Re: [HTCondor-users] DAGman point - although no big deal

[Ian Cottam <Ian.Cottam@xxxxxxxxxxxxxxxx>]

On 19/02/2013 15:45, "Nathan Panike" <nwp@xxxxxxxxxxx> wrote:

>On Tue, Feb 19, 2013 at 02:39:48PM +0000, Ian Cottam wrote:
>> On 19/02/2013 14:30, "Nathan Panike" <nwp@xxxxxxxxxxx> wrote:
>> 
>> >My thought is that such a check does not make sense for a scheduler or
>> >local universe job such as DAGMan. Adding something to your script to
>> >check that the job universe is neither scheduler nor local would be a
>> >workaround.
>> OK, will do this. Thx.
>> >
>> >Out of curiosity: What happens when condor_dagman turns around and
>> >submits to the schedd? How does the user receive the warning message?
>
>> Don't follow you here.
>
>> -Ian



>
>DAGMan does not edit the submit file before submitting; my question is,
>"What
>happens if the user does not include a Request_Memory in the submit
>file?"  condor_lint will then issue a warning, but it seems like there is
>no way for the user to get the warning message.
>
>Nathan Panike
>
>
>

It looks like the warnings our condor_lint produce end up in a dagman .out
file, rather buried amongst lots of other messages. Thanks for pointing
this out: I'll think some more.
-Ian


-- 
Ian Cottam
IT Services -- supporting research
Faculty of Engineering and Physical Sciences
The University of Manchester
"The only strategy that is guaranteed to fail is not taking risks." Mark
Zuckerberg