[GE users] How to clear internal hostname cache?

Joe Landman landman at scalableinformatics.com
Tue Apr 4 17:21:39 BST 2006

    [ 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. ]

Kim Leng Goh wrote:
> Hi Joe,
>   Some update. First, thanks for your script. I edited it to become:
> #!/usr/bin/perl
> @files = `find /`;
> foreach $file (@files)
>  {
>   chomp($grep = `grep -i --exclude="/dev/*" network-0-0 $file`);
>   printf "%s\n",$file if ($grep ne "");
>  }
> Here's the output at time of writing:
> /proc/kcore

Ok... this is just the memory of the machine.  You shouldn't be looking 
in there, as you could be regexing the image of the program that 
contains that string.

I would suggest adding --exclude="/proc/*" --exclude="/sysfs/*"

At this point I think you have two options.

Option 1 is, using Rocks methodology, to re-install the head node and 
re-install the compute nodes.

Option 2 is to decommission 5.3, and hand install 6.0u7.

I don't recommend "1" apart from exceptional cases, it usually causes 
far more interruption than you need.  Option "2" is usually much 
easier/safer, and you can (as long as you run them on separate ports, 
and don't use the same compute nodes as execution hosts) run two 
schedulers at once for a cutover to the new installation.  Given the 
issues in the Rocks SGE (qrsh not working among other bits), I would 
recommend "2".



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 or +1 866 888 3112
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