alex

Hi!

I have the same problem as described above. I use a mac os x server as file server with the number of client connections set to unlimited.

The mayabatch on the render machines are pointing to a remote maya.env configuration file located on the file server. The mental ray shaders are also located on the file server.

The muster user has read and write access to all the files on the server. I also disabled all the error checking (and actions) to prevent clients from getting stuck on one chunk. Instead we go through the client logs manually if something goes wrong.

Still, it randomly throws 211’s from about 50% of the frames. Some of the lost scenes are able to rerender but other are impossible. And they’re still inherited from the same scene file.

Since all the clients are xp machines I tried to mount the network share with a persistent flag to prevent disconnection when the machine enters sleep mode, although it seemes to reduce the 211 error I still loose a lot of frames. Another mysterious phenomena is that a lot of logfiles from the clients are not completed. It looks like the client tend to just skipp everything and walk to next shot without logging anything.

Does that mean that muster throws a 211 not only due to permission issues but with corrupt scene files as well?
If that is the case maybe the error codes should be separated if possible.

If let say the mayabatch run into issues with a scene file. How does muster deal with it? Will an unknown mayabatch error simply return in a 211?

I guess my case could be a combination of client to server connection issues and scene file problems. But some of the muster errors doesnt make any sense which makes errors very hard to track and debug.

…

that is all.

any suggestions?