I’ve gotten Swarm to run across my network and it works great minus one thing.
Let say I’ve got 3 machines, A, B & C - and N (20) other nodes. We’ll say C is the coordinator.
A,B & C all have ‘AvoidLocalExecution’ set to true.
N & C are in group ‘RenderGroup’
B is in it’s own group ‘RenderGroupB’
A is in it’s own group ‘RenderGroupA’
If I run a build from machine C (the coordinator) everything works perfect - Only N handles the rendering (C respects ‘AvoidLocalExecution’)
If I run a build from Machine A then all of N are used as render nodes (not B or C) but A itself is also used to render even though it’s set to ‘AvoidLocalExecution’
If I run a build from Machine B then all of N are used as render nodes (not A or C) but B itself is also used to render even though it’s set to ‘AvoidLocalExecution’
Is there a way that I can ensure that the non-coordinator machines that are set to ‘AvoidLocalExecution’ actually render using only network nodes?