[Condor-users] Good way to start failed jobs from large cluster?

[Carsten Aulbert <carsten.aulbert@xxxxxxxxxx>]

Hi,

as an admin I'm out of condor submit file magic for some time and would
like to know if there is an easy way to accomplish this:

Imagine a user using vanilla universe and large clusters using a submit
file like this:

universe                = vanilla
Arguments               = -j $(Process)
log                     = /home/user/log/$(Process).log
error                   = /home/user/log/$(Process).err
executable              = /home/user/bin/IWillFindIt.exe
notification            = Never
queue 45345

Now imagine this ran for a while but 134 jobs with more or less random
numbers failed, e.g.

5.6, 5.1345, 5.8733, ...

What is a good way to restart only these? So far I help me with this:

for i in `magic_which_will_outpuy_me_process_ids_only`; do
cat <<EOF | condor_submit
universe                = vanilla
Arguments               = -j $i
log                     = /home/user/log/$i.log
error                   = /home/user/log/$i.err
executable              = /home/user/bin/IWillFindIt.exe
notification            = Never
queue
EOF
done

Is there a better way to get this?

Please note: I need to get the log, error as well argument line correctly.

Cheers

Carsten