with lam and gridengine

Date view Thread view Subject view Author view Attachment view

From: Jerry Mersel (jerry.mersel_at_weizmann.ac.il)
Date: Thu Nov 20 2008 - 05:40:47 PST

  • Next message: Jerry Mersel: "Re: callback function for parallel apps." 
    Hi all:


 I need to checkpoint my applications. To do that I integrated BLCR 0.6.4
 with lam 7.1.4.

  From the command line checkpointing/restarting works okay.

  However when I try doing so from gridengine I can't restart the
application.
  Even from the command line, using the checkpointed files, gotten under
gridengine do not work.


 In my /var/log/messages I see a message kernel: socket skipped.


  Any advice/help would be greatly appreciated.

                                  Best regards,
                                    Jerry
  • Next message: Jerry Mersel: "Re: callback function for parallel apps."
    Date view Thread view Subject view Author view Attachment view