[Condor-users] ERROR: Failed to create cluster

[Tom Ghekiere <tom.ghekiere@xxxxxxxx>]

Hi all,


I configured a Condor master node and a slave node. On both sides, I can 
see the queue with 'condor_q' and 'condor_status'.

When I try to submit a job (simple "hello Condor") from the master, I get:
$ condor_submit hello.submit
Submitting job(s)
ERROR: Failed to create cluster

Looking into the scheduler's log, reveals:
$ tail -n 10 /home/condor/local.xxxx/log/SchedLog
3/8 16:35:16 (pid:32266) ZKM: setting default map to tom@mo99pool
3/8 16:35:16 (pid:32266) QMGT command failed: no WRITE permission for 
192.168.0.102
3/8 16:47:57 (pid:32266) ZKM: setting default map to tom@mo99pool
3/8 16:47:57 (pid:32266) QMGT command failed: no WRITE permission for 
192.168.0.102
3/8 16:50:59 (pid:32266) ZKM: setting default map to tom@mo99pool
3/8 16:50:59 (pid:32266) QMGT command failed: no WRITE permission for 
192.168.0.102
3/8 16:52:59 (pid:32266) ZKM: setting default map to tom@mo99pool
3/8 16:52:59 (pid:32266) QMGT command failed: no WRITE permission for 
192.168.0.102
3/8 17:44:25 (pid:32266) ZKM: setting default map to tom@mo99pool
3/8 17:44:25 (pid:32266) QMGT command failed: no WRITE permission for 
192.168.0.102

However, this IP address is included in HOSTALLOW_WRITE on both master 
and slave:
$ condor_config_val HOSTALLOW_WRITE
192.168.0.*, 127.0.0.1

When I try to submit from slave, it does manage to submit and execute 
the job:
$ condor_submit hello.submit
Submitting job(s).
Logging submit event(s).
1 job(s) submitted to cluster 3.


Can somebody help me with this, or give me a hint?


Kind regards,

Tom



begin:vcard
fn:Tom Ghekiere
n:Ghekiere;Tom
org:Ghent University-IBBT-IBiTech;Department of Electronics and Information Systems - MEDISIP
adr:;;De Pintelaan 185 (block B);Ghent;;B-9000 ;Belgium
email;internet:tom.ghekiere@xxxxxxxx
tel;work:+32 9 332 43 22 
x-mozilla-html:TRUE
version:2.1
end:vcard