Announcement

Collapse
No announcement yet.

Distributed Rendering "Ping not received..."

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Distributed Rendering "Ping not received..."

    I just started using DR when rendering, and first time round on a rather complex model it came up with this:

    Ping not received from host 0.0.0.0 for 50188 milliseconds, dropping connection.
    Render host unknown (192.168.1.XX) is not responding

    *XX substituted for actual number*

    Second time round I tried with a less complex model and it worked like a treat.

    Third time I tried the complex model again (all same settings as second time round), and it once again gave me the 'Ping not received...' etc.

    The this particular slave computer is an older computer, and could just be taking longer than ~50 seconds to receive and ping back.

    Is it possible to increase the time vray allows for the ping?

    if not

    What could be causing this problem?

    Any other methods to solve it?

  • #2
    Re: Distributed Rendering "Ping not received..."

    Same problem - any solution?

    Comment


    • #3
      Re: Distributed Rendering "Ping not received..."

      Don't know for sure if this is your issue, but we got the same error once in a while on a couple of our nodes and it was driving us batty because it was hit or miss.

      The two things we did that stopped those "not responding" errors was this:

      1. Make sure all materials and bump/disp map images are accessible across your network to all nodes. (find a common folder for everything/ The complex models have so many maps it's easy to have one that is unreachable.) I think it is suppose to push them from the main to the nodes, but if one map is overly large, this may slow that down.

      *2. Make sure you turn off the sleep functions of your nodes. We noticed a couple of the render nodes going into sleep mode and they wouldn't wake up when Vray requested them to, at least not fast enough...because they had to wake up first-if they did at all. Usually after this happened we had to restart the node, try again, and it would work fine.

      Like I said, not sure if this is your issue, but it helped us stop that from happening.

      So tune the render nodes to run full bore, turning off any sleep/hibernate settings.

      Comment

      Working...
      X