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

Re: [Condor-users] jobs are not being distributed among the machines



Alex,
 
If your jobs are writing rank or constraint expressions that reference "vm#" you need to change them to "slot#". The Condor team switched to "slot" to remove the ambiguity with machine virtualization and "virtual machines" done by software like Xen and VMWare.
 
You can force Condor back to using "vm#" by setting:
 
STARTD_RESOURCE_PREFIX = vm
 
I also found some pecularities in privledge separation in 7.0.5 vs 6.8.6. See my emails to the list from Friday evening. If you're using dedicated domain accounts to run jobs the syntax has changed for setting up the accounts. And I've found preemption to be many, many times slower in 7.0.5. Too slow for my needs -- I've had to revert back to 6.8.6.
 
- Ian


From: condor-users-bounces@xxxxxxxxxxx [mailto:condor-users-bounces@xxxxxxxxxxx] On Behalf Of Alas, Alex [FEDI]
Sent: Monday, November 24, 2008 11:01 AM
To: Condor-Users Mail List
Subject: [Condor-users] jobs are not being distributed among the machines

I recently upgraded my windows condor pool from 6.8.7 to 7.0.5 my central manager and all the execute/submit nodes. I can submit/execute jobs fine locally and through the network by using condor_store_cred configuration. On all the nodes after the upgrade I copied the old condor_config file to guarantied my previous configuration will remain intact and it worked because the functionality of my pool remains operational.

On some jobs that require to run a batch file multiple times I noticed once the job claims the first available node it will not move to the next available node. Since the upgrade the systems that in the past were listed as vm1, vm2 are now being displayed as slot1, slot2… I don’t know if the new upgrade had something to do with it but want to formulate the question if there is anything I need to addition in the condor_config file to make  

Thanks for your time in advance,

 

Respectfully,

Alex Alas

Systems Administrator
Fugro EarthData Inc.

Tel. 301-948-8550 x219 Fax 301-963-2064 E-mail: aalas@xxxxxxxxxxxxx

7320 Executive Way, Frederick, MD  21704

Website: http://www.fugroearthdata.com

 



Confidentiality Notice.
This message may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient, you are hereby notified that any use, disclosure, dissemination, distribution, or copying of this message, or any attachments, is strictly prohibited. If you have received this message in error, please advise the sender by reply e-mail, and delete the message and any attachments. Thank you.