date:20110408

Re: [one-users] Can't restart or resume VM when VM is in save status. And MEMORY = 0 when VM is running .

2011-04-08 Thread Prakhar Srivastava

Hi,
If by saving you intend to 'resume' that VM at a later point in time,your
steps are correct.

Regards,
Prakhar

2011/4/8 Khoa Nguyen 

>
> Hi Prakhar Srivastava
> Thank you for your reply. If i want to save a VM .
> 1. onevm stop id
> 2. Wait until changing from saving to stopping state.
> 3. onevm resume id.
>
> Is that right?
>
>
>
> 2011/4/8 Prakhar Srivastava 
>
>> Hi Khoa
>> As already stated by Carlos, the save state is an intermediate state in
>> which the running VM image is saved to the disk. This might take a little
>> time and in the the mean time, the VM state is save (means it is saving the
>> image so that it can be resumed from the state at which it was stopped). You
>> cannot perform resume  at save (saving...)  state. It will always give an
>> error. Once the VMs image is saved completely, the state changes to stop
>> (means the VM is stopped). You can only resume a VM from a stopped or
>> suspended state. So if you are not getting to 'stop' state, please check
>> your vm.log file for any error. You can find the log file in
>> /var/log//vm.log. Do post back the logs in case you find it cryptic
>> !!!
>>
>> Regards,
>> Prakhar
>>
>>
>> 2011/4/8 Khoa Nguyen 
>>
>>> Hi Carlos Martín Sánchez
>>>
>>> When the VM is in save status, How to change from saving  to running
>>> status?
>>> I try to enable VM , but it display Erros: wrong state.
>>> I can't solve problem.
>>>
>>>
>>>
>>>
>>> 2011/4/5 Carlos Martín Sánchez 
>>>
 Hi there,

 If you take a look at the Virtual Machine Life-cycle diagram [1], the
 'save' state is a intermediate state, and the VM should end in either the
 'suspended' or 'stopped' state. If the VM is stuck in 'save', you may find
 some error messages in oned.log or the VM's log (in /vm.log).

 About #2, it may be a monitorization problem. Can you check using virsh
 or xentop the actual memory consumption?

 Best regards,
 Carlos.

 [1]
 http://opennebula.org/documentation:rel2.2:vm_guide#virtual_machine_life-cycle

 --
 Carlos Martín, MSc
 Project Major Contributor
 OpenNebula - The Open Source Toolkit for Cloud Computing
 www.OpenNebula.org  | cmartin@opennebula
 .org


 On 17 March 2011 05:01, Khoa Nguyen  wrote:

> I have two question but I can't explain .
> 1.When i suspend or stop a VM, It changes to save status. Unfortunaly,I
> want to resume or restart  VM to changing running status , It display
> "Error: Wrong state to perform action". I don't know why. Is OpenNebula
> wrong??
>
> 2. Status's VM is ruunig but MEMORY = 0.
>
> Anyone know? Please help me.
>
>
> oneadmin1@oneadmin-desktop:/home/oneadmin$ onevm list
>ID USER NAME STAT CPU MEMHOSTNAMETIME
>39 oneadmin ttylinux runn   0 * 0K*   172.29.70.137 00 00:05:05
> oneadmin1@oneadmin-desktop:/home/oneadmin$ onevm suspend 39
> oneadmin1@oneadmin-desktop:/home/oneadmin$ onevm list
>ID USER NAME STAT CPU MEMHOSTNAMETIME
>39 oneadmin ttylinux save   0 * 0K*   172.29.70.137 00 00:05:21
> oneadmin1@oneadmin-desktop:/home/oneadmin$ *onevm resume 39
> Error: Wrong state to perform action*
> oneadmin1@oneadmin-desktop:/home/oneadmin$ *onevm restart 39
> Error: Wrong state to perform action*
>
> Thank you for advance.
>
> ,--
> Nguyễn Vũ Văn Khoa
> Đại học Khoa Học Tự Nhiên TP HCM
>
>
>
> ___
> Users mailing list
> Users@lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>
>

>>>
>>>
>>> --
>>> Nguyễn Vũ Văn Khoa
>>> Đại học Khoa Học Tự Nhiên TP HCM
>>>
>>>
>>>
>>> ___
>>> Users mailing list
>>> Users@lists.opennebula.org
>>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>>
>>>
>>
>
>
> --
> Nguyễn Vũ Văn Khoa
> Đại học Khoa Học Tự Nhiên TP HCM
>
>
>
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

Re: [one-users] VM creation stuck in "pending" state

2011-04-08 Thread Upendra Sharma


Hi Daniel,

sorry for late response ... 
here is the result


  4
  obelix20
  2
  im_xen
  vmm_xen
  tm_ssh
  1302289702
  cluster1
  
4
0
0
0
0
4193280
400
0
3091456
399
0
1101824
1
0
  
  













  



thanks,
-upendra

> From: dmol...@opennebula.org
> Date: Tue, 5 Apr 2011 11:44:42 +0200
> Subject: Re: [one-users] VM creation stuck in "pending" state
> To: upend...@hotmail.com
> CC: users@lists.opennebula.org
> 
> Hi Upendra,
> 
> Would you mind to send us the output of:
> onehost show 4 -x
> 
> Regards
> 
> On 4 April 2011 15:58, Daniel Molina  wrote:
> > Hi,
> >
> > Would you mind to try changing the Requirements statement to:
> > REQUIREMENTS = "HOSTNAME=obelix20"
> >
> > Regards
> >
> > On 2 April 2011 20:56, Upendra Sharma  wrote:
> >> Hi Tino,
> >>
> >> something similar is happening to me as well.
> >>
> >> the VM template looks is shown below.
> >>
> >>  vm template ==
> >> NAME   = TPCW-L
> >> CPU= 1
> >> MEMORY = 2000
> >> REQUIREMENTS = "HOSTNAME=\"obelix20\""
> >> RANK = FREECPU
> >> OS = [ bootloader=/usr/bin/pygrub ]
> >> DISK   = [
> >>   source   = "disk-vm1.img",
> >>   target   = "xvda",
> >>   readonly = "no" ]
> >> DISK   = [
> >>   source   = "disk-tpcw-db.img",
> >>   target   = "xvda5",
> >>   readonly = "no" ]
> >> NIC= [ NETWORK = "Public VLAN" ]
> >> RAW= [
> >>   type  = "xen",
> >>   data  = "vcpus=1"
> >> ]
> >> ==
> >>
> >>
> >>
> >>
> >> The main problem is this: my vm-template has a specific REQUIREMENTS .. 
> >> i.e.
> >> HOSTNAME. That particular host has a no VMs running on it and a 3 G of RAM
> >> free on it. The list of hosts on the cluster are:
> >>
> >>
> >>
> >>
> >> $ onehost list
> >>   ID NAME  CLUSTER  RVM   TCPU   FCPU   ACPUTMEMFMEM
> >> STAT
> >>4 obelix20  cluster1  0400  399 400
> >> 4G   2.9Gon
> >>
> >> the machine simply sits on the pending state.
> >>
> >> When I comment out the REQUIREMENTS line, it works fine and schedules it on
> >> one of the machines.
> >>
> >> I could not find a reason in any of the logs to find the reason.
> >>
> >>
> >>
> >> thanks,
> >> -upendra
> >>
> >>> From: tin...@opennebula.org
> >>> Date: Fri, 1 Apr 2011 15:01:42 +0100
> >>> To: zas...@pdc.kth.se
> >>> CC: users@lists.opennebula.org
> >>> Subject: Re: [one-users] VM creation stuck in "pending" state
> >>>
> >>> Hi,
> >>>
> >>> Any REQUIREMENT section in the VM templates?
> >>>
> >>> Regards,
> >>>
> >>> -Tino
> >>>
> >>> --
> >>> Constantino Vázquez Blanco, MSc
> >>> OpenNebula Major Contributor
> >>> www.OpenNebula.org | @tinova79
> >>>
> >>>
> >>>
> >>> On Fri, Apr 1, 2011 at 1:38 PM, Zeeshan Ali Shah 
> >>> wrote:
> >>> > in Our case they are monitorized and are not in error state as shown in
> >>> > one
> >>> > host list .  neither the memory cpu requirement are exhausted ..
> >>> > any other hint ?
> >>> > it was working fine after that I launched 5vm and it stucked (Although)
> >>> > I
> >>> > have 32 cores with 160GB RAM left.   but later i deleted vms restarted
> >>> > ONEd
> >>> > and SCHED:
> >>> > even after that when no vms left the new vms in pending state.I had
> >>> > to
> >>> > use manual one vm deploy everytime..
> >>> > Zeeshan
> >>> > On Fri, Apr 1, 2011 at 1:22 PM, Tino Vazquez  wrote:
> >>> >>
> >>> >> Hi,
> >>> >>
> >>> >> Please take a look at [1].
> >>> >>
> >>> >> Regards,
> >>> >>
> >>> >> -Tino
> >>> >>
> >>> >> [1]
> >>> >>
> >>> >> http://opennebula.org/documentation:community:faq#my_machine_stays_in_pending_state_what_can_i_do
> >>> >>
> >>> >> --
> >>> >> Constantino Vázquez Blanco | dsa-research.org/tinova
> >>> >> Virtualization Technology Engineer / Researcher
> >>> >> OpenNebula Toolkit | opennebula.org
> >>> >>
> >>> >>
> >>> >>
> >>> >> On Fri, Apr 1, 2011 at 11:59 AM, Zeeshan Ali Shah 
> >>> >> wrote:
> >>> >> > Happen to me as well , I can manually deploy, but by default it is
> >>> >> > stuck
> >>> >> > in
> >>> >> > Pending state forever ..   I restarted SCHED as well. but same ..
> >>> >> > any hint ?
> >>> >> >
> >>> >> > Zeeshan
> >>> >> >
> >>> >> > On Sun, Mar 27, 2011 at 5:34 PM, Héctor Sanjuán
> >>> >> > 
> >>> >> > wrote:
> >>> >> >>
> >>> >> >> Hi,
> >>> >> >>
> >>> >> >> make sure there is a host available to receive the VM. Have you
> >>> >> >> tried
> >>> >> >> to
> >>> >> >> manually deploy it by:
> >>> >> >>
> >>> >> >> $ onevm deploy vmid hostid
> >>> >> >>
> >>> >> >> Maybe the logs will offer more information then.
> >>> >> >>
> >>> >> >> Héctor
> >>> >> >>
> >>> >> >> El 27/03/11 17:24, Madhurranjan Mohaan escribió:
> >>> >> >> > Hi All,
> >>> >> >> >
> >>> >> >> > I am unable to create VMs any further because they are stuck in
> >>> >> >> > pending
> >>> >> >> > state.
> >>> >> >> >
> >>> >> >> > onevm list
> >>> >> >> >39 oneadminvm-ex pend   0  0K

Re: [one-users] Can't restart or resume VM when VM is in save status. And MEMORY = 0 when VM is running .

2011-04-08 Thread Khoa Nguyen

Hi Prakhar Srivastava
Thank you for your reply. If i want to save a VM .
1. onevm stop id
2. Wait until changing from saving to stopping state.
3. onevm resume id.

Is that right?


2011/4/8 Prakhar Srivastava 

> Hi Khoa
> As already stated by Carlos, the save state is an intermediate state in
> which the running VM image is saved to the disk. This might take a little
> time and in the the mean time, the VM state is save (means it is saving the
> image so that it can be resumed from the state at which it was stopped). You
> cannot perform resume  at save (saving...)  state. It will always give an
> error. Once the VMs image is saved completely, the state changes to stop
> (means the VM is stopped). You can only resume a VM from a stopped or
> suspended state. So if you are not getting to 'stop' state, please check
> your vm.log file for any error. You can find the log file in
> /var/log//vm.log. Do post back the logs in case you find it cryptic
> !!!
>
> Regards,
> Prakhar
>
>
> 2011/4/8 Khoa Nguyen 
>
>> Hi Carlos Martín Sánchez
>>
>> When the VM is in save status, How to change from saving  to running
>> status?
>> I try to enable VM , but it display Erros: wrong state.
>> I can't solve problem.
>>
>>
>>
>>
>> 2011/4/5 Carlos Martín Sánchez 
>>
>>> Hi there,
>>>
>>> If you take a look at the Virtual Machine Life-cycle diagram [1], the
>>> 'save' state is a intermediate state, and the VM should end in either the
>>> 'suspended' or 'stopped' state. If the VM is stuck in 'save', you may find
>>> some error messages in oned.log or the VM's log (in /vm.log).
>>>
>>> About #2, it may be a monitorization problem. Can you check using virsh
>>> or xentop the actual memory consumption?
>>>
>>> Best regards,
>>> Carlos.
>>>
>>> [1]
>>> http://opennebula.org/documentation:rel2.2:vm_guide#virtual_machine_life-cycle
>>>
>>> --
>>> Carlos Martín, MSc
>>> Project Major Contributor
>>> OpenNebula - The Open Source Toolkit for Cloud Computing
>>> www.OpenNebula.org  | cmar...@opennebula.org
>>>
>>>
>>> On 17 March 2011 05:01, Khoa Nguyen  wrote:
>>>
 I have two question but I can't explain .
 1.When i suspend or stop a VM, It changes to save status. Unfortunaly,I
 want to resume or restart  VM to changing running status , It display
 "Error: Wrong state to perform action". I don't know why. Is OpenNebula
 wrong??

 2. Status's VM is ruunig but MEMORY = 0.

 Anyone know? Please help me.


 oneadmin1@oneadmin-desktop:/home/oneadmin$ onevm list
ID USER NAME STAT CPU MEMHOSTNAMETIME
39 oneadmin ttylinux runn   0 * 0K*   172.29.70.137 00 00:05:05
 oneadmin1@oneadmin-desktop:/home/oneadmin$ onevm suspend 39
 oneadmin1@oneadmin-desktop:/home/oneadmin$ onevm list
ID USER NAME STAT CPU MEMHOSTNAMETIME
39 oneadmin ttylinux save   0 * 0K*   172.29.70.137 00 00:05:21
 oneadmin1@oneadmin-desktop:/home/oneadmin$ *onevm resume 39
 Error: Wrong state to perform action*
 oneadmin1@oneadmin-desktop:/home/oneadmin$ *onevm restart 39
 Error: Wrong state to perform action*

 Thank you for advance.

 ,--
 Nguyễn Vũ Văn Khoa
 Đại học Khoa Học Tự Nhiên TP HCM



 ___
 Users mailing list
 Users@lists.opennebula.org
 http://lists.opennebula.org/listinfo.cgi/users-opennebula.org


>>>
>>
>>
>> --
>> Nguyễn Vũ Văn Khoa
>> Đại học Khoa Học Tự Nhiên TP HCM
>>
>>
>>
>> ___
>> Users mailing list
>> Users@lists.opennebula.org
>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>>
>>
>


-- 
Nguyễn Vũ Văn Khoa
Đại học Khoa Học Tự Nhiên TP HCM
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

Re: [one-users] root user creates checkpoint directory on cluster node instead of oneadmin

2011-04-08 Thread Pieter Simoens


Dear all

sorry for spamming the list, but I want to report the solution for my 
issues as well, just to close this thread


Apparently, I had misconfigured the 
/etc/apparmor.d/abstractions/libvirt-qemu...

First, there was a typo in the line I had added.

Second, I had configured a specific VM_DIR, and you have to add an 
additional exception to the apparmor.d file!
In my case, this was VM_DIR=/work/onimages, and I had to add an 
additional line:

---> owner /srv/cloud/one/var/** rw,
--> owner /work/onimages/** rw,

Of course, restart the apparmor daemon again.

I do not know whether this is a generic "issue", maybe it should be 
added to the documentation on the OpenNebula website? Now, there is only 
a warning for the /srv/cloud/one/var directory...


Best regards,
Pieter



Op 8/04/2011 10:52, Pieter Simoens schreef:

Dear all,

following up on my e-mail below: according to 
http://dev.opennebula.org/issues/131, this should be solved in my 
version (I am running 2.0).


It seems it has to do with the way you connect to the virsh socket 
(qemu:///system ?)


So, I think the question is: where is the variable $LIBVIRT_URI 
specified? At least, that is the variable that is used in the 
/var/remotes/vmm/kvm/save script?


Best regards,
Pieter

Op 8/04/2011 10:18, Pieter Simoens schreef:

Dear all,

when I issue a command "onevm stop VM_ID'', the Virtual Machine stays 
in the save state. On my front-end node, the log-files (both oned.log 
and vm.log) show no specific error.


However, when I have a look at the virsh log-file on my cluster node 
(/var/log/libvirt/qemu/one-22.log), I see the error: "sh: cannot 
create /work/onimages/VM_ID/images/checkpoint: Permission denied".


Strangely enough, the permissions of /work/onimages/VM_ID/images 
directory are set correctly (or so I think...): oneadmin is the 
owner, and cloud is the group. Also the uid and gid are identical to 
the oneadmin user on my frontend. But the images/checkpoint directory 
seems to be owned by the root user, belonging to the root group.


Does anybody have an idea how this can be possible? Did I configure 
my cluster node in the wrong way?


Thank you,
Pieter

Op 5/04/2011 14:49, Carlos Martín Sánchez schreef:

Hi Pieter, please find my comments inline:


On 5 April 2011 13:42, Pieter Simoens > wrote:


Dear all,

I am running OpenNebula 2.0. I am able to successfully deploy a
VM on a cluster node. The image is (currently) transferred over
SSH. The VM boots successfully, and I am able to log-in.

When I do "onevm shutdown VM_ID", the VM is correctly shut down.
But it does no longer appear in the output of "onevm list all".


That's the normal behaviour, VMs in DONE state are not listed. 
However, you can still see the extended information using 'onevm 
show ', because they are stored in the DB for accounting purposes.


However, as far as I can recall from my previous experiments, I
thought that the VM image would we transferred back, somewhere
to VM_DIR/VM_ID directory.

Unfortunately, this is not happening. Does anybody has any idea
what can be the cause? Do I need to specify any additional
parameters? The log-files show me no specific problem (I checked
the oned.log and the /var/VM_ID/vm.log).


Are you using the Image Repository [1]? If so, is the Image persistent?
If you are defining the DISKs in the VM templates as in previous 
versions, make sure the attribute SAVE=YES is present.



Thank you,


Regards,
Carlos.

[1] http://opennebula.org/documentation:rel2.2:img_guide

--
Carlos Martín, MSc
Project Major Contributor
OpenNebula - The Open Source Toolkit for Cloud Computing
www.OpenNebula.org  | 
cmar...@opennebula.org 


-- 
Pieter Simoens, Ph.D.

Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W :www.ibcn.intec.UGent.be  


___
Users mailing list
Users@lists.opennebula.org 
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org




--
Pieter Simoens, Ph.D.
Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W :www.ibcn.intec.UGent.be


___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org


--
Pieter Simoens, Ph.D.
Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommen

[one-users] Problem monitoring Xen host 2.2

2011-04-08 Thread Javier Alcazar Luque

Hello,

I have installed opennebula 2.2 in a Centos 5.6 server, and everything works 
fine except for the hosts. The Kvm hosts seems to be monitoring correctly, but 
the Xen hosts show that the memory used is at 98% and there is no running vm in 
it.

Sunstone information:
Max Mem 24G
Used Mem (real) 23.6G

Top in the xen host:
Mem:  24259584k total,  1012020k used, 23247564k free

Any help?

Thanks
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

[one-users] A question (maybe bug) on parsing xentop outputs in var/remotes/vmm/xen/poll

2011-04-08 Thread Jiachen Yang

Hello everyone using OpenNEbular:

I recently installed ONE on my xen hosts and begin to study how to use
this excellent cloud management tool.
When I configured my hosts according to the user guide, I happily
submitted my first vm to one. After prologing, I got this kind of
error message of my vm.log:

Fri Apr  8 15:36:26 2011 [VMM][I]: Command execution fail: 'if [ -x
"/var/tmp/one/vmm/xen/poll" ]; then /var/tmp/one/vmm/xen/poll one-2;
else  exit 42; fi'
Fri Apr  8 15:36:26 2011 [VMM][I]: STDERR follows.
Fri Apr  8 15:36:26 2011 [VMM][I]: Error executing sudo /usr/sbin/xentop -bi2
Fri Apr  8 15:36:26 2011 [VMM][I]: /var/tmp/one/vmm/xen/poll:186:in
`get_vm_info': undefined method `[]' for nil:NilClass (NoMethodError)
Fri Apr  8 15:36:26 2011 [VMM][I]: from
/var/tmp/one/vmm/xen/poll:310:in `print_one_vm_info'
Fri Apr  8 15:36:26 2011 [VMM][I]: from /var/tmp/one/vmm/xen/poll:344
Fri Apr  8 15:36:26 2011 [VMM][I]: ExitCode: 1
Fri Apr  8 15:36:26 2011 [VMM][E]: Error monitoring VM, Error
executing sudo /usr/sbin/xentop -bi2

I typed the ruby code of poll in irb manually and found that
get_all_vm_info returned Nil.

When I tried to read and understand the source of poll then I met this line:

def self.get_all_vm_info
begin
 # ...
!   domain_lines=valid_lines[4..-1]
# ..

I want to ask why domain_lines start from 4th line in valid_lines?
When I typed these codes into irb, valid_lines.length == 3. There were
no 3rd and 4th items in valid_lines.

I changed it to:
 domain_lines=valid_lines[1..-1]

And all seems to work perfect on my hosts.

I want to know whether this is a bug or it has its intentions to start
from the 4th line in valid_lines.


About my host configuration:
I got 2 host installed SLES 11 sp1 and Xen3.3.
I added these hosts to ONE using:
onehost linux-front im_xen vmm_xen tm_ssh

-
Jiachen Yang
School of Software Engineering
Shanghai Jiaotong University
Tel. (+86) 13774494600

farsee...@sjtu.edu.cn
firechild...@163.com
farsee...@gmail.com (gtalk)
firechild...@hotmail.com (msn)
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

Re: [one-users] root user creates checkpoint directory on cluster node instead of oneadmin

2011-04-08 Thread Pieter Simoens


Dear all,

following up on my e-mail below: according to 
http://dev.opennebula.org/issues/131, this should be solved in my 
version (I am running 2.0).


It seems it has to do with the way you connect to the virsh socket 
(qemu:///system ?)


So, I think the question is: where is the variable $LIBVIRT_URI 
specified? At least, that is the variable that is used in the 
/var/remotes/vmm/kvm/save script?


Best regards,
Pieter

Op 8/04/2011 10:18, Pieter Simoens schreef:

Dear all,

when I issue a command "onevm stop VM_ID'', the Virtual Machine stays 
in the save state. On my front-end node, the log-files (both oned.log 
and vm.log) show no specific error.


However, when I have a look at the virsh log-file on my cluster node 
(/var/log/libvirt/qemu/one-22.log), I see the error: "sh: cannot 
create /work/onimages/VM_ID/images/checkpoint: Permission denied".


Strangely enough, the permissions of /work/onimages/VM_ID/images 
directory are set correctly (or so I think...): oneadmin is the owner, 
and cloud is the group. Also the uid and gid are identical to the 
oneadmin user on my frontend. But the images/checkpoint directory 
seems to be owned by the root user, belonging to the root group.


Does anybody have an idea how this can be possible? Did I configure my 
cluster node in the wrong way?


Thank you,
Pieter

Op 5/04/2011 14:49, Carlos Martín Sánchez schreef:

Hi Pieter, please find my comments inline:


On 5 April 2011 13:42, Pieter Simoens > wrote:


Dear all,

I am running OpenNebula 2.0. I am able to successfully deploy a
VM on a cluster node. The image is (currently) transferred over
SSH. The VM boots successfully, and I am able to log-in.

When I do "onevm shutdown VM_ID", the VM is correctly shut down.
But it does no longer appear in the output of "onevm list all".


That's the normal behaviour, VMs in DONE state are not listed. 
However, you can still see the extended information using 'onevm show 
', because they are stored in the DB for accounting purposes.


However, as far as I can recall from my previous experiments, I
thought that the VM image would we transferred back, somewhere to
VM_DIR/VM_ID directory.

Unfortunately, this is not happening. Does anybody has any idea
what can be the cause? Do I need to specify any additional
parameters? The log-files show me no specific problem (I checked
the oned.log and the /var/VM_ID/vm.log).


Are you using the Image Repository [1]? If so, is the Image persistent?
If you are defining the DISKs in the VM templates as in previous 
versions, make sure the attribute SAVE=YES is present.



Thank you,


Regards,
Carlos.

[1] http://opennebula.org/documentation:rel2.2:img_guide

--
Carlos Martín, MSc
Project Major Contributor
OpenNebula - The Open Source Toolkit for Cloud Computing
www.OpenNebula.org  | 
cmar...@opennebula.org 


-- 
Pieter Simoens, Ph.D.

Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W :www.ibcn.intec.UGent.be  


___
Users mailing list
Users@lists.opennebula.org 
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org




--
Pieter Simoens, Ph.D.
Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W :www.ibcn.intec.UGent.be


___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org


--
Pieter Simoens, Ph.D.
Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W : www.ibcn.intec.UGent.be

___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

[one-users] root user creates checkpoint directory on cluster node instead of oneadmin

2011-04-08 Thread Pieter Simoens


Dear all,

when I issue a command "onevm stop VM_ID'', the Virtual Machine stays in 
the save state. On my front-end node, the log-files (both oned.log and 
vm.log) show no specific error.


However, when I have a look at the virsh log-file on my cluster node 
(/var/log/libvirt/qemu/one-22.log), I see the error: "sh: cannot create 
/work/onimages/VM_ID/images/checkpoint: Permission denied".


Strangely enough, the permissions of /work/onimages/VM_ID/images 
directory are set correctly (or so I think...): oneadmin is the owner, 
and cloud is the group. Also the uid and gid are identical to the 
oneadmin user on my frontend. But the images/checkpoint directory seems 
to be owned by the root user, belonging to the root group.


Does anybody have an idea how this can be possible? Did I configure my 
cluster node in the wrong way?


Thank you,
Pieter

Op 5/04/2011 14:49, Carlos Martín Sánchez schreef:

Hi Pieter, please find my comments inline:


On 5 April 2011 13:42, Pieter Simoens > wrote:


Dear all,

I am running OpenNebula 2.0. I am able to successfully deploy a VM
on a cluster node. The image is (currently) transferred over SSH.
The VM boots successfully, and I am able to log-in.

When I do "onevm shutdown VM_ID", the VM is correctly shut down.
But it does no longer appear in the output of "onevm list all".


That's the normal behaviour, VMs in DONE state are not listed. 
However, you can still see the extended information using 'onevm show 
', because they are stored in the DB for accounting purposes.


However, as far as I can recall from my previous experiments, I
thought that the VM image would we transferred back, somewhere to
VM_DIR/VM_ID directory.

Unfortunately, this is not happening. Does anybody has any idea
what can be the cause? Do I need to specify any additional
parameters? The log-files show me no specific problem (I checked
the oned.log and the /var/VM_ID/vm.log).


Are you using the Image Repository [1]? If so, is the Image persistent?
If you are defining the DISKs in the VM templates as in previous 
versions, make sure the attribute SAVE=YES is present.



Thank you,


Regards,
Carlos.

[1] http://opennebula.org/documentation:rel2.2:img_guide

--
Carlos Martín, MSc
Project Major Contributor
OpenNebula - The Open Source Toolkit for Cloud Computing
www.OpenNebula.org  | 
cmar...@opennebula.org 


-- 
Pieter Simoens, Ph.D.

Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W :www.ibcn.intec.UGent.be  


___
Users mailing list
Users@lists.opennebula.org 
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org




--
Pieter Simoens, Ph.D.
Department of Information Technology
Internet Based Communication Networks and Services (IBCN)
Ghent University - IBBT
Gaston Crommenlaan 8 (Bus 201), B-9050 Gent, Belgium
T: +32 9 33 14878 ; T Secr: +32 9 33 14900
F: +32 9 33 14899
W : www.ibcn.intec.UGent.be

___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

Re: [one-users] Sunstone VS Onemc

2011-04-08 Thread Héctor Sanjuán

Hi,

There are not many differences. Sunstone is perhaps more up to date
regarding the OpenNebula API though. Nevertheless there are both have
different features and you should choose the one that fits you best.

Regarding Sunstone documentation, you can check [1].

We think it is a very good idea to include a VNC client in Sunstone, and
so I have opened a ticket [2] to put it in practice (thanks!).

Hector

[1] http://www.opennebula.org/documentation:rel2.2:sunstone
[2] http://dev.opennebula.org/issues/564

El 07/04/11 08:01, KINGLABS escribió:
> Hi
> 
> I have recently installed OpenNebula 2.2 for mainly checking sunstone. I
> would like to
> know how it would compare with OneMC .
> 
> Also I realized that I do not have option to  console(VNC) to the vm
> from sunstone interface which is very basic need , unlike onemc.
> 
> What are the advantages of using sunstone ? Is there any guide for using
> Sunstone ?
> 
> 
> Regards
> ___
> Users mailing list
> Users@lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
> 


-- 
Héctor Sanjuan
OpenNebula.org
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org

Re: [one-users] Can't restart or resume VM when VM is in save status. And MEMORY = 0 when VM is running .

Re: [one-users] VM creation stuck in "pending" state

Re: [one-users] Can't restart or resume VM when VM is in save status. And MEMORY = 0 when VM is running .

Re: [one-users] root user creates checkpoint directory on cluster node instead of oneadmin

[one-users] Problem monitoring Xen host 2.2

[one-users] A question (maybe bug) on parsing xentop outputs in var/remotes/vmm/xen/poll

Re: [one-users] root user creates checkpoint directory on cluster node instead of oneadmin

[one-users] root user creates checkpoint directory on cluster node instead of oneadmin

Re: [one-users] Sunstone VS Onemc

9 matches

Site Navigation

Mail list logo

Footer information