Re: [Dpp-devel] Review Request: Socks5Proxy initialisation improved for multiple IP addresses

[bjornson <bjo...@in...>]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://saros-build.imp.fu-berlin.de/reviews/r/28/#review95
-----------------------------------------------------------

Ship it!


- bjornson


On April 14, 2011, 3:04 p.m., gustavs wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://saros-build.imp.fu-berlin.de/reviews/r/28/
> -----------------------------------------------------------
> 
> (Updated April 14, 2011, 3:04 p.m.)
> 
> 
> Review request for Saros.
> 
> 
> Summary
> -------
> 
> Currently, the local Socks5Proxy is set up with 1 local IP by "InetAddress.getLocalHost().getHostAddress()".
> This call only return one local IP and sometimes (java version dependend) even the loopback device IP which is of no use for external clients trying to contact me.
> 
> On several tests in our FU LAN, a Socks5 data connection was only seldom possible and then fell back to slow IBB.
> 
> 
> The patch retrieves all (non loopback) IP addresses of all network devices. So also with multiple network adapters, Saros is accessable.
> Same tests as above now make Saros instances Socks5 accessable in many cases.
> 
> E.g., Debian system with two network adapters:
> Original IP list:[127.0.1.1]
> Now IP list:    :[192.168.1.1, 192.168.110.132, fe80:0:0:0:20c:29ff:fe89:2682%3, fe80:0:0:0:20c:29ff:fe89:2678%2]
> 
> 
> Diffs
> -----
> 
>   /trunk/dpp/de.fu_berlin.inf.dpp/src/de/fu_berlin/inf/dpp/Saros.java 3255 
> 
> Diff: http://saros-build.imp.fu-berlin.de/reviews/r/28/diff
> 
> 
> Testing
> -------
> 
> manual testing on several platforms, different network szenarios
> 
> 
> Thanks,
> 
> gustavs
> 
>