My understanding is that this is not an error, it's an interruption in the communication with a client. This should be harmless if it doesn't happen at a massive rate. You probably have many more of these in your logs.

On Jun 14, 2017 11:04 PM, "evan clark" <eclark28@fau.edu> wrote:
That may work, but now we are getting this error.

Exception happened during processing of request from ('127.0.0.1', 43030)
Traceback (most recent call last):
  File "/cm/shared/apps/galaxy/prod/galaxy/.venv/lib/python2.7/site-packages/paste/httpserver.py", line 1085, in process_request_in_thread
    self.finish_request(request, client_address)
  File "/cm/shared/apps/galaxy/requirements/python/lib/python2.7/SocketServer.py", line 331, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/cm/shared/apps/galaxy/requirements/python/lib/python2.7/SocketServer.py", line 654, in __init__
    self.finish()
  File "/cm/shared/apps/galaxy/requirements/python/lib/python2.7/SocketServer.py", line 713, in finish
    self.wfile.close()
  File "/cm/shared/apps/galaxy/requirements/python/lib/python2.7/socket.py", line 283, in close
    self.flush()
  File "/cm/shared/apps/galaxy/requirements/python/lib/python2.7/socket.py", line 307, in flush
    self._sock.sendall(view[write_offset:write_offset+buffer_size])
error: [Errno 32] Broken pipe

June 14, 2017 4:31 PM
Hi Evan,

If this is something that you rarely see and your filesystem is on a network drive,
you could try setting `retry_job_output_collection` to 10 in the galaxy.ini file.
This will let galaxy try to collection the output files 10 times,
which should help if there is some delay for the filesystem to be updated.

Best,
Marius