DR slaves very slow to be ready for next render

V-Ray V-Ray for Maya Issues
1

Hi,
I started testing out DR this morning, and it worked fine, except that after the image renders, the slaves do not seem to be aware that the render is done until several minutes later.
[2010/Jun/15|10:18:21] Unshaded rays: 0
[2010/Jun/15|10:22:23] Closing DR session from 192.168.70.64
Clearing command queue.

Any ideas what I may be doing wrong, or if there is something I can send to the slaves to tell them the image is rendered?

2

We experience the same problem right now with the latest builts from today. It takes up to four minutes until the machines are free again. Any ideas on that?

3

I’m using a nightly build from June 13th I believe. It’s pretty consistent with 3 to 4 minute delay between the image being completed, and the slaves being freed up

4

Anything new on this? Our renderclients are getting bored :slight_smile:

(BTW, are the nightly builts officially supported?)

5

I cannot check this right now, but it came to my mind that the writing of the file to disc, which takes pretty long over a standard-network with float images, could be part of the problem.

6

that does not seem to be the issue for me. The files are complete and accessible long before the render slaves acknowledge that the image is done

7

You can try a newer build and see if the issue is still there. If you look at the output from the render servers, what are they doing at that time?

Best regards,
Vlado

8

Hi Vlado,
I installed the June 22 Windows X64 build, and tested again. Still the same results. Here’s the relevant bits from the output:
Render server:
[2010/Jun/22|09:50:56] V-Ray: Render complete
[2010/Jun/22|09:50:56] V-Ray: ========================
[2010/Jun/22|09:50:56] V-Ray: Clearing exporter memory…
[2010/Jun/22|09:50:56] V-Ray: Total time clearing exporter memory 0h 0m 0.0s (0.0 s)
[2010/Jun/22|09:50:56] V-Ray debug: vrend command executed
[2010/Jun/22|09:50:56] V-Ray debug: vrend: -q

Render slave:
[2010/Jun/22|09:51:33] warning: DR request received from 192.168.70.64, but currently busy with 0.0.0.0 (Tried to kick off another DR to see what would happen)

[2010/Jun/22|09:54:10] Closing DR session from 192.168.70.64. clearing command queue