<br><font size=2 face="Arial">HI, &nbsp;everyone</font>
<br><font size=2 face="Arial"><br>
</font><font size=2 color=#000080 face="Arial">I used blcr-0.7.3 and torque
</font><a href="http://www.clusterresources.com/downloads/torque/temp/torque-2.4.0-snap.200809111541.tar.gz"><font size=2 color=blue face="Arial"><u>torque-2.4.0-snap.200809111541.tar.gz</u></font></a><font size=2 face="Arial">
&nbsp;to test the checkpoint/restart function according to </font>
<br>
<br><font size=2 face="Arial">the wiki: http://www.clusterresources.com/wiki/doku.php?id=torque:2.6_job_checkpoint_and_restart</font>
<br>
<br>
<br><font size=2 face="Arial">I found an insteresting question, when I
qhold the job, I'll see the checkpoint file located at /var/spool/torque/checkpoint/4817.node24.CK/ckpt.4817.node24.1221666102
</font>
<br>
<br><font size=2 face="Arial">but when I qrls the same job 4817, the pbs_mom
daemon at the compute node will down (killed by something). &nbsp;Any clues?
&nbsp;</font>
<br>
<br><font size=2 face="Arial">Thank you very much.</font>
<br>
<br>
<br><font size=2 face="Arial">dolphin ,qin </font>
<br><font size=2 color=#000080 face="&#23435;&#20307;">&nbsp;</font>
<br><font size=2 color=#000080 face="&#23435;&#20307;">&nbsp;</font>