[cvsnt] Strange behaviour of cvs status on a VMWare server...

JP Vossen jp at jpsdomain.org
Fri Oct 27 07:42:48 BST 2006


Bo Berglund wrote:
> On Thu, 26 Oct 2006 19:53:17 -0300, Gerhard Fiedler
>> Another thing may be to compare the network traffic of the command that
>> works with the command that hangs. 
>>
> How does one compare network traffic? I have no idea about that...
> In fact I have often wanted to be able to "see" what my own PC at home
> is doing towards the network when it seems to be slowing down for
> unknown reasons...

http://wireshark.org/

Reasonably easy to use and runs on everything.  Do a capture while 
running the command that works, stop it and save it, then do a capture 
while running the command that fails.  On Windows you'll need to install 
WinPCap (which is in the installer) and be admin to do it.  On other 
systems you may have to install libpcap and you'll have to be root.

If you are not on a switched network or if your computer is doing a lot 
of network related things, you may need to figure out a capture filter 
using destination IP and/or ports or you may get overwhelmed.  In my 
opinion capture filters are harder than they should be, but they aren't 
that hard to figure out.

If you are on a switched network you have to run Wireshark on the same 
machine from which you are doing the CVS commands (else you won't even 
see the traffic), unless you have a switch with a span or mirror port 
you can use.  But it sounds like you can easily do everything from the 
same machine and if so, you should as it'll be a lot easier.

Once you have captures, find the packets that begin each connection to 
the server, then use the Analyze, Follow TCP Stream menu to open a 
window containing sort-of a transcript of the session.  You may need to 
open and refer to the capture of the command that works to see what it 
should look like, then find the differences in the broken one.

You can save out parts of the capture in a plain text form (the default 
is a binary form), and you may attempt to diff it.  I suspect so much 
will differ, like sequence numbers, time stamps, etc., that the real 
differences will be drowned out in the noise.

HTH,
JP
----------------------------|:::======|-------------------------------
JP Vossen, CISSP            |:::======|        jp{at}jpsdomain{dot}org
My Account, My Opinions     |=========|      http://www.jpsdomain.org/
----------------------------|=========|-------------------------------
Microsoft has single-handedly nullified Moore's Law.
Innate design flaws of Windows make a personal firewall, anti-virus
and anti-malware software mandatory. The resulting software arms race
has effectively flattened Moore's Law on hardware running Windows.


More information about the cvsnt mailing list