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

Bo Berglund bo.berglund at telia.com
Fri Oct 27 23:28:19 BST 2006


On Fri, 27 Oct 2006 12:37:13 +0100, Tony Hoyle
<tony.hoyle at march-hare.com> wrote:

>Those symptoms look like packet loss of large packets - normally caused
>by misconfigured MTU across tunnels and/or broken PMTU discovery.  Other
>possibilities are a broken network card or switch.

Well, since the network "card" is a virtual adapter (VMWare) I can't
see how it can actually be broken...

Regarding the MTU I just went to the server (using remote desktop) and
changed (had to add it) the MTU setting on the NIC to 1300 (which is a
value that makes PPTP VPN connections work such that remote desktop is
OK.
Restarted the server and then tested it with status -v, but no change.
Then I reduced it to 1100 but then I thought I detected a worsening of
the symptoms (no data at all) so I changed it back to 1300.

PMTU is something I don't even have a clue as to what it is....

But now I am fingering on stuff I don't know anything about, really...

>Compression would make this worse since it builds up a chunk of data
>before sending it, rather than sending line by line.

Then it is at least a cause behind it. Strange though.

What can I do next (the Wireshark suggestion is tried but I don't
really understand what to do with it)?


HTH

/Bo
(Bo Berglund, developer in Sweden)


More information about the cvsnt mailing list