[one-users] SSH error

[Zeeshan Ali Shah]

for shortrun i am transfering manually .. but for long run if we have 
the possibility to register kernel and initrd as well like we do in euca 
, that will be great.. may be something to be added in wish list.


Zeeshan

On 02/09/2011 05:36 PM, Javier Fontan wrote:
> That is possible but that will lead to huge amount of hacks as you
> have to do this (at least):
>
> * Parse the deployment.0 file or call back to opennebula to ask for
> the source kernel/initrd files. TM knows nothing about kernel/initrd
> * Rewrite the deployment.0 file to point to the newly copied
> kernel/initrd in the node
> * Take care somewhere in the stack to copy the kernel/initrd file for
> livemigration (it wont work without shared storage anyway)
>
> Why don't you just use pygrub?
>
> OS = [ bootloader = "/usr/bin/pygrub" ]
>
> http://wiki.xensource.com/xenwiki/PyGrub
>
> On Wed, Feb 9, 2011 at 1:32 PM, Zeeshan Ali Shah<zashah at pdc.kth.se>  wrote:
>    
>> what about  extending tm driver for ssh which copy the needed kernel
>> alongwith images to node ?
>>
>> Zeeshan
>>
>> On 02/09/2011 01:30 PM, Ruben Diez wrote:
>>      
>>> Hi Zeeshan
>>>
>>> You MUST put the kernels in all the nodes. You can use the parallel-scp
>>> stuff or export them to all the nodes by NFS... It is your decision... The
>>> concrete mechanism is not the matter... The important stuff is to make
>>> kernels accessible by all the physical nodes....
>>>
>>> Regards
>>>
>>> Zeeshan Ali Shah escribió:
>>>        
>>>> Hi Ruben  ,
>>>> thanks for answer ,
>>>>
>>>> But if i have 10 kernels and 10 nodes , is there any easy way to transfer
>>>> kernels beside NFS ..
>>>>
>>>> in Eucalyptus we can upload kernel and initrd alongwith image in
>>>> repository , any way to do the same in opennebula ..?
>>>>
>>>>
>>>> BR
>>>>
>>>> Zeeshan
>>>>
>>>> On 02/08/2011 06:31 PM, Ruben Diez wrote:
>>>>          
>>>>> You must copy the kernel on the correct path to every physical nodes....
>>>>>
>>>>> You can use a tool like parallel-scp (
>>>>> http://code.google.com/p/parallel-ssh/ ) for do this... (perhaps your distro
>>>>> have a package for parallel-ssh)
>>>>>
>>>>> Other possibility is use full virtualization and not
>>>>> paravirtualization....
>>>>>
>>>>> Regards
>>>>>
>>>>> Zeeshan Ali Shah escribió:
>>>>>            
>>>>>> Hi, When creating vm i am getting this
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh:
>>>>>> frontnebula:/srv/cloud/one/var//images/8625d68b699fd30e64360471eb2c38fed47fcfb6
>>>>>> nebula1:/srv/cloud/one/var//0/images/disk.0
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh: DST:
>>>>>> /srv/cloud/one/var//0/images/disk.0
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh: Creating directory
>>>>>> /srv/cloud/one/var//0/images
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh: Executed "/usr/bin/ssh
>>>>>> nebula1 mkdir -p /srv/cloud/one/var//0/images".
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh: Cloning
>>>>>> frontnebula:/srv/cloud/one/var//images/8625d68b699fd30e64360471eb2c38fed47fcfb6
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh: Executed "/usr/bin/scp
>>>>>> frontnebula:/srv/cloud/one/var//images/8625d68b699fd30e64360471eb2c38fed47fcfb6
>>>>>> nebula1:/srv/cloud/one/var//0/images/disk.0".
>>>>>> Tue Feb  8 11:43:07 2011 [TM][I]: tm_clone.sh: Executed "/usr/bin/ssh
>>>>>> nebula1 chmod a+rw /srv/cloud/one/var//0/images/disk.0".
>>>>>> Tue Feb  8 11:43:07 2011 [LCM][I]: New VM state is BOOT
>>>>>> Tue Feb  8 11:43:07 2011 [VMM][I]: Generating deployment file:
>>>>>> /srv/cloud/one/var/0/deployment.0
>>>>>> Tue Feb  8 11:43:08 2011 [VMM][I]: Command execution fail: 'if [ -x
>>>>>> "/var/tmp/one/vmm/xen/deploy" ]; then /var/tmp/one/vmm/xen/deploy
>>>>>> /srv/cloud/one/var//0/images/deployment.0; else
>>>>>>   exit 42; fi'
>>>>>> Tue Feb  8 11:43:08 2011 [VMM][I]: STDERR follows.
>>>>>> Tue Feb  8 11:43:08 2011 [VMM][I]: Error: Kernel image does not exist:
>>>>>> */srv/cloud/one/ttylinux-xen/vmlinuz-xen*
>>>>>> Tue Feb  8 11:43:08 2011 [VMM][I]: ExitCode: 1
>>>>>> Tue Feb  8 11:43:08 2011 [VMM][E]: Error deploying virtual machine:
>>>>>> Error: Kernel image does not exist: /srv/cloud/one/ttylinux-xen/vmlinuz-xen
>>>>>> Tue Feb  8 11:43:08 2011 [DiM][I]: New VM state is FAILED
>>>>>>
>>>>>>
>>>>>> In my vmtemplate it has kernel   =
>>>>>> "/srv/cloud/one/ttylinux-xen/vmlinuz-xen",
>>>>>>   initrd   = "/srv/cloud/one/ttylinux-xen/initrd.gz",
>>>>>>
>>>>>>
>>>>>> Now as i am using SSH (with out NFS) on Nodes the location
>>>>>> /srv/cloud/one/ttylinux-xen does not exist since I only untar it on Front
>>>>>> End.
>>>>>>
>>>>>> Am I missing something ?
>>>>>>
>>>>>> Thanks in advance.
>>>>>>
>>>>>> Zeeshan
>>>>>>
>>>>>>
>>>>>> ------------------------------------------------------------------------
>>>>>>
>>>>>> _______________________________________________
>>>>>> Users mailing list
>>>>>> Users at lists.opennebula.org
>>>>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>>>>>              
>>>>>            
>>>>
>>>>          
>>>        
>>
>> --
>> Regards
>>
>> Zeeshan Ali Shah
>> System Administrator
>> PDC-Center for High Performance Computing
>> KTH-Royal Institute of Technology , Sweden
>> +46 8 790 9115
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.opennebula.org
>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>
>>      
>
>
>    


-- 
Regards

Zeeshan Ali Shah
System Administrator
PDC-Center for High Performance Computing
KTH-Royal Institute of Technology , Sweden
+46 8 790 9115