RebusFarm upload looking for old CoronaProxies

[ant_march]

I'm perplexed!  I've submitted a ticket to Rebus and waiting to hear back from them, so thought why not come here also in case this has happened to anyone else.  I've done some searching but couldn't find anything similar.

We have about 5/6 animation scenes left to render and are looking to send them to Rebus.  I'm trying to upload a scene for test frames, as you can see from my attached image, it's throwing up errors. 

I've highlighted one such proxy where the error is being thrown up.  Essentially, it's looking for CoronaProxies in old locations.  This project currently has every scene and proxy asset using absolute local paths.  This is my preferred way of working, as it's neat and tidy, and easy to keep track of the various scene assets. I understand that I can workaround this problem, by finding all my scene proxies, in the different Xrefs, and converting them back to meshes.   But obviously I don't want to do this as all the scenes are set and ready to render and the waste of time doing so.  In fact they do render fine, on my pc, my colleagues pc and our office workstations, just not when I try to upload to Rebus.

All my xrefs have been optimised for animation, stripped of unnecessary geometry for the scene camera.  I've made sure there's no missing links or maps.  I use Forest and Railclone.  I was able to determine that old Corona Proxies no longer used inside a Forest object, one that was merged in from another project, were also coming up with the same error.  I converted this Forest object to instances and deleted it, and obviously those errors went away, but I found it strange that these Forest instances were showing up on an object even though they weren't even used.  I guess that's a Forest thing, but these errors are not, they are separate Corona Proxies.

Other than these Corona proxies throwing up errors, I had only 5 maps throw up errors.  This was unusual in that they were where they were supposed to be, asset tracker could see them, but when I found it in my material browser and clicked on it, it disappeared.  It was a max bitmap and on an imported bought asset.  I fixed this by creating a new coronabitmap and relinking to those maps.

So these are the errors that are left, but this is only one camera, I'm sure it will be the same headache through the others.

Anyone know what's going on? 

[ant_march]

Well I finally got an actual description of the problem from Rebus, rather than the standard "just collate all your files into one easy to find directory" answer.

Here's the explanation, maybe it will help someone.  We've decided to use another renderfarm to get this project finished.


"Unfortunately there is one part of 3ds max scene we have not achieved to work out, and it is exactly what you are experiencing. Somewhere in the memory of the 3ds max scene still have the old paths written you are using anymore, and farminizer still recognizes it as a problem. In some cases works not only changing the path in the asset tracker but also in the node object itself (in this case the proxy file). In some cases it only works with delete and reload. The problem about the "memory" of the scene not only apply for this, but also for plugins: e.g. is you create a scene in max19 but later keep working on it on max20, then the farminizer will start mentioning an error about not knowing corona2019.dlr (this does not exist, is an example), but we also have small fixes for this. "

Sorry for the trouble, we are still trying to make it work smoother, so workflow like yours do not face this problems.