Hi,
NIKHEF has been in CT status on the SFTs for the last two runs ... from
the RM supertest. I managed to "catch" the ops job right before it
exited this time.
The report says that the problem is a timeout ... indeed. The command I
'caught' was an lcg-rep command to castorsrm.cern.ch.
I tried first an lcg-cr command to our local SRM (tbn18.nikhef.nl). It
was slower than I expected (several tens of seconds before I got a
command prompt back) but it worked fine.
Then I tried the exact same command with castorsrm.cern.ch, and I am
still waiting for the prompt. So yes the RM command times out, but I
it's not clear to me that it's a problem on-site. Any ideas?
JT
tbn18:
> node17-23:~> time lcg-cr -v --vo atlas -d tbn18.nikhef.nl -l lfn:/grid/atlas/jeff.tst.49 file:///home/templon/rmdir_if_empty.sh
> Using grid catalog type: lfc
> Using grid catalog : prod-lfc-atlas-central.cern.ch
> Source URL: file:///home/templon/rmdir_if_empty.sh
> File size: 656
> VO name: atlas
> Destination specified: tbn18.nikhef.nl
> Destination URL for copy: gsiftp://hooivork.nikhef.nl/hooivork.nikhef.nl:/export/cache2/atlas/2006-10-20/file3007ec7d-6f72-47e9-8da1-4142fb24ec9d.209484.0
> # streams: 1
> # set timeout to 0 seconds
> Alias registered in Catalog: lfn:/grid/atlas/jeff.tst.49
> 656 bytes 0.35 KB/sec avg 0.35 KB/sec inst
> Transfer took 2050 ms
> Destination URL registered in Catalog: srm://tbn18.nikhef.nl/dpm/nikhef.nl/home/atlas/generated/2006-10-20/file3007ec7d-6f72-47e9-8da1-4142fb24ec9d
> guid:d62666c3-a1f8-4938-bd0b-4a490be42211
> lcg-cr -v --vo atlas -d tbn18.nikhef.nl -l lfn:/grid/atlas/jeff.tst.49 0.58s user 0.07s system 3% cpu 16.655 total
castorsrm:
> node17-23:~> lcg-cr -v --vo atlas -d castorsrm.cern.ch -l lfn:/grid/atlas/jeff.tst.51 file:///home/templon/rmdir_if_empty.sh
> Using grid catalog type: lfc
> Using grid catalog : prod-lfc-atlas-central.cern.ch
> Source URL: file:///home/templon/rmdir_if_empty.sh
> File size: 656
> VO name: atlas
> Destination specified: castorsrm.cern.ch
> Destination URL for copy: gsiftp://lxfsra2405.cern.ch//castor/cern.ch/grid/atlas/generated/2006-10-20/file1bb2cdfc-a364-4fcd-a242-c9c7538e251e
> # streams: 1
> # set timeout to 0 seconds
> Alias registered in Catalog: lfn:/grid/atlas/jeff.tst.51
[ control-C after several minutes waiting ... ]
> Copy Cancelled...es 0.01 KB/sec avg 0.01 KB/sec inst
> Copy Failed: Unregistering alias from catalog.
>
|