[one-users] New Ceph Datastore reports "Not enough space in datastore" after upgrade to 4.1.80

[Jon]

Hi Jamie,

Ok, I did that, my sudoers for opennebula now looks like:

    oneadmin    ALL=(ALL) NOPASSWD: /usr/bin/ovs-ofctl, /usr/bin/ovs-vsctl,
/usr/bin/rados

the oneadmin user also has the ability to run all ceph/rados commands
without sudo (I created a ceph group, added the oneadmin user to the group,
then chgrp ceph /etc/ceph/ceph.keyring ).

I am able to run the rados command as oneadmin, and running what looks to
be the monitor command is successful:

> oneadmin at red6:~$ sudo rados df | awk '{
>     if ($1 == "total") {
>
>         space = int($3/1024)
>
>         if      ($2 == "used")  {var = "USED_MB"}
>         else if ($2 == "avail") {var = "FREE_MB"}
>         else if ($2 == "space") {var = "TOTAL_MB"}
>
>         print var "=" space
>     }
> }'
> USED_MB=1221077
> FREE_MB=3544932
> TOTAL_MB=4766009

However, when I run the full command found in the logs, I get an error:

oneadmin at red6:~$ /var/lib/one/remotes/datastore/ceph/monitor
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
101
ERROR: monitor: Command "" failed:
ERROR MESSAGE --8<------
Cannot monitor sudo rados df | awk '{
    if ($1 == "total") {

        space = int($3/1024)

        if      ($2 == "used")  {var = "USED_MB"}
        else if ($2 == "avail") {var = "FREE_MB"}
        else if ($2 == "space") {var = "TOTAL_MB"}

        print var "=" space
    }
}'
ERROR MESSAGE ------>8--

The full log message is:

Fri Jul 19 21:16:45 2013 [ImM][I]: Datastore files (2) successfully
monitored.
Fri Jul 19 21:16:45 2013 [ImM][I]: Datastore default (1) successfully
monitored.
Fri Jul 19 21:16:45 2013 [ImM][I]: Command execution fail:
/var/lib/one/remotes/datastore/ceph/monitor
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
101
Fri Jul 19 21:16:45 2013 [ImM][I]: ExitCode: 255
Fri Jul 19 21:16:45 2013 [ImM][E]: Error monitoring datastore 101: -

I'm really not sure where to go from here.

Is there a way to set a hard limit on a datastore, or would storage limits
be configured only at a user level?

Thanks,
Jon A



On Fri, Jul 19, 2013 at 2:50 AM, Jaime Melis <jmelis at opennebula.org> wrote:

> Hi Jon,
>
> thanks for the feedback.
>
> Can you try again adding the rados command to the NOPASSWD sudoers
> commands in the $HOST of the Ceph datastore?
>
> cheers,
> Jaime
>
>
> On Fri, Jul 19, 2013 at 7:47 AM, Jon <three18ti at gmail.com> wrote:
>
>> Hello,
>>
>> I've basically performed a fresh install of the New OpenNebula.
>>
>> I'm attempting to create a new image on a ceph datastore.  When I click
>> submit in sunstone, I get an  error "Not enough space in datastore"
>>
>> The config of the datastore clearly shows 0/0/0M total/used/free.  Off
>> the top of my head, I also don't remember there being a path for RBD
>> datastores in 4.0.
>>
>> Is there a way to set the size of the datastore?  I'm unable to edit it
>> via sunstone and there is no size parameter when I run `onedatastore update
>> 100`.
>>
>> I don't see anything in the docs about setting the size for a ceph
>> datastore: http://opennebula.org/documentation:rel4.2:ceph_ds
>> but maybe the docs haven't been updated yet?
>>
>> Any ideas as to what I'm missing?
>>
>> Thanks,
>> Jon A
>>
>>
>> oneadmin at red6:~$ onedatastore show 100
>> DATASTORE 100 INFORMATION
>>
>> ID             : 100
>> NAME           : rbd-1
>> USER           : oneadmin
>> GROUP          : oneadmin
>> CLUSTER        : -
>> TYPE           : IMAGE
>> DS_MAD         : ceph
>> TM_MAD         : ceph
>> BASE PATH      : /var/lib/one/datastores/100
>> DISK_TYPE      : RBD
>>
>> DATASTORE CAPACITY
>>
>> TOTAL:         : 0M
>> USED:          : 0M
>> FREE:          : 0M
>>
>> PERMISSIONS
>>
>> OWNER          : um-
>> GROUP          : u--
>> OTHER          : ---
>>
>> DATASTORE TEMPLATE
>>
>> DS_MAD="ceph"
>> TM_MAD="ceph"
>> TYPE="IMAGE_DS"
>>
>> IMAGES
>>
>>
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.opennebula.org
>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>
>>
>
>
> --
> Join us at OpenNebulaConf2013 <http://opennebulaconf.com/> in Berlin, 24-26
> September, 2013
> --
> Jaime Melis
> Project Engineer
> OpenNebula - The Open Source Toolkit for Cloud Computing
> www.OpenNebula.org | jmelis at opennebula.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20130719/b0f0293b/attachment-0001.htm>