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

[Condor-users] condor_q fails to fetch classads: Connection refused



I'm setting up a small condor pool and am having trouble with condor_q. I'm using version 7.2 with the collector,negotiator,scheduler on Red Hat 5 and my two test execute nodes running OS X Leopard. I'm trying to setup a single machine to handle all the scheduling. If i run "condor_q -name master.my.domain" from an execute node I get the error message:

########################### BEGIN ##########################
2/23 15:58:11 CONNECT src="" fd=3 dst=<master:9618>
2/23 15:58:11 SECMAN: command 6 QUERY_SCHEDD_ADS to collector at <master:9618> from TCP port 63323 (blocking).
2/23 15:58:11 SECMAN: using session master:30346:1235426291:122 for {<master:9618>,<6>}.
2/23 15:58:11 condor_write(fd=3 <master:9618>,,size=554,timeout=180,flags=0)
2/23 15:58:11 SECMAN: resume, other side is $CondorVersion: 7.2.0 Dec 19 2008 BuildID: 121001 $, NOT reauthenticating.
2/23 15:58:11 SECMAN: startCommand succeeded.
2/23 15:58:11 condor_write(fd=3 <master:9618>,,size=74,timeout=180,flags=0)
2/23 15:58:11 condor_read(fd=3 <master:9618>,,size=5,timeout=180,flags=0)
2/23 15:58:12 condor_read(fd=3 <master:9618>,,size=1056,timeout=180,flags=0)
2/23 15:58:12 CLOSE <worker:63323> fd=3
2/23 15:58:12 attempt to connect to <master:49824> failed: Connection refused (connect errno = 61).
2/23 15:58:12 CLOSE <worker:63326> fd=3
########################### END ##########################

I know the scheduler is running on master and in the CollectorLog I see:

########################### BEGIN ##########################
2/23 15:38:01 condor_read(fd=10 <worker:63244>,,size=5,timeout=30,flags=0)
2/23 15:38:01 condor_read(fd=10 <worker:63244>,,size=105,timeout=30,flags=0)
2/23 15:38:01 Got QUERY_SCHEDD_ADS
2/23 15:38:01 (Sending 1 ads in response to query)
2/23 15:38:01 CLOSE <master:9618> fd=10
2/23 15:38:01 condor_write(fd=10 <worker:63244>,,size=1061,timeout=60,flags=0)
2/23 15:38:01 ACCEPT from=<worker:63245> newfd=10 to=<master:9618>
2/23 15:38:01 condor_read(fd=10 <worker:63245>,,size=4,timeout=1,flags=2)
2/23 15:38:01 condor_read(fd=10 <worker:63245>,,size=5,timeout=1,flags=0)
2/23 15:38:01 condor_read(fd=10 <worker:63245>,,size=549,timeout=1,flags=0)
2/23 15:38:01 condor_read(fd=10 <worker:63245>,,size=5,timeout=30,flags=0)
2/23 15:38:01 condor_read(fd=10 <worker:63245>,,size=69,timeout=30,flags=0)
2/23 15:38:01 Got QUERY_SCHEDD_ADS
2/23 15:38:01 (Sending 1 ads in response to query)
2/23 15:38:01 CLOSE <master:9618> fd=10
########################### END ##########################

Any idea as to why condor_q isn't seeing the class ads being sent to it?

Thanks,
NTroutman