennebula-frontend:/var/lib/one//datastores/146/efe78e0d61224a0b459e6cc423b789a5
>>>>> to one2-hetzner.syseleven.de:/var/lib/one/datastores/143/55/disk.0
>>>>> Wed May 7 12:58:14 2014 [TM][I]: ExitCode: 1
>>>>> Wed May 7 12:58:14 2014 [TM][E]: Error executing image transfer script:
>>>>> Error copying
>>>>> opennebula-frontend:/var/lib/one//datastores/146/efe78e0d61224a0b459e6cc423b789a5
>>>>> to one2:/var/lib/one/datastores/143/55/disk.0
>>>>> Wed May 7 12:58:14 2014 [DiM][I]: New VM state is FAILED
>>>>>
>>>>> I don't get it.
>>>>>
>>>>> best regards
>>>>> t.
>>>>>
>>>>>> On Wed, May 7, 2014 at 11:58 AM, Thomas Stein <himbeere at meine-oma.de> wrote:
>>>>>>> Hello.
>>>>>>>
>>>>>>> I deleted all datastores and recreated the neccessary datastores again.
>>>>>>> Now i get:
>>>>>>>
>>>>>>> oneadmin at opennebula-frontend:~/datastores$ onedatastore list
>>>>>>>   ID NAME                SIZE AVAIL CLUSTER      IMAGES TYPE DS       TM
>>>>>>>
>>>>>>>  139 glusterds             0M -     -                 0 img  shared   shared
>>>>>>>  140 datastore_sys         0M -     fluffy            0 sys  -        shared
>>>>>>>  141 datastore_fil      47.2G 34%   fluffy            0 fil  fs       shared
>>>>>>> oneadmin at opennebula-frontend:~/datastores$ cd
>>>>>>>
>>>>>>> oneadmin at opennebula-frontend:~$ oneimage create -d 139 --name gentoo
>>>>>>> --path /opt/gentoo-template.qcow2 --type OS --driver qcow2 --description
>>>>>>> "gentoo"
>>>>>>>
>>>>>>> [ImageAllocate] Cannot determine Image SIZE. Datastore driver 'shared'
>>>>>>> not available
>>>>>>> oneadmin at opennebula-frontend:~$
>>>>>>>
>>>>>>> Someone has an idea how to solve this error?
>>>>>>>
>>>>>>> best regards
>>>>>>> t.
>>>>>>>
>>>>>>>
>>>>>>> Am 06.05.14 17:19, schrieb Thomas Stein:
>>>>>>>> Hello.
>>>>>>>>
>>>>>>>> I'm playing around with glustersfs and opennebula. Followed the quick
>>>>>>>> howto here:
>>>>>>>>
>>>>>>>> http://opennebula.org/native-glusterfs-image-access-for-kvm-drivers/
>>>>>>>>
>>>>>>>> Alltough it's not everything clear to me, i managed to create the
>>>>>>>> gluster datastore on the frontend.
>>>>>>>>
>>>>>>>> oneadmin at opennebula-frontend:~$ onedatastore show 135
>>>>>>>> DATASTORE 135 INFORMATION
>>>>>>>>
>>>>>>>> ID             : 135
>>>>>>>> NAME           : gluster_image
>>>>>>>> USER           : oneadmin
>>>>>>>> GROUP          : oneadmin
>>>>>>>> CLUSTER        : fluffy
>>>>>>>> TYPE           : IMAGE
>>>>>>>> DS_MAD         : fs
>>>>>>>> TM_MAD         : shared
>>>>>>>> BASE PATH      : /var/lib/one//datastores/135
>>>>>>>> DISK_TYPE      :
>>>>>>>>
>>>>>>>> DATASTORE CAPACITY
>>>>>>>>
>>>>>>>> TOTAL:         : 492.2G
>>>>>>>> FREE:          : 459.4G
>>>>>>>> USED:          : 7.6G
>>>>>>>> LIMIT:         : -
>>>>>>>>
>>>>>>>> PERMISSIONS
>>>>>>>>
>>>>>>>> OWNER          : um-
>>>>>>>> GROUP          : u--
>>>>>>>> OTHER          : ---
>>>>>>>>
>>>>>>>> DATASTORE TEMPLATE
>>>>>>>>
>>>>>>>> BASE_PATH="/var/lib/one//datastores/"
>>>>>>>> CLONE_TARGET="SYSTEM"
>>>>>>>> DISK_TYPE="GLUSTER"
>>>>>>>> DS_MAD="fs"
>>>>>>>> GLUSTER_HOST="one-volume1"
>>>>>>>> LN_TARGET="NONE"
>>>>>>>> TM_MAD="shared"
>>>>>>>> TYPE="IMAGE_DS"
>>>>>>>>
>>>>>>>> Problem is. When i try to create an image from a template which resides
>>>>>>>> in that datastore i get:
>>>>>>>>
>>>>>>>> Tue May  6 17:18:00 2014 [TM][D]: Message received: TRANSFER FAILURE 48
>>>>>>>> Error copying
>>>>>>>> opennebula-frontend:/var/lib/one//datastores/135/9ba000010fc657ac57edc0af308e4fa7
>>>>>>>> to remotehost:/var/lib/one/datastores/129/48/disk.0
>>>>>>>>
>>>>>>>> Why does opennebula to transfer the image?
>>>>>>>>
>>>>>>>> Thanks and best regards
>>>>>>>> t.
>>>>>>>> _______________________________________________
>>>>>>>> Users mailing list
>>>>>>>> Users at lists.opennebula.org
>>>>>>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Users mailing list
>>>>>>> Users at lists.opennebula.org
>>>>>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>
>>> _______________________________________________
>>> Users mailing list
>>> Users at lists.opennebula.org
>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>> 
>> 
>> 
> 
> _______________________________________________
> Users mailing list
> Users at lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
> 