DR: Server not found in waiting list, ignoring

[olly]

We are using max design 2014 and vray 3.10.01.

I have 3 out of 20 nodes that won't render in a DR job since we upgraded vray.

Checking the vraylog on the host pc I see the following entries for each of the 3 pcs:

TM_SERVER_STARTED received from 192.168.1.3:20204
Server not found in waiting list, ignoring

Yet the pc's are ticked in the dr list for the machine that sent the backburner job and the host pc that is rendering it.

Any ideas?

Cheers,
Olly

[tashko.zashev]

Hi Olly,

Please check the Max servers parameter in the DR setting window. It should be set to "0".
Could you also post or send us the full V-Ray log from the Main machine and one of the problematic render nodes.
Does the issue persist if Backburner is not involved in the rendering process?

[olly]

The nodes took part in the next job as expected - so the vraylog for the problem job was overwritten.

Not sure if helps, but the affected nodes were turned on after the frame had already begun. Prior to the upgrade this was not a problem though.

I will keep an eye on things and send logs if it happens again.

[tashko.zashev]

Thanks for the quick update.

I will test this here.

Update: Everything works as expected in my test, even if I start the V-Ray spawner (manually and as a service) in the middle of the BB job.
If it happens again, please send us the log files and step by step instructions for recreating the setup.

[olly]

Update: Everything works as expected in my test, even if I start the V-Ray spawner (manually and as a service) in the middle of the BB job.

Is that with the same environment of max design 2014 SP5 and v-ray 3.10.01?

This seems to happen every time a render node is powered up in the middle of a job. I have a log for the host machine I will email it to support quoting this thread.

[tashko.zashev]

Is that with the same environment of max design 2014 SP5 and v-ray 3.10.01?

This seems to happen every time a render node is powered up in the middle of a job. I have a log for the host machine I will email it to support quoting this thread.

Thank you Olly.
We will also need the log from the problematic render nodes. I will send you a mail with some additional details.

Meanwhile I will run another tests to see if this issue can be reproduced here.

[olly]

Before sending us the existing log file please do the following:
Open "Distributed rendering" settings in your scene, and make sure that the option "Restart servers on render end" is not checked.

Hi Tashko,

We actually have that setting unchecked, yet the logs are overwritten with each new render.

I will try and grab one from a node next time I see the behaviour.

Cheers,
Olly

[tashko.zashev]

Hi Tashko,

We actually have that setting unchecked, yet the logs are overwritten with each new render.

I will try and grab one from a node next time I see the behaviour.

Cheers,
Olly

Thank you, it will be very helpful to see this log, because usually there are the errors or warnings, which are not available in the host machine log.
I will post here if we manage to reproduce the issue.

[tashko.zashev]

Managed to reproduce the issue here. It seems that the DR broadcasting is broken and I will send a report to our developers about it.
Could you please confirm that the issue in your environment occurs only when the Spawners has been started after the renderer?

[olly]

Could you please confirm that the issue in your environment occurs only when the Spawners has been started after the renderer?

That is correct.

I look forward to a fix!

Thanks for looking in to it.

[tashko.zashev]

That is correct.

I look forward to a fix!

Thanks for looking in to it.

Thank you. I will post here when we have any news.