incubator-tashi-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From hari narayanan <hari.zla...@gmail.com>
Subject Re: Problem with starting node manager
Date Mon, 18 Apr 2011 23:28:39 GMT
This is actually the error that i get in primitive.py

Failed to schedule or activate hari
Traceback (most recent call last):
  File "./bin/primitive.py", line 162, in start
    hook.preCreate(inst)
  File "/home/hari/tashi/src/tashi/agents/dhcpdns.py", line 223, in
preCreate
    ip = self.allocateIP(nic)
  File "/home/hari/tashi/src/tashi/agents/dhcpdns.py", line 93, in
allocateIP
    if (requestedIP <= self.ipMax[network] and requestedIP >=
self.ipMin[network] and (requestedIP not in self.usedIPs)):
KeyError: 272




On Mon, Apr 18, 2011 at 6:53 PM, hari narayanan <hari.zlatan@gmail.com>wrote:

> Yes, I started the scheduler. It says "Failed to schedule or activate DC" .
> Key error 272...
>
> On Mon, Apr 18, 2011 at 6:37 PM, Richard Gass <richardgass@gmail.com>wrote:
>
>> Did you start the scheduler?
>>
>>
>> --
>> Sent from my mobile device
>>
>> Richard
>>
>> On Apr 19, 2011, at 0:35, hari narayanan <hari.zlatan@gmail.com> wrote:
>>
>> Yes... The VM hasnt started .... the message that i get is "[/home/akshay/Downloads/tashi/src/tashi/nodemanager/vmcontrol/qemu.pyc:INFO]
>> *No vm information found in /var/tmp/VmCont**rolQemu/*" ... This same msg
>> is given in the documentation too .... I started the host as mentioned in
>> the documentation using the cluster manager. How do i fix this ?
>>
>> On Mon, Apr 18, 2011 at 6:26 PM, Richard Gass < <richardgass@gmail.com>
>> richardgass@gmail.com> wrote:
>>
>>> Looks like the vm did not start. You are doing a stand alone setup?  Look
>>> at the nodemanger console and look for the qemu command that got printed.
>>> Copy that line and run it in a terminal and see if it starts. You will most
>>> likely find your problem there.
>>>
>>>
>>> --
>>> Sent from my mobile device
>>>
>>> Richard
>>>
>>> On Apr 19, 2011, at 0:24, hari narayanan < <hari.zlatan@gmail.com>
>>> hari.zlatan@gmail.com> wrote:
>>>
>>> id hostId name user   state    disk             memory cores
>>> -------------------------------------------------------------
>>> 11 None   DC   akshay Pending  gutsy-server.img 128    1
>>>
>>> On Mon, Apr 18, 2011 at 6:20 PM, Richard Gass < <richardgass@gmail.com><richardgass@gmail.com>
>>> richardgass@gmail.com> wrote:
>>>
>>>> Can you show output of getinstances
>>>>
>>>>
>>>> --
>>>> Sent from my mobile device
>>>>
>>>> Richard
>>>>
>>>> On Apr 19, 2011, at 0:11, hari narayanan < <hari.zlatan@gmail.com><hari.zlatan@gmail.com>
>>>> hari.zlatan@gmail.com> wrote:
>>>>
>>>>
>>>> I tried using it. I get an error that "No such hostid:None"
>>>>
>>>> Thanks
>>>>
>>>> On Mon, Apr 18, 2011 at 5:59 PM, Richard Gass < <richardgass@gmail.com><richardgass@gmail.com><richardgass@gmail.com>
>>>> richardgass@gmail.com> wrote:
>>>>
>>>>> If you don't setup a dhcp server, you can access the console of the vm
>>>>> through the startVnc arg with tashi.
>>>>>
>>>>> --
>>>>> Sent from my mobile device
>>>>>
>>>>> Richard
>>>>>
>>>>> On Apr 18, 2011, at 23:49, hari narayanan < <hari.zlatan@gmail.com><hari.zlatan@gmail.com><hari.zlatan@gmail.com>
>>>>> hari.zlatan@gmail.com> wrote:
>>>>>
>>>>> Is it true that the VM will be allocated an IP automatically when it
is
>>>>> created using the "*tashi-client.py createVm*" command if i have setup
>>>>> a bridge and DHCP server in my host system ?? In the documentation, this
is
>>>>> not indicated very clearly . Sorry for troubling with repeated emails
>>>>>
>>>>> Thanks
>>>>>
>>>>> On Mon, Apr 18, 2011 at 2:31 PM, Michael Stroucken < <mxs@cmu.edu><mxs@cmu.edu><mxs@cmu.edu><mxs@cmu.edu>
>>>>> mxs@cmu.edu> wrote:
>>>>>
>>>>>> hari narayanan wrote:
>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> I have setup QEMU/kvm running ubuntu 10.10.... I started the
cluster
>>>>>>> manager and it worked. However, when i tried starting the node
manager ,i
>>>>>>> get the following error
>>>>>>>
>>>>>>>  File "./bin/nodemanager.py", line 73, in <module>
>>>>>>>    main()
>>>>>>>  File "./bin/nodemanager.py", line 47, in main
>>>>>>>    vmm = instantiateImplementation(config.get("NodeManager",
"vmm"),
>>>>>>> config, dfs, None)
>>>>>>>  File "/home/hari/tashi/src/tashi/util.py", line 191, in
>>>>>>> instantiateImplementation
>>>>>>>    exec cmd in locals()
>>>>>>>  File "<string>", line 2, in <module>
>>>>>>>  File "/home/hari/tashi/src/tashi/nodemanager/vmcontrol/qemu.py",
>>>>>>> line 108, in __init__
>>>>>>>    self.scratchDir = self.config.get("Qemu", "scratchDir")
>>>>>>>  File "/usr/lib/python2.6/ConfigParser.py", line 540, in get
>>>>>>>    raise NoOptionError(option, section)
>>>>>>> ConfigParser.NoOptionError: No option 'scratchdir' in section:
'Qemu'
>>>>>>>
>>>>>>> Can someone tell me what the problem is ?
>>>>>>>
>>>>>>>  Hi Hari,
>>>>>>
>>>>>> A parameter "scratchDir" is now required to direct copy-on-write
disk
>>>>>> output to a specified directory. Before, this data was written (by
qemu) to
>>>>>> /tmp alone, which may not have enough space.
>>>>>>
>>>>>> Greetings,
>>>>>> Michael.
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>

Mime
View raw message