[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Condor-users] an issue in heterogeneous pool

I submit a job to worker nodes which have different architecture from that of submit box, and the submited jobs is always in idle state.
The following is the log of sched:
9/18 23:21:21 (pid:5089) condor_read(): recv() returned -1, errno = 104, assuming failure reading 21 bytes from unknown source.
9/18 23:21:21 (pid:5089) IO: Failed to read packet header
9/18 23:21:21 (pid:5089) Response problem from startd on <> (match <>#1250678162#1#...).
9/18 23:21:21 (pid:5089) Sent RELEASE_CLAIM to startd at <>
9/18 23:21:21 (pid:5089) Match record (<>, 8083, 0) deleted
Is there anyone who can give me some suggestions? Thank you.