[Rocks-Discuss] rocks 6.1 kickstart compute nodes stops at language screen

Marcin Sliwowski marcin at renci.org
Mon Jun 10 07:37:30 PDT 2013


Hello Dimitris,

Thanks for sharing your solution, I had the same problem of the missing 
password in the updated file.

Also thanks to Luca and Philip for the suggestion to try the rocks 
commands as user apache.

Thanks,
Marcin

On 06/09/2013 04:52 AM, Dimitris Mavroeidis wrote:
> Hi Philip,
>
> You are right, I ran the command as "apache" and here is what I got:
>
> ************************************************
> Traceback (most recent call last):
>   File "/opt/rocks/bin/rocks", line 301, in <module>
>     command.runWrapper(name, args[i:])
>   File 
> "/opt/rocks/lib/python2.6/site-packages/rocks/commands/__init__.py", 
> line 2194, in runWrapper
>     self.run(self._params, self._args)
>   File 
> "/opt/rocks/lib/python2.6/site-packages/rocks/commands/list/host/profile/__init__.py", 
> line 301, in run
>     for host in self.getHostnames(args):
>   File 
> "/opt/rocks/lib/python2.6/site-packages/rocks/commands/__init__.py", 
> line 773, in getHostnames
>     min,max = self.db.fetchone()
> TypeError: 'NoneType' object is not iterable
> ************************************************
>
> So, I looked around a bit and found that /opt/rocks/etc/my.cnf has 
> been replaced by the update (so, the mysql password was missing). 
> Restored it, and now I can successfully run the command and also get 
> the correct kickstart file.
>
> Thanks for the useful pointer Philip!
>
> Best regards,
> Dimitris
>
>
>
> Στις 7/6/2013 7:24 μμ, ο/η Philip Papadopoulos έγραψε:
>> on the frontend
>> # sudo -u apache /opt/rocks/bin/rocks list host profile compute-0-0
>> and tell us what errors you see
>>
>> -P
>>
>>
>>
>>
>> On Fri, Jun 7, 2013 at 7:23 AM, Dimitris Mavroeidis 
>> <dmavroeidis at csri.gr> wrote:
>>> Hi Marcin,
>>>
>>> I have the same problem after upgrading the CentOS packages. My compute
>>> nodes cannot pxe boot.
>>>
>>> Apache seems to be working, since `rocks list host profile compute-0-0`
>>> does not give me an error.
>>>
>>> Though, when running "wget --no-check-certificate
>>> https://localhost/install/sbin/public/kickstart.cgi", I get an empty 
>>> file.
>>>
>>> I have tried several suggestions from previous posts, but nothing seems
>>> to work.
>>>
>>> If you finally come up with a solution, please share it with the list.
>>>
>>> Best regards,
>>> Dimitris
>>>
>>>
>>>
>>> On 31/05/2013 10:56 μμ, Marcin Sliwowski wrote:
>>>> Luca,
>>>>
>>>> I have the following rolls installed
>>>>
>>>> ganglia:      6.1     x86_64 yes
>>>> web-server:   6.1     x86_64 yes
>>>> perl:         6.1     x86_64 yes
>>>> kernel:       6.1     x86_64 yes
>>>> python:       6.1     x86_64 yes
>>>> area51:       6.1     x86_64 yes
>>>> updates:      6.4     x86_64 yes
>>>> base:         6.1     x86_64 yes
>>>> java:         6.1     x86_64 yes
>>>> os:           6.4     x86_64 yes
>>>> service-pack: 6.1     x86_64 yes
>>>>
>>>> I think you are right about apache not being able to connect to the
>>>> DB, as apache a `rocks list host profile compute-0-0` dumps that exact
>>>> error.
>>>>
>>>> Now to try and figure out what exactly is breaking it.
>>>>
>>>> Thanks
>>>>
>>>> Marcin Sliwowski | SysAdmin at RENCI | (919) 445-0479
>>>>
>>>> On 05/31/2013 03:43 PM, Luca Clementi wrote:
>>>>> On Fri, May 31, 2013 at 10:05 AM, Marcin Sliwowski <marcin at renci.org>
>>>>> wrote:
>>>>>> I am pretty confidant that httpd is still functional and doing its
>>>>>> job. It
>>>>>> does log both the attempts made and the errors involved in pulling
>>>>>> down the
>>>>>> kickstart file.
>>>>>>
>>>>>> I thought that maybe someone would recognize the error that is being
>>>>>> logged
>>>>>> in /var/log/httpd/ssl_error_log:
>>>>>>
>>>>>> Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] Traceback (most
>>>>>> recent
>>>>>> call last):
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] File
>>>>>> "/opt/rocks/bin/rocks", line 301, in <module>
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] File
>>>>>> "/opt/rocks/bin/rocks", line 301, in <module>
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1]
>>>>>> command.runWrapper(name, args[i:])
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] File
>>>>>> "/opt/rocks/lib/python2.6/site-packages/rocks/commands/__init__.py",
>>>>>> line
>>>>>> 2194, in runWrapper
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1]
>>>>>> self.run(self._params,
>>>>>> self._args)
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] File
>>>>>> "/opt/rocks/lib/python2.6/site-packages/rocks/commands/list/host/attr/__init__.py", 
>>>>>>
>>>>>>
>>>>>> line 129, in run
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1]     for 
>>>>>> host in
>>>>>> self.getHostnames(args):
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] File
>>>>>> "/opt/rocks/lib/python2.6/site-packages/rocks/commands/__init__.py",
>>>>>> line
>>>>>> 773, in getHostnames
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] min,max =
>>>>>> self.db.fetchone()
>>>>>> [Wed May 29 14:32:02 2013] [error] [client 127.0.0.1] TypeError:
>>>>>> 'NoneType'
>>>>>> object is not iterable
>>>>>>
>>>>> I would say that apache can not connect anymore to the rocks mysql 
>>>>> DB.
>>>>> It is difficult to say why though...
>>>>>
>>>>> Which roll do you have installed? (rocks list roll)
>>>>>
>>>>> You could try to enable the apache user, so you can "su - apache", 
>>>>> and
>>>>> then try to run some "rocks list" command.
>>>>>
>>>>> rocks list host profile compute-0-0-0
>>>>>
>>>>> I suppose that some update to mysql/mysql-python/mysql-libs changed
>>>>> something which broke this.
>>>>>
>>>>> Luca
>>> -------------- next part --------------
>>> A non-text attachment was scrubbed...
>>> Name: dmavroeidis.vcf
>>> Type: text/x-vcard
>>> Size: 385 bytes
>>> Desc: not available
>>> Url : 
>>> https://lists.sdsc.edu/pipermail/npaci-rocks-discussion/attachments/20130607/3ab890b0/dmavroeidis.vcf
>>>
>>
>>
>



More information about the npaci-rocks-discussion mailing list