Wrongly reported error: "Port 51347 seems to be used by another program"

[tasptz]

When opening the udp server an error is reported Port 51347 seems to be used by another program however the exception actually says OSError: [WinError 10049] The requested address is not valid in its context.

The address in my case is ('ff05::1', 51347). There is some code addressing this in __init__.py:804 but it is commented out.

if addrinfo.broadcast == '<broadcast>':  # or addrinfo.broadcast == 'ff05::1'
  bind_addr = ''
else:
  bind_addr = addrinfo.broadcast

[pgiri]

Does it work if you uncomment or part (so bind_addr is `` in your case)?

Exception message can be fixed by checking errno (will commit this fix in the next few days).

[tasptz]

It works but then I ran into other errors related to this.

But if I specify an ip4 address it works

    # create cluster
    cluster = dispy.JobCluster(compute, ip_addr='127.0.0.1', depends=[C])

[pgiri]

Instead of ff05::1, can you try configured interface address (starts with fd)?

[vesnikos]

It seems that I am affected by the same error. I had the 'Port 51347 seems to be used by another program' and after I installed netifaces the error changed to 'OSError: [WinError 10049] The requested address is not valid in its context'

it could be related with https://stackoverflow.com/questions/23857942/winerror-10049-the-requested-address-is-not-valid-in-its-context as I was trying to bind a remote server as node in an external network.

windows 10 + python 3.6 and dispy in a ve

[pgiri]

If your problem is also because the address used is ff05::1, then, as mentioned above, can you explicitly use configured interface address (e.g., for eth0) with both dispynode (with -i option) and JobCluster in client (with 'ip_addr=' parameter)? ff05::1 is a multicast address and it can't be used for binding. May be there is a way to deal with this issue, but I don't know enough about IPv6 to figure out.