Hello,
I have a script which create 2 jobs A & B.
B has a dependency on A. (B should start when A is over).
B has a higher priority than A. (A prio is 50, and B prio is 90)
I launch my script 2 times, with at least 1 second between each.
I get jobs : A1, B1 and A2, B2
On our repository job scheduling order is ‘Pool, Priority, First-In First-Out’.
I thought my job would be treated in that order :
A1, B1, A2, B2
In Practice, I have :
A1, A2, B1, B2
Did I miss something ?
Thanks
Nico
Hello Nico,
So I did some initial testing here, and it looks like it’s happening as expected here. Can you export a set of those jobs and send them in to support? I ask because we would like to try to figure out why the jobs are acting like that for you. Thanks.
Cheers,
Hello,
‘Can you export a set of those jobs and send them in to support?’
Is there a way to do this, from Monitor, select one job and export it in a Deadline format ?
Thanks
Hello Nico,
I apologize, my wording was a bit off. I should have said archive the jobs so we can have you send them over and we can take a look. You can archive a job by right-clicking on it in the job list in the Monitor and selecting Archive Job, then giving it the path you want to save it to. Once you have them archived, can you email it to support@thinkboxsoftware.com. Thanks!
Hello,
I found a useful panel ‘Job Details’ from which it is possible to copy/paste easily.
I have attached these descriptor, I think you should find all information you need.
A1 & A2 goes together, and also B1 & B2.
Thanks
Hello Nico,
So I looked at those files you sent, and here is what I noticed. A1 priority 20. B1 priority 20. A2 Priority 90. That pretty much explains why A2 picked up after A1 was done, because the priority was set such that B1 was not the logical next job.
Sorry, I did a mistake inside text file names, it is important to look at ‘Job Dependencies’:
There are:
- A1 (53980873f7f03e06c495479b), B1 (53980873f7f03e06c495479c dependent on 53980873f7f03e06c495479b)
- A2 (539808a8f7f03e06c495479e), B2 (539808a8f7f03e06c495479f dependent on 539808a8f7f03e06c495479e)
(I put all jobs in a ZIP, because it is not possible to attach more than 3 files)
There are dates inside, moreover I put a portion of repository history., it gives chronology of submissions.
And I give the portion of the slave which did the job (In fact we only have ONE machine set for these proper Pool & Group.
In principle we should have that order :
A1, B1, A2, B2
You may read that we have
A1, B1, B2
I hope you have enough information to investigate
Thanks
Hello Nico,
Just to let you know, I have removed all attachments from this post as we don’t want to se you get in trouble for posting identifiable stuff here. It might be better to email stuff in for future?
As for this issue, I think the best thing for us would be to have you recreate the setup of this on your end with jobs that won’t take very long to render, and we can have a remote session to poke around and see it actually running. Is this possible for early next week? Thanks.
Hello Dwight,
Thanks for all.
I hoped you would find a solution with all my indications.
I will look for a work around.
Thanks.
Hello Nico,
The issue is that we haven’t been able to recreate this issue. Every test done on this issue performed like we expected. This is why we would like to take a look via remote session as this is happening, so we can try to get a better idea of what could be causing it.