torque+maui问题-任务不运行

这次因停电服务器关机重启后发现pbs系统不能正常工作了，提上去任务显示Q状态，就是不运行。

pbs_server进程、pbs_mom进程及trqauthd进程重启多次都无果，无意中使用了showq命令，提示
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

ERROR:    cannot send request to server logon:42559 (server may not be running)
ERROR:    cannot request service (status)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

然后就在这里找到了答案：http://www.supercluster.org/pipermail/torqueusers/2011-June/012986.html

=================================================================================

[torqueusers] still problem with FQDN and torque


Mahmood Naderan nt_mahmood at yahoo.com 
Wed Jun 8 23:07:30 MDT 2011
Previous message: [torqueusers] still problem with FQDN and torque
Next message: [torqueusers] TORQUE 2.4.14 available. Contains fix for bufferoverflow risk
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
>that is maui, not torque. does maui run?


You are right. I didn't start maui.
it is now ok. Thanks for your help :)


// Naderan *Mahmood;




----- Original Message -----
From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
To: Mahmood Naderan <nt_mahmood at yahoo.com>
Cc: torque cluster <torqueusers at supercluster.org>
Sent: Thursday, June 9, 2011 12:48 AM
Subject: Re: [torqueusers] still problem with FQDN and torque


On Wed, Jun 8, 2011 at 1:18 PM, Mahmood Naderan <nt_mahmood at yahoo.com> wrote:
>>did you restart the servers?
>
> Even I reboot the server but I get
>
> root at srv:/var/spool/maui# /etc/init.d/pbs_mom restart
>  * Restarting TORQUE mom pbs_mom                                                       [ OK ]
>
> root at srv:/var/spool/maui# showq
> ERROR:    cannot send request to server srv:42559 (server may not be running)
> ERROR:    cannot request service (status)


that is maui, not torque. does maui run?




>>depends on your torque configuration.
>
> Exactly what?


for example: /var/spool/torque/server_name


axel.


>
>
> // Naderan *Mahmood;
>
>
> ----- Original Message -----
> From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
> To: Mahmood Naderan <nt_mahmood at yahoo.com>
> Cc: torque cluster <torqueusers at supercluster.org>
> Sent: Wednesday, June 8, 2011 8:29 PM
> Subject: Re: [torqueusers] still problem with FQDN and torque
>
> On Wed, Jun 8, 2011 at 11:18 AM, Mahmood Naderan <nt_mahmood at yahoo.com> wrote:
>>
>> How about changeing  srv.domain.com
>>
>> to srv-ex.domain.com ? I think it is much easier. Do you agree with that?
>
> yes.
>
>> I change that however  the server_log still says
>>
>> 06/08/2011 19:43:52;0001;PBS_Server;Svr;PBS_Server;LOG_ERROR::is_request, bad attempt to connect from 192.168.1.1:1023 (address not trusted - check entry in server_priv/nodes)
>
> did you restart the servers?
>
>
>> It is very strange for me and I can not understand.
>>
>> root at srv:/var/spool/pbs/server_logs# ping srv
>> PING srv (192.168.1.1) 56(84) bytes of data.
>> 64 bytes from srv (192.168.1.1): icmp_seq=1 ttl=64 time=0.092 ms
>>
>>
>> As you can see the IP address of srv is 192.168.1.1
>> also the server_priv/nodes contains
>>
>> srv np=14
>>
>> however I am really confused why it says "192.168.1.1 address not trusted" !!!!
>
> depends on your torque configuration.
> again, did you restart after changing /etc/hosts and DNS entries?
>
> axel.
>
>> Please assume
>>  192.168.1.1     srv
>>  196.215.64.105 srv-ex.domain.com
>>
>> // Naderan *Mahmood;
>>
>>
>> ----- Original Message -----
>> From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
>> To: Mahmood Naderan <nt_mahmood at yahoo.com>
>> Cc: torque cluster <torqueusers at supercluster.org>
>> Sent: Wednesday, June 8, 2011 7:25 PM
>> Subject: Re: [torqueusers] still problem with FQDN and torque
>>
>> On Wed, Jun 8, 2011 at 9:11 AM, Mahmood Naderan <nt_mahmood at yahoo.com> wrote:
>>> root at srv:~# cat /etc/hosts
>>> 127.0.0.1       localhost.localdomain localhost
>>> 192.168.1.1     srv-internal
>>> 196.215.64.105 srv.domain.com
>>>
>>>
>>> root at srv:~# cat /var/spool/pbs/server_priv/nodes
>>> srv-internal np=14
>>> ...
>>>
>>>
>>> root at srv:~# cat /var/spool/pbs/server_name
>>> srv-internal
>>>
>>> root at srv:~# /etc/init.d/pbs_mom restart
>>>  * Restarting TORQUE mom pbs_mom                                                       [ OK ]
>>>
>>> root at srv:~# showq
>>> ERROR:    cannot connect to 'srv' port 42559
>>
>>>
>>> Still it is looking for srv.....
>>
>> sure. did you change your configuration to
>> connect to srv-internal?
>> you have to do this and the corresponding
>> changes in the hosts files on _all_ compute
>> nodes, mom config files, pbs_server and so on.
>> you want the batch system communication only
>> on the internal network, so you have to use
>> the proper names to get it to use the right
>> ip numbers. elementary.
>>
>> axel.
>>
>>>
>>> // Naderan *Mahmood;
>>>
>>>
>>> ----- Original Message -----
>>> From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
>>> To: Mahmood Naderan <nt_mahmood at yahoo.com>
>>> Cc: torque cluster <torqueusers at supercluster.org>
>>> Sent: Wednesday, June 8, 2011 5:30 PM
>>> Subject: Re: [torqueusers] still problem with FQDN and torque
>>>
>>> On Wed, Jun 8, 2011 at 8:50 AM, Mahmood Naderan <nt_mahmood at yahoo.com> wrote:
>>>>>have you tried using
>>>>>something like this instead?
>>>>>192.168.1.1 srv-internal
>>>>
>>>> You mean the hostname for the invalid ip should be different from valid ip?
>>>>
>>>>
>>>>  127.0.0.1           localhost.localdomain localhost
>>>>  192.168.1.1        srv-internal
>>>>  196.215.64.105 srv.domain.com
>>>>
>>>> you mean that?
>>>
>>> yes.
>>>
>>>>
>>>> // Naderan *Mahmood;
>>>>
>>>>
>>>> ----- Original Message -----
>>>> From: Axel Kohlmeyer <akohlmey at cmm.chem.upenn.edu>
>>>> To: Mahmood Naderan <nt_mahmood at yahoo.com>; Torque Users Mailing List <torqueusers at supercluster.org>
>>>> Cc:
>>>> Sent: Wednesday, June 8, 2011 4:58 PM
>>>> Subject: Re: [torqueusers] still problem with FQDN and torque
>>>>
>>>> On Wed, Jun 8, 2011 at 6:09 AM, Mahmood Naderan <nt_mahmood at yahoo.com> wrote:
>>>>> Dear all
>>>>>
>>>>> I have asked about FQDN before while torque is running however somethings are still vague that causes the server node to be "down".
>>>>>
>>>>> The server has two NIC. One is connected to internet with valid IP and the other is connected to a local switch with invalid IP. The content of /etc/hosts looks like
>>>>>
>>>>> 127.0.0.1           localhost.localdomain localhost
>>>>> 192.168.1.1        srv
>>>>> 196.215.64.105  srv.domain.com
>>>>
>>>> i think it is a bad idea to have "srv" for the internal name
>>>> and "srv.<something" for the external one. this will defeat
>>>> attempts to canonicalize hostnames. have you tried using
>>>> something like this instead?
>>>>
>>>> 192.168.1.1 srv-internal
>>>>
>>>> axel.
>>>>
>>>>>
>>>>> some information:
>>>>>
>>>>> mahmood at srv:~$ cat /var/spool/pbs/server_name
>>>>> srv
>>>>>
>>>>>
>>>>> root at srv:~# cat /var/spool/pbs/server_priv/nodes
>>>>> srv np=14
>>>>> ...
>>>>>
>>>>>
>>>>> mahmood at srv:~$ pbsnodes -a srv
>>>>> srv
>>>>>      state = down
>>>>>      np = 14
>>>>>      ntype = cluster
>>>>>      status = rectime=1307513072,varattr=,jobs=26463.srv,state=free,netload=359025735787,gres=,loadave=0.34,ncpus=16,physmem=33013828kb,availmem=11987350430564kb,totmem=12345261654628kb,idletime=8,nusers=1,nsessions=4,sessions=21685 21758 22420 27547,uname=Linux srv 2.6.32-24-server #39-Ubuntu SMP Wed Jul 28 06:21:40 UTC 2010 x86_64,opsys=linux
>>>>>      mom_service_port = 15002
>>>>>      mom_manager_port = 15003
>>>>>      gpus = 0
>>>>>
>>>>>
>>>>>
>>>>> My question is hy the srv is "down". What is wrong?
>>>>>
>>>>> // Naderan *Mahmood;
>>>>> _______________________________________________
>>>>> torqueusers mailing list
>>>>> torqueusers at supercluster.org
>>>>> http://www.supercluster.org/mailman/listinfo/torqueusers
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Dr. Axel Kohlmeyer    akohlmey at gmail.com
>>>> http://sites.google.com/site/akohlmey/
>>>>
>>>> Institute for Computational Molecular Science
>>>> Temple University, Philadelphia PA, USA.
>>>>
>>>>
>>>
>>>
>>>
>>> --
>>> Dr. Axel Kohlmeyer    akohlmey at gmail.com
>>> http://sites.google.com/site/akohlmey/
>>>
>>> Institute for Computational Molecular Science
>>> Temple University, Philadelphia PA, USA.
>>>
>>>
>>
>>
>>
>> --
>> Dr. Axel Kohlmeyer    akohlmey at gmail.com
>> http://sites.google.com/site/akohlmey/
>>
>> Institute for Computational Molecular Science
>> Temple University, Philadelphia PA, USA.
>>
>>
>
>
>
> --
> Dr. Axel Kohlmeyer    akohlmey at gmail.com
> http://sites.google.com/site/akohlmey/
>
> Institute for Computational Molecular Science
> Temple University, Philadelphia PA, USA.
>
>






-- 
Dr. Axel Kohlmeyer    akohlmey at gmail.com
http://sites.google.com/site/akohlmey/


Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.


Previous message: [torqueusers] still problem with FQDN and torque
Next message: [torqueusers] TORQUE 2.4.14 available. Contains fix for bufferoverflow risk
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the torqueusers mailing list