[Moabusers] communication error ... cannot load packet size

[Laurence Dawson]

OK, so with loglevel 7 the error disappeared. I turned the log level 
back to 2 and it still works. I don't know of anything else that has 
changed. I will reintegrate the nodes into our cluster for now and keep 
an eye on them.

chris wrote:
> Thanks,
>
> Can you now set your loglevel to 7 & 9 and send us the output that
> generates.
>
> ---
> Chris Vaughan
>
> On Fri, 2006-06-23 at 14:57 -0500, Laurence Dawson wrote:
>   
>> It looks OK. They match - vmpsched is our scheduler, vmp002 is one of 
>> the new compute nodes.
>>
>> [root at vmpsched root]# uname -a
>> Linux vmpsched 2.4.21-32.0.1.ELsmp #1 SMP Thu Jun 30 10:52:34 CDT 2005 
>> i686 i686 i386 GNU/Linux
>>
>> [root at vmpsched root]# MOABTEST=checksum:bababa,HMAC64,dog moab
>> CS: 'iEQIQX1nhHaXN8gyn5/TYCbJstk='/'5JUSUk9HtBONhQydnYWXKScoHaA='
>>
>>
>> [root at vmp002 sbin]# MOABTEST=checksum:bababa,HMAC64,dog ./moab
>> CS: 'iEQIQX1nhHaXN8gyn5/TYCbJstk='/'5JUSUk9HtBONhQydnYWXKScoHaA='
>>
>> [root at vmp002 config]# uname -a
>> Linux vmp002.vampire 2.4.21-40.ELsmp #1 SMP Wed Mar 15 14:44:44 EST 2006 
>> i686 athlon i386 GNU/Linux
>>
>>
>>
>> chris wrote:
>>     
>>> Hi Laurence,
>>>
>>> Could you take a look at this posting
>>> http://www.supercluster.org/pipermail/moabusers/2006-May/000174.html and
>>> run those tests on your Moab installation.  Let us know what your
>>> results are.
>>>
>>>
>>> Thanks,
>>>
>>> ---
>>> Chris Vaughan
>>>
>>> _______________________________________________
>>> moabusers mailing list
>>> moabusers at supercluster.org
>>> http://www.supercluster.org/mailman/listinfo/moabusers
>>>   
>>>       
>
>