- Posts: 5
- Joined: Fri Dec 16, 2016 12:04 pm
Fri Jan 06, 2017 3:56 pm
Hi,
We have narrowed down some bad Fusion crashing to broken or incomplete openEXRs.
The issue is that we are currently trying to run through a lot of comps. So we are looking at around 10 days for 2000 comps / per node. So we need the jobs to continue running to use all the computing time. But currently it will crash the entire application and queue if there is a single bad frame. We have added some better verification to our pipeline but it would be very helpful to get a better handling from fusion. I have collected some of these bad frames. Will it be possible to email these to a developer for testing perhaps?
Thank you,
Marco
We have narrowed down some bad Fusion crashing to broken or incomplete openEXRs.
The issue is that we are currently trying to run through a lot of comps. So we are looking at around 10 days for 2000 comps / per node. So we need the jobs to continue running to use all the computing time. But currently it will crash the entire application and queue if there is a single bad frame. We have added some better verification to our pipeline but it would be very helpful to get a better handling from fusion. I have collected some of these bad frames. Will it be possible to email these to a developer for testing perhaps?
Thank you,
Marco