Discussion:
Accounting File Missing
(too old to reply)
adarsh
2010-11-29 06:42:45 UTC
Permalink
Raw Message
Hi,

Yesterday I ran several simple jobs in SGE and monitor through QMON for their successful completion.

But today when I ran a simple job sleeper.sh,it remain in hqw state. FOllowing is my configuration :
Qmaster
OS - CentOS
Hadoop - Hadoop-0.20.2
Execd hosts
OS - Ubuntu 10.4

When i ran qacct command it shows
***@ws19-sub-lin:~# /opt/sge-root/bin/lx24-amd64/qacct -j 6
/opt/sge-root/default/common/accounting: No such file or directory
no jobs running since startup

When I go through Execution Host directories accounting file is missing.

I don't how it happens and to do ?

Thanks & Regards
Adarsh Sharma

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=300145

To unsubscribe from this discussion, e-mail: [users-***@gridengine.sunsource.net].
reuti
2010-11-29 09:16:39 UTC
Permalink
Raw Message
Hi,
Post by adarsh
Hi,
Yesterday I ran several simple jobs in SGE and monitor through QMON for their successful completion.
But today when I ran a simple job sleeper.sh,it remain in hqw state.
you will have to investigate why. Did you submit it with a hold, of jobid_hold?
Post by adarsh
Qmaster
OS - CentOS
Hadoop - Hadoop-0.20.2
Execd hosts
OS - Ubuntu 10.4
When i ran qacct command it shows
/opt/sge-root/default/common/accounting: No such file or directory
no jobs running since startup
When I go through Execution Host directories accounting file is missing.
There is only one accounting file, which is central. Is /opt/sge-root/default/common shared between all nodes?

-- Reuti
Post by adarsh
I don't how it happens and to do ?
Thanks & Regards
Adarsh Sharma
------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=300145
------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=300165

To unsubscribe from this discussion, e-mail: [users-***@gridengine.sunsource.net].
adarsh
2010-12-07 04:38:05 UTC
Permalink
Raw Message
Is it necessary to share defaut directory through NFS , i simply copied qmaster's defaut directory through scp to all execution hosts.

Does it affects or it is wrong configuration.

However, my jobs are running.


Thanks & Regards

Adarsh Sharma

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=302643

To unsubscribe from this discussion, e-mail: [users-***@gridengine.sunsource.net].
reuti
2010-12-07 13:12:54 UTC
Permalink
Raw Message
Hi,
Post by adarsh
Is it necessary to share defaut directory through NFS ,
no, not necessarily.
Post by adarsh
i simply copied qmaster's defaut directory through scp to all execution hosts.
But some details you can find here:

http://gridengine.sunsource.net/howto/nfsreduce.html

As the accounting file is written by the qmaster, only on the master node the `qacct` will work, because there is no accounting information on any node.

-- Reuti
Post by adarsh
Does it affects or it is wrong configuration.
However, my jobs are running.
Thanks & Regards
Adarsh Sharma
------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=302643
------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=302771

To unsubscribe from this discussion, e-mail: [users-***@gridengine.sunsource.net].
adarsh
2010-12-08 04:59:27 UTC
Permalink
Raw Message
Thanx for your reply.


Best Regards

Adarsh Sharma

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=303036

To unsubscribe from this discussion, e-mail: [users-***@gridengine.sunsource.net].
adarsh
2010-12-07 04:37:04 UTC
Permalink
Raw Message
Is it necessary to share defaut directory through NFS , i simply copied qmaster's defaut directory through scp to all execution hosts.

Does it affects or it is wrong configuration.

However, my jobs are running.


Thanks & Regards

Adarsh Sharma

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=302642

To unsubscribe from this discussion, e-mail: [users-***@gridengine.sunsource.net].
Loading...