Recently update to .293 and now I can't get past this error:

…………………….
return self.parse_response(response)
File “/opt/hfs17.5.293/python/lib/python2.7/xmlrpclib.py”, line 1493, in parse_response
return u.close()
File “/opt/hfs17.5.293/python/lib/python2.7/xmlrpclib.py”, line 800, in close
raise Fault(**self._stack)
Fault: &lt;Fault 1: “<type ‘exceptions.TypeError’&gt;:check_ready_batch() takes exactly 4 arguments (3 given)”>

terminate called after throwing an instance of ‘std::runtime_error’
what(): mutex: pthread_mutex_lock failed: Invalid argument

Seems to happen across the board. Previously was working.

I trashed the server and client folders and did fresh installs. Only touched the server address and mount directory in the .ini files. Clean scene file, fresh top rops, and still persists.

Server and clients running Linux (ubuntu 18.04 server, centos 7 and fedora 30). No firewalls and they can all see each other by hostname. HQueue runs sims and renders fine from /out via the old way.

Anyone else having trouble?

Seems to happen across the board. Previously was working.

I trashed the server and client folders and did fresh installs. Only touched the server address and mount directory in the .ini files. Clean scene file, fresh top rops, and still persists.

Server and clients running Linux (ubuntu 18.04 server, centos 7 and fedora 30). No firewalls and they can all see each other by hostname. HQueue runs sims and renders fine from /out via the old way.

Anyone else having trouble?

Sorry about that - it was a temporary regression between ~.291 to .293. I recommend trying a more recent build.

Well that's good it's not me

I'll give the daily a go. Thanks!

Update: Updated everything to 17.5.300 and all is well again. Thanks for the info!

Same problem here.

In every Production Build since 17.5.299 (17.5.299 / 17.5.258 / 17.5.293)
PDG and HQueue were broken.

See here as well:
https://www.sidefx.com/forum/topic/66698/ [www.sidefx.com]

17.5.320 fixes it for me but I can't use it because then Redshift won't work anymore because apparently there have been updates to the HDK.
Don't know if 17.5.300 would still work with redshift-2.6.44.
But anyways I can't download it anymore because there is always only the latest 2 daily builds available.


This is really causing me issues recently.
It would be great if this stuff could be tested better before releasing the Production Builds.

Any idea how I could work around this?

Thanks

I can make 17.5.294 available to you if you wish - it has the fix for the hqueue bug and doesn't have any HDK changes.

That would help me a lot!
Thanks

PS. would also be awesome though if this would be fixed in the next production build.
As our IT dep. only deploys prod. builds on the farm, and right now I use the messy workaround to use a Houdini Version that doesn't have the bug as the Hqueue Server and the Production Build on the clients.
That works, but is quite tedious and is also causing some other issues.

I am also running into the same problem. PDG with distributed hqueue sims seem broken in 17.5.293 but that's the latest version for which Redshift is currently available.

It looks like 294 is not available as a build at the moment.

Niels Provos
I am also running into the same problem. PDG with distributed hqueue sims seem broken in 17.5.293 but that's the latest version for which Redshift is currently available.

It looks like 294 is not available as a build at the moment.

You can get .294 here: ftp://ftp.sidefx.com/public/Houdini17.5/Build.294/