[GE users] SGE6 does not backfill

Juha Jäykkä juhaj at iki.fi
Mon Apr 18 11:40:47 BST 2005


    [ The following text is in the "ISO-8859-15" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

> That is good news. Thank you very much. We expected it. However, the bug
> is not fixed. We will need the fix real bug behind it. It is just very,
> very unlikely now. 

The real bug being that extra signal the job gets while in "t" state? Just
curious.

> Is everything else in the 6.0u4beta working for you?
> Anything you noticed?
> Pe jobs should start almost immediately. Can you confirm that?

In short: yes, no, yes. :)

Everything seems to work fine, except that one parallel job got stuck in
MPI_Barrier (because it thought one of the processes in the MPI universe
had died) while I did a cluster-wide "killall sge_execd", reinstall and
restart. It may not be sge-related, though I think it is: it was my own
code, it's never behaved like that before and I see no reason it should
have now.

Also, I have not been able to see a parallel job in "t" state since I
installed the newest CVS HEAD. Thus it is at least faster than I can say
qstat.

> Again, thanks for you cooperation.

Thank you, too. This has been both instructive and very helpful. And the
cluster can now be fully used, once again. Thanks!

-- 
                 -----------------------------------------------
                | Juha Jäykkä, juolja at utu.fi			|
		| Laboratory of Theoretical Physics		|
		| Department of Physics, University of Turku	|
                | home: http://www.utu.fi/~juolja/              |
                 -----------------------------------------------


    [ Part 2, Application/PGP-SIGNATURE 196 bytes. ]
    [ Unable to print this part. ]



More information about the gridengine-users mailing list