AY-6230 - Submitting pointcache product type to farm is always a single file

MustafaJafar commented 3 months ago

Is there an existing issue for this?

[X] I have searched the existing issues

Current Behavior:

It happens with Alembics and Bgeos

So, I've added $F4 to my alembic rop output file to have a separate sequence of alembic files. It works as expected with local publishing However when submitting to farm I always get one file Although, I've set the frame per task to 1

With Bgeo, it fails completely :(

2024-07-31 18:21:16:  0: STDOUT: Missing expected files: ['pointcacheBgeoCache.0999.bgeo', 'pointcacheBgeoCache.1000.bgeo', 'pointcacheBgeoCache.1002.bgeo', 'pointcacheBgeoCache.1003.bgeo']
2024-07-31 18:21:16:  0: STDOUT: Expected files: ['pointcacheBgeoCache.0999.bgeo', 'pointcacheBgeoCache.1000.bgeo', 'pointcacheBgeoCache.1001.bgeo', 'pointcacheBgeoCache.1002.bgeo', 'pointcacheBgeoCache.1003.bgeo']

Expected Behavior:

No response

Version

1.0.0

What platform you are running on?

Windows

Steps To Reproduce:

publish abc (with .$F4 token in output path) or Bgeo to farm

Are there any labels you wish to add?

[X] I have added the relevant labels to the bug report.

Relevant log output:

No response

Additional context:

No response

[cuID:AY-6230]

SzabadosA commented 3 months ago

looking at the manifest, it seems like the frame always gets set to 0 and the output dir/file does not get resolved:

SzabadosA commented 3 months ago

I just did some test.

Bgeo does work without issues
ABC breaks in deadline, but the manual fix is manually setting the frame range properly from 0 to the correct range = > will compare the code differences between bgeo and abc and come back to you

SzabadosA commented 3 months ago

https://github.com/ynput/ayon-deadline/blob/67fce57874ba293a754c1adae93202c1d071a20c/client/ayon_deadline/plugins/publish/houdini/submit_houdini_cache_deadline.py#L80

This line of code seems to be the problem, if i disable the condition, the caches go through as expected. Do we know why that is in there? Maybe @iLLiCiTiT ?

iLLiCiTiT commented 3 months ago

I'm sorry, but I'm not familiar with houdini, would be question for @BigRoy or @antirotor .

SzabadosA commented 3 months ago

Ah ok, I just tagged you because I saw your name in the code blame for this code

BigRoy commented 2 months ago

https://github.com/ynput/ayon-deadline/blob/67fce57874ba293a754c1adae93202c1d071a20c/client/ayon_deadline/plugins/publish/houdini/submit_houdini_cache_deadline.py#L80

This line of code seems to be the problem, if i disable the condition, the caches go through as expected. Do we know why that is in there?

No idea - @MustafaJafar any idea why that might be there? If not, let's remove it - and then test the different cases.

MustafaJafar commented 2 weeks ago

I was checking https://github.com/ynput/ayon-deadline/pull/49 and this issue came to my mind.

No idea - @MustafaJafar any idea why that might be there? If not, let's remove it - and then test the different cases.

I think this most probably assumes that alembic is always generating a single file. however, alembic nodes can generate file sequence if $F token was used.

So, we need a more accurate condition.

BigRoy commented 1 week ago

Something similarly broken came up on community forum here: https://community.ynput.io/t/error-houdini-deadline-cache-submission/1987/4?u=bigroy

Not sure if 1:1 the same thing.

MustafaJafar commented 4 days ago

Something similarly broken came up on community forum here: https://community.ynput.io/t/error-houdini-deadline-cache-submission/1987/4?u=bigroy

Not sure if 1:1 the same thing.

The log in that post comes from an older Houdini addon, more info check https://github.com/ynput/ayon-houdini/commit/37a4e91b4a47eb02eb66dbcad3633f0a4d3b0fda

ynput / ayon-deadline