I defaulted back the review and OIIO values, updated the AYON server and it is now working - I don’t have any explicit solution unfortunately.
I thank you both again for your time and patience!
Hey!
I just came across a similar problem with the Extract Review with .exrs from Houdini and Redshift where I ran into the “Channel doesn’t match layer: R” etc…
The solution that fixed it for me was setting the Handle Missing Frame option in the Extract Review option in core settings to Use only rendered.
Having it set to Use closest exsisting seems to have caused the issue.
I have no idea how that switch would affect this particular error - so you’ve made me very curious. Can you confirm switching it back still reproduces the error? And if so, how do I reproduce this?
Yes, confirmed now that switching it back reproduced the error…
I’m working on a custom product type and custom publishing plugins for houdini, so I’m not sure if it’s coming from that I’ve done something very weird on my end… But setting it to “Use only rendered” fixes it.
I can send you my ayon publish logs + working and not working deadline log if you want?
Yes please, always good to know what’s going on.
Here’s an extract of the logs. Not sure if this is enough info for you. But the left hand side one is when it is set to “Use Closest Existing” and the right hand side is with “Use only rendered”
Looks like it’s doing a completely different FFMPEG call. Perhaps @Petr_Kalis may know more about this part of the codebase.
Only thing that is(should be) different is range of frames to be rendered.
-start_number 0 -framerate 2.0 -to 5.5000000000 -i C:\Users\dev\AppData\Local\Temp\ay_tmp_w6222n9x\testing.%04d.exr
vs
-f concat -safe 0 -i C:\Users\dev\AppData\Local\Temp\explicit_framesgvfcmxn4.txt -r 2.0
Everything else should be same. Not sure if that -r 2.0 creates a difference. (It should not in regard of format, afaik.)
But the issue is that it creates different/wrong .exrs? (In regards of channel.)