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

[Condor-users] condor flocking problem



Hi,

First of all thanks a lot to Prashant, Jaime for the quick response, it did solve my problem.

I have go this peculiar problem with my condor install. I am trying to flock some jobs from one machine to another. This a simple job with Vanilla Universe. It does no require much resources and its just a simple shell script. The jobs are flocking alright to one machine but I am getting this problem on 2 machines on the network.

Here the machine 10.1.30.105 is trying to flock jobs to 10.1.30.25, 10.1.30.99, 10.1.30.105

Jobs run successfully on 10.1.30.25 but not on the other 2 machines.

###########################################################################

3/14 18:49:09 DaemonCore: Command received via TCP from host <10.1.30.105:1934>
3/14 18:49:09 DaemonCore: received command 444 (ACTIVATE_CLAIM), calling handler (command_activate_claim)
3/14 18:49:09 Got activate_claim request from shadow (<10.1.30.105:1934>)
3/14 18:49:09 Remote job ID is 211.27
3/14 18:49:09 Got universe "VANILLA" (5) from request classad
3/14 18:49:09 State change: claim-activation protocol successful
3/14 18:49:09 Changing activity: Idle -> Busy
3/14 18:49:09 Starter pid 20286 exited with status 1
3/14 18:49:09 State change: starter exited
3/14 18:49:09 Changing activity: Busy -> Idle
3/14 18:49:09 DaemonCore: Command received via UDP from host <10.1.30.105:1048>
3/14 18:49:09 DaemonCore: received command 443 (RELEASE_CLAIM), calling handler (command_handler)
3/14 18:49:09 State change: received RELEASE_CLAIM command
3/14 18:49:09 Changing state and activity: Claimed/Idle -> Preempting/Vacating
3/14 18:49:09 State change: No preempting claim, returning to owner
3/14 18:49:09 Changing state and activity: Preempting/Vacating -> Owner/Idle
3/14 18:49:09 State change: IS_OWNER is false
3/14 18:49:09 Changing state: Owner -> Unclaimed
3/14 18:49:09 DaemonCore: Command received via UDP from host <10.1.30.105:1048>
3/14 18:49:09 DaemonCore: received command 443 (RELEASE_CLAIM), calling handler (command_handler)
3/14 18:49:09 Error: can't find resource with capability (<10.1.30.99:2705>#1421033947)
3/14 18:54:02 DaemonCore: Command received via UDP from host <10.1.30.99:1543>
3/14 18:54:02 DaemonCore: received command 440 (MATCH_INFO), calling handler (command_match_info)
3/14 18:54:02 match_info called


###########################################################################\


3/14 18:35:40 DaemonCore: Command received via TCP from host <10.1.30.105:2001>
3/14 18:35:40 DaemonCore: received command 444 (ACTIVATE_CLAIM), calling handler (command_activate_claim)
3/14 18:35:40 Got activate_claim request from shadow (<10.1.30.105:2001>)
3/14 18:35:40 Remote job ID is 211.27
3/14 18:35:40 Got universe "VANILLA" (5) from request classad
3/14 18:35:40 State change: claim-activation protocol successful
3/14 18:35:40 Changing activity: Idle -> Busy
3/14 18:35:40 Starter pid 2712 exited with status 1
3/14 18:35:40 State change: starter exited
3/14 18:35:40 Changing activity: Busy -> Idle
3/14 18:35:40 DaemonCore: Command received via UDP from host <10.1.30.105:1048>
3/14 18:35:40 DaemonCore: received command 443 (RELEASE_CLAIM), calling handler (command_handler)
3/14 18:35:40 State change: received RELEASE_CLAIM command
3/14 18:35:40 Changing state and activity: Claimed/Idle -> Preempting/Vacating
3/14 18:35:40 State change: No preempting claim, returning to owner
3/14 18:35:40 Changing state and activity: Preempting/Vacating -> Owner/Idle
3/14 18:35:40 State change: IS_OWNER is false
3/14 18:35:40 Changing state: Owner -> Unclaimed
3/14 18:35:40 DaemonCore: Command received via UDP from host <10.1.30.105:1048>
3/14 18:35:40 DaemonCore: received command 443 (RELEASE_CLAIM), calling handler (command_handler)
3/14 18:35:40 Error: can't find resource with capability (<10.1.30.206:1027>#7940822160)


##########################################################################
Does the network traffic affect the flocking of jos ?

Thanks in advance,

Chaitanya Hazarey