[GE users] sge v6.0u3 new installation issue with more than 1021 hosts.

Joe Landman landman at scalableinformatics.com
Thu Mar 10 15:47:41 GMT 2005


    [ The following text is in the "ISO-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Hi Mac:

   What does  "cat /proc/sys/fs/file-nr"  report before and after you 
run  qmaster?

   Also, if "cat /proc/sys/fs/file-max" is under 100,000, you might want 
to boost it up a bit.

   Also look at "cat /proc/sys/kernel/shmmax", and "cat 
/proc/sys/kernel/shmall" .  You want shmmax about the same size as your 
system memory, and shmall should be similarly sized.

Joe


McCalla, Mac wrote:
>>I assume you run the qmaster as root...
> 
> Yes.
> 
> 
>>Can you set the "descriptors" limit to something large, for both the
> 
> soft
> 
>>and hard limit, and then start qmaster from the interactive session??
> 
> Yes, although I thought 4096 was "large" considering the recommendation
> of something
> like number of hosts + number of dynamic connections + 20 i found in the
> archives....8~).
> 
> ulimit -Hn 32759; ulimit -Sn 32759 .
> Startup successful..  qmaster msg |I|qmaster will use max.32739 file
> descriptors for communication.
> 
> results are same.........8~(
> 
> mac 
> 
> 
> 
> -----Original Message-----
> From: raysonho at eseenet.com [mailto:raysonho at eseenet.com] 
> Sent: Thursday, March 10, 2005 9:09 AM
> To: users at gridengine.sunsource.net
> Subject: Re: [GE users] sge v6.0u3 new installation issue with more than
> 1021 hosts.
> 
> I assume you run the qmaster as root...
> 
> Can you set the "descriptors" limit to something large, for both the
> soft
> and hard limit, and then start qmaster from the interactive session??
> 
> Rayson
> 
> 
> 
>>when the number of hosts actually
>>connected by execd passed from 1021 to 1022,
>>i noticed that qmaster stopped responding on port 538 to any further
>>requests from additional execd's or commands (qstat,qhost
>>,etc).   the ulimit for fd's is set at 4096 at qmaster startup (the
> 
> info
> 
>>message at qmaster startup says qmaster will use 4076 file
>>descriptors for communication).  Has anyone else see this problem or
>>have a 6.0u3 installation with more hosts?  
> 
> ---------------------------------------------------------
> Get your FREE E-mail account at http://www.eseenet.com !
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net

-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
phone: +1 734 786 8423
fax  : +1 734 786 8452
cell : +1 734 612 4615


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list