Page MenuHome

[netrender] Broken pipe when sending result frame
Closed, ResolvedPublic


System Information

ArchLinux (kernel 4.15.2-2-ARCH)

Blender Version
Broken: version 2.79 Hash: 8ef39d5c882

Short description of error

Broken pipe when a slave running on the same host of the master node try to send his result:

Here are the logs:

Saved: '/tmp/slave_eec2333792ca4a514a1768c665e9d353/job_12/000106.exr'                                                                                     
 Time: 11:02.61 (Saving: 00:00.97)                                                                                                                         
Blender quit                                                                                                                                               
status 0                                                                                                                                                   
Traceback (most recent call last):                                                                                                                         
  File "/usr/share/blender/2.79/scripts/addons/netrender/", line 367, in render                                                                   
    raise e                                                                                                                                                
  File "/usr/share/blender/2.79/scripts/addons/netrender/", line 360, in render                                                                   
  File "/usr/share/blender/2.79/scripts/addons/netrender/", line 387, in render_slave                                                             
    slave.render_slave(self, scene.network_render, scene.render.threads)                                                                                   
  File "/usr/share/blender/2.79/scripts/addons/netrender/", line 390, in render_slave                                                              
    conn.request("PUT", "/render", f, headers=headers)                                                                                                     
  File "/usr/lib/python3.6/http/", line 1239, in request                                                                                          
    self._send_request(method, url, body, headers, encode_chunked)                                                                                         
  File "/usr/lib/python3.6/http/", line 1285, in _send_request                                                                                    
    self.endheaders(body, encode_chunked=encode_chunked)                                                                                                   
  File "/usr/lib/python3.6/http/", line 1234, in endheaders                                                                                       
    self._send_output(message_body, encode_chunked=encode_chunked)                                                                                         
  File "/usr/lib/python3.6/http/", line 1065, in _send_output                                                                                     
  File "/usr/lib/python3.6/http/", line 986, in send                                                                                              
BrokenPipeError: [Errno 32] Broken pipe                                                                                                                    
location: <unknown location>:-1                                                                                                                            
location: <unknown location>:-1                                                                                                                            
Error: [Errno 32] Broken pipe

It only seems to happen when the the master and the slave are on the same host. I have two other slaves running on my network a window (version 2.79 8928d99270f) and a Mac (version 2.79 8928d99) and they does not seems to be affected by this issue.

Exact steps for others to reproduce the error

Just send a job on the network and let the slave that is running on the same host crash