Hello,
We're (well our users) are having srm troubles again, they're seeing
error messages like this:
2006-08-04 03:26:41(PDT) lcg0477.gridpp.rl.ac.uk lcgCopy ERROR lcg-cp
failed: lcg-cp --vo atlas -t 1800
srm://fal-pygrid-20.lancs.ac.uk/pnfs/lancs.ac.uk/data/atlas//dq2/csc11/csc11.005200.T1_McAtNlo_Jimmy.evgen.EVNT.v11004205/csc11.005200.T1_McAtNlo_Jimmy.evgen.EVNT.v11004205._00246.pool.root.4
file:`pwd`/csc11.005200.T1_McAtNlo_Jimmy.evgen.EVNT.v11004205._00246.pool.root.4
256 lcg_cp: Transport endpoint is not connected
In our billing logs we see:
08.04 15:21:26 [pool:fal-pygrid-28_5@fal-pygrid-28Domain:transfer]
[00030000000000000093F7D0,1043289667] atlas:atlas@osm 805044224
10892552 false {GFtp-1.0 fal-pygrid-24.lancs.ac.uk 48721}
{33:"Unexpected Exception : java.net.SocketException: Connection
reset"}
08.04 19:03:45 [pool:fal-pygrid-25_5@fal-pygrid-25Domain:transfer]
[0003000000000000009A9840,869754754] atlas:atlas@osm 291504128 82090
false {GFtp-1.0 fal-pygrid-25.lancs.ac.uk 48918} {33:"Unexpected
Exception : java.net.SocketException: Connection reset"}
I have yet to personally see this trouble myself. Any tests or
anything people can prescribe to investigate this further? Our srm is
being moved again on monday morning, so we'll be restarting
everything, which might get rid of the problem before it can be
diagnosed.
Thanks in advance for any sagely wisdom,
Matt
|