[openbeosnetteam] Re: Progress report
- From: Peter Moore <peter@xxxxxxxxx>
- To: openbeosnetteam@xxxxxxxxxxxxx
- Date: Thu, 14 Mar 2002 22:05:13 +1100
>Peter Moore wrote:
>> if you use windows at work then try HTTPort http://www.htthost.com
>/
>> i was able to use it to http tunnel through a proxy server and
>checkout the
>> Apache 2 sources.
>
>Will this HTTPort soft let me commit into OpenBeOS CVS from office
>computer?
>I mean, using SSH and CVS over this HTTP tuneling to
>cvs.sourceforge.net?
not sure about whether it allows ssh as i didnt have the need to.
but from memory some https sites i accessed worked through it, but hotmail
didnt for example.
>
>> you can configure it to allow other computers to go through it as
>well, so
>> your BeOS PC should have no issues if you run it on a windows PC and
>access
>> it via BeOS or Linux or even Solaris.
>
>Next step: install BeOS into my office computer, and try to hide this
>fact
>to the sysadmin... :-)
exactly what i did!! you can configure it to say that you are using Netscape
Navigator or MS IE even though you might be accessing something from Telnet :
-)
unless you have someone who is monitoring the traffic (down to the packet
level) on the local subnet then they wont even know. and i think even if they
were packet sniffing the fact that the HTTPort software conforms to the http
rfcs it passes as normal http traffic.
>
>> will i need a "normal" BeOS 5.03 install to test this ? (ie not BONE)
>
>As we don't base the new network stack code on any previous BeOS net
>stuffs,
>it should works same in vanilla BeOS (R5.0.x) and BONified BeOS
>systems.
>Compilation, however, may have some difference, because of differents
>public include files/path...
>
>The only problem I could see is the select() support offer by the
>kernel.
>Manuel Jesus Petit de Gabriel, former BeOS kernel engineer tell me
>that,
>if he recall correctly, the vfs select implementation in vanilla r5 has
>some memory
>corruptions bugs, which were fixed for BONE. It could have impact the
>stability
>of our select() support for sockets with the vaniall kernel.
>We will see...
>
>Some here code under R5.0.x (David, I guess), others have BONE (you, I
>guess! ;-) ),
>some have both (me). Same things for SMP hardware machines.
ok that's good to know. I'll re-install my R5.x Pro and it will be on a dual
PIII 1GHz PC so i can test the SMP side of things both on "vanilla" and on
BONE.
peter
>
>-Philippe.
>
>--
>Fortune Cookie Says:
>
>Philosophy will clip an angel's wings.
> -- John Keats
Other related posts:
