[Freeciv-Dev] Re: civclient loops infinitely with corrupted network data
[Top] [All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: |
freeciv-dev@xxxxxxxxxxx |
Subject: |
[Freeciv-Dev] Re: civclient loops infinitely with corrupted network data (PR#1247) |
From: |
Christian Knoke <ChrisK@xxxxxxxx> |
Date: |
Wed, 30 Jan 2002 11:48:29 +0100 |
On Wed, Jan 30, 2002 at 09:57:30AM +0100, Raimar Falke wrote:
> On Tue, Jan 29, 2002 at 07:32:04PM -0800, Vasco Alexandre Da Silva Costa
> wrote:
> > On Tue, 29 Jan 2002, Raimar Falke wrote:
> >
> > > On Tue, Jan 29, 2002 at 07:03:04AM -0800, Christian Knoke wrote:
> > > > Recent CVS + CMA 3.4
> > > >
> > > > Civclient loops, spoiling
> > > >
> > > > 1: Received unknown packet (type 0) from server!
> > > > 1: received short packet (type 0, len 0) from server
> > > >
> > > > again and again, seemingly because it receives data
> > > > it doesn't expect, at a time when the other side
> > > > already has aborted. I report it, though it won't
> > > > happen with an intact civserver:
> > >
> > > It is known that the network part isn't robust. It is low priority
> > > since nobody complained about it and nobody is interrested in it.
> >
> > Erm, i don't consider this low priority... Unless CMA is merely optional
> > at runtime. (i believe it is?)
>
> Christian has a program which acts changes/corrupts the data stream
> between client and server. Lets see what can happen with the current
> (non-CMA) code:
ACK. This has nothing to do with CMA. And of course I don't expect
civclient to work properly when the data stream is corrupted.
My thoughts when I reported this were:
civclient shouldn't core dump (PR1225)
civclient shouldn't hang in an infinite loop without
blocking
Christian
--
http://www.enter.de/~c.knoke/ * * * * * * * * * * * * * * *
* * * * * * * * * Ceterum censeo Microsoft esse dividendum.
|
|