[torqueusers] checkpoint/restart mpi-job on different compute nodes

[TingtingYang]

hi all:   I want to use torque's checkpoint/restart function.I wonder if it possible to checkpoint/restart mpi jobs which run on multi-nodes  with -l nodes=2:ppn=2. right now I can checkpoint/restart mpi jobs which run on one node with  -l nodes=1；ppn=4 and counter some error when I try to checkpoint/restart mpi jobs running on multi-nodesthank you 

                                                                                Tingting Yang from Beihang university
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20120723/9c8090e8/attachment.html