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

Re: [Condor-users] pool drainoff



> condor_off -peaceful <nodename>
> on a single node does work for me for this purpose. (and also 
> worked in 6.8.x and 6.9.x) If it works right the nodes status 
> should show up in condor_status as "retiring", all of the 
> VM's both busy and idle.  If it doesn't work right, that 
> means that the command failed quietly due to lack of 
> privilege to run it.  If your pool is gsi-authenticated then 
> you have to have a grid proxy to run that command.
> If it's normal host/ip authentication, you have to be on the 
> collector/negotiator to make the command work, it won't work 
> if you are logged into the startd node itself (and it won't 
> give you any errors saying that it didn't work).

I don't use GSI authentication in my pool and all administrative
commands are set to work from the local machine or from our negotiator.
And all other command (like condor_vacate) do indeed work properly from
the localhost. But not condor_off -peaceful. Waiting for 1 long job to
finish the machine starts to run other jobs in the empty slots.

- Ian


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.