FFAStrans forum

The lock file has an automatic timeout...so if it persists for more than 45 minutes we ignore it and continue with monitoring. Starting and stopping should solve the problem, and clearing monitor cache should definately solve the problem. If they don't, please check if the .mxf files are actually ...

@emcodem

i play with the thought about making it default or even dump support for light skin in near future

No you don't!

-steinar

Hi DCCentR,

I'm afraid this is one looks like a dreaded concurrency issue with the way we use file system as db in a farming environment. I would like you to make sure these registry parameters are set on each host and server hold the FFAStrans installation:

Windows Registry Editor Version 5.00 ...

An already running job will change priority at runtime. This also means it will try to change the priority of child jobs, like ffmpeg encoding. However, to avoid stalling the system with high priority on CPU-intensive operations, the maximum possible priority for child processes will be "normal ...

Sweet mother of Jesus...it's a bug! Will be fixed for the next release. Thanks for reporting!

-steinar

Yes, 2016 or better. But it will likely also work on 2012....

-steinar

Hi djv6ru,

Thank you for using FFAStrans and welcome to the forum!

This is a bug :-( But thanks for reporting. To work around the problem (until a new release is out) you could try and put a priority change to the job right after you get the job id. That should work. There might be a small delay ...

Hi DCCentR,

We really appreciate your help with this issue, which is not an easy one to solve. Can I ask for you to also send some logs of jobs that was OK? That is jobs that did NOT have the stuck monitor files.

Thanks!

-steinar

So knk, it's basically the same use case as we have where we use FFAStrans to drive Whisper transcriptions. If you leave your hosts at 24 slots and set the GPU intensive node to utilize 13 slots, then you will still have 11 free slots for other operations but not enough room for two 13 slot nodes on ...

Hi knk,

You cannot limit a workflow like this. The closest (without adding another host for this one workflow) might be to use the feature where you tell a node (or even all nodes in your workflow) to occupy more slots in order to limit the total executed nodes per host. But it's limited occupy 16 ...