Changing Network Group clears Jobs list

Render Manager Issues
#1

This is on a single machine acting as the Workstation and Admin.

If you submit a job(s) to Pulze, then go to Settings > Network Group and create a new group, the job queue will be cleared and become unrecoverable. The scene files remain on disk in the Repository but I can’t find any way to restore the queue itself. Switching the Network Group back, ie to “default”, does not restore the queue. If you never adjust the Network Group then the job queue appears to be stable and is retained across multiple sessions.

If I am doing something wrong here please let me know. I think that there should always be a way to restore a job queue from files, or if that is not possible a warning that it is about to be cleared. The documentation around Network Groups is also very short and could explain their purpose better.

-Josh

#2

Hi Josh,

I’m just curious if there is any specific reason that you change the Working Group on one PC or you just test it randomly?

Because we did not mention in the documentation that it is not recommended to switch working groups with only on PC. Sorry for the inconvenience.

On the other hand we have been working hard to make the Network Groups feature more user friendly and robust in the upcoming release.

Cheers,

Attila

#3

Attila,

I was just testing features, kicking the proverbial tires. Could you explain what the Network Groups are used for? I imagine it’s to allow a user to connect to different render queue managers but am not sure.

-Josh

#4

Hi Josh,

First of all there is no render queue manager in case of Pulze Render Manager. It is an equal network instead of a pyramid structure. So you just have to have at least one Workstation alive and it serves as a manager. It makes the communication quicker and the system is more reliable.
The Network Groups are for dividing your farm into isolated groups. In this way different groups of people and groups of nodes can be created and these will be invisible to each other.
In your case the problem was that only one PC jumped from one group to another and there was no PC in the leftower group to remember the jobs. Pulze Render Manager was not prepared for this test because there is no point to do this alone.

Please let me know if you have further questions, I’d be happy to help.

Attila

#5

Ah, ok I understand. I think that there should be a warning that the current job list will be lost if the only PC in a group is trying to switch Groups, since this would be possible even if the system is being used correctly.

Thank you for explaining things, Attila!
-Josh

#6

You are right, we put this on the list.
Thanks Josh for your input!

Regards,

Attila

#7

You’re welcome, and thank you for listening and explaining!