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

RE: [Condor-users] Does using LOWPORT/HIGHPORT restrict the number ofjobs a schedd can run?



> >From section 3.3.2, if I use LOWPORT and HIGHPORT to 
> restrict the ports
> that condor can use does this limit the number of jobs I can 
> run from a single schedd since there are a limited number of 
> ports now for condor_shadow's to choose from? Or do 
> LOWPORT/HIGHPORT not affect condor_shadow? What about the 
> number of open connections that condor_schedd can maintain? 
> Would they be limited by the LOWPORT/HIGHPORT settings?

I think I can answer my own question: jobs will fail to start. I set
LOWPORT=9600, HIGHPORT=9650 on my schedd machine and now I'm seeing:

2/9 19:11:41 (6.59) (3276): Sock::bindWithin - failed to bind any port
within (9600 ~ 9650)
2/9 19:11:41 (6.59) (3276): DCStartd::activateClaim: Failed to send
command ACTIVATE_CLAIM to the startd
2/9 19:11:41 (6.59) (3276): Job 6.59 is being evicted
2/9 19:11:41 (6.59) (3276): Sock::bindWithin - failed to bind any port
within (9600 ~ 9650)
2/9 19:11:41 (6.59) (3276): RemoteResource::killStarter(): Could not
send command to startd
2/9 19:11:41 (6.59) (3276): logEvictEvent with unknown reason (108),
aborting
2/9 19:11:41 (6.59) (3276): Sock::bindWithin - failed to bind any port
within (9600 ~ 9650)
2/9 19:11:41 (6.59) (3276): Can't connect to queue manager:
CEDAR:6001:Failed to connect to <137.57.142.135:9615>
2/9 19:11:41 (6.59) (3276): Failed to update job queue!
2/9 19:11:41 (6.59) (3276): **** condor_shadow (condor_SHADOW) EXITING
WITH STATUS 108

But it looks like the schedd is hanging on to a good 75% of the ports in
that range while running jobs are preempted for this machine:

condor_master.exe:2688	TCP	ttc-ichesal3.altera.com:9638
TTC-ICHESAL3:0	LISTENING	
condor_master.exe:2688	UDP	ttc-ichesal3.altera.com:9638	*:*

condor_master.exe:2688	TCP	ttc-ichesal3.altera.com:9621
TTC-ICHESAL3:0	LISTENING	
condor_master.exe:2688	TCP	ttc-ichesal3.altera.com:9631
TTC-ICHESAL3:0	LISTENING	
condor_master.exe:2688	UDP	ttc-ichesal3.altera.com:9621	*:*

condor_master.exe:2688	UDP	ttc-ichesal3.altera.com:9631	*:*

condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9635
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9640
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9620
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9623
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9633
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9643
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9636
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9611
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9644
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9642
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9607
ttc-jpeters.altera.com:4501	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9648
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9615
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9615	*:*

condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9600
ttc-bs1000-044.altera.com:2418	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9601
ttc-bs1000-072.altera.com:1442	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9602
ttc-bs1000-044.altera.com:2418	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9604
ttc-bs1000-074.altera.com:2010	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9605
ttc-bs1000-034.altera.com:3421	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9606
ttc-bs1000-048.altera.com:4717	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9609
ttc-bs1000-034.altera.com:3421	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9610
ttc-bs1000-066.altera.com:3143	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9613
ttc-bs1000-082.altera.com:4191	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9614
ttc-bs1000-066.altera.com:3143	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9616
ttc-bs1000-036.altera.com:4724	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9617
ttc-bs1000-076.altera.com:3509	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9619
ttc-bs1000-071.altera.com:2751	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9624
ttc-bs1000-052.altera.com:3294	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9625
ttc-bs1000-085.altera.com:2629	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9626
ttc-bs1000-067.altera.com:2467	ESTABLISHED	
condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9650
ttc-bs1000-062.altera.com:4147	ESTABLISHED	
condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9648	*:*

condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9632
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9632	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9642	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9644	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9611	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9636	*:*

condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9631
ttc-condorsrv.altera.com:42282	ESTABLISHED	
condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9643	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9633	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9620	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9623	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9640	*:*

condor_schedd.exe:3140	TCP	ttc-ichesal3.altera.com:9645
TTC-ICHESAL3:0	LISTENING	
condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9645	*:*

condor_schedd.exe:3140	UDP	ttc-ichesal3.altera.com:9635	*:*

condor_shadow.exe:1428	TCP	ttc-ichesal3.altera.com:9628
ttc-bs1000-051.altera.com:4846	ESTABLISHED	
condor_shadow.exe:1684	TCP	ttc-ichesal3.altera.com:9618
ttc-bs3066-190.altera.com:1062	ESTABLISHED	
condor_shadow.exe:2396	TCP	ttc-ichesal3.altera.com:9612
ttc-eahmed3.altera.com:1648	ESTABLISHED	
condor_shadow.exe:2788	TCP	ttc-ichesal3.altera.com:9639
ttc-bs1000-071.altera.com:2751	ESTABLISHED	
condor_shadow.exe:2884	TCP	ttc-ichesal3.altera.com:9641
ttc-bs3066-183.altera.priv.altera.com:1274	ESTABLISHED	
condor_shadow.exe:2928	TCP	ttc-ichesal3.altera.com:9634
ttc-bs3066-183.altera.priv.altera.com:1274	ESTABLISHED	
condor_shadow.exe:3388	TCP	ttc-ichesal3.altera.com:9637
ttc-bs1000-067.altera.com:2467	ESTABLISHED	
condor_shadow.exe:3684	TCP	ttc-ichesal3.altera.com:9627
ttc-bs1000-055.altera.com:4310	ESTABLISHED	
condor_shadow.exe:372	TCP	ttc-ichesal3.altera.com:9608
ttc-bs1000-032.altera.com:2482	ESTABLISHED	
condor_shadow.exe:3764	TCP	ttc-ichesal3.altera.com:9646
ttc-tvanderh2.altera.com:1052	ESTABLISHED	
condor_shadow.exe:3812	TCP	ttc-ichesal3.altera.com:9622
ttc-bs3066-187.altera.com:1124	ESTABLISHED	
condor_shadow.exe:4076	TCP	ttc-ichesal3.altera.com:9629
ttc-bs1000-049.altera.com:3948	ESTABLISHED	
condor_shadow.exe:528	TCP	ttc-ichesal3.altera.com:9649
ttc-bs3066-179.altera.com:1034	ESTABLISHED	

Why won't the schedd give up a port for the shadow so it can start
running a job?

- Ian