joinmarketd.py running on 0.0.0.0

[0xB44EFD8751077F97]

I know this was supposed to be fixed, but it's still happening for me on master. I am running a fairly unique setup with daemon and wallet running in separate VMs. I'm happy to provide any further information that is helpful.

user@host:~/joinmarket-clientserver$ git show HEAD
commit 5e05c62be39c5923abe831176df3a40dc1ad5907
Merge: 52264a7 bf1a816
Author: AdamISZ <ekaggata@gmail.com>
Date:   Fri Nov 3 22:19:24 2017 +0100

    Merge #103: add --no-gpg-validation flag, default it in travis
    
    bf1a816 add --no-gpg-validation flag, default it in travis (fivepiece)

user@host:~/joinmarket-clientserver$ sudo netstat -antlop | grep 27183
tcp        0      0 0.0.0.0:27183           0.0.0.0:*               LISTEN      10587/python     off (0.00/0/0)
user@host:~/joinmarket-clientserver$ ps aux | grep 10587 | head -1
user     10587  0.0  3.7  97668 44760 pts/0    S+   04:07   0:00 python joinmarketd.py

[AdamISZ]

Thanks for reporting/catching this. I should have of course noted it at the time of the previous fix, but for context:

joinmarketd.py is kind of experimental (as warned here until such time as someone works on it a bit more. I mean this specifically in the sense that, the whole point of it is to do something like you mention (have the daemon on a separate host to the client), and/or to be able to use one daemon with multiple clients. The first of those two requires security consideration (so use ssl at least; but the ssl support is rudimentary at the moment, and not tested much; about all I can say is I am able to get it running with a self-signed cert, and I've heard at least one other person doing it successfully. See also #3 ), the second of those two however has not really been addressed at all yet (see #44 for related comments, but the basic point is it hasn't been done/tested).

Having said all that I fully understand why people want to use it in this mode already. I just hope that more eyes will look at this and make the (hopefully not too substantial) improvements required to make it robust.

So thanks to @undeath for providing the patch (I've manually tested it in separated daemon mode, so ACKed from me).