[Freeciv-Dev] Re: core file on civserver, http://civserver.freeciv.org/games/162970

[Raimar Falke <i-freeciv-lists@xxxxxxxxxxxxx>]

On Sun, Nov 30, 2003 at 01:37:21PM +0000, Per I. Mathisen wrote:
> On Sun, 30 Nov 2003, Raimar Falke wrote:
> > > The backtraces generated from such asserts are usually useless
> >
> > I agree.
> >
> > > , while the errors are often recoverable if they didn't assert. Much
> > > better to output some useful error logging, IMHO.
> >
> > We had this discussion (die or recover) in the past. I'm against doing
> > this in CVS HEAD
> 
> When we had this discussion in the past - we agreed that the importance of
> catching and fixing bugs was more important than recovery. However, in
> this case, it doesn't let us fix any bugs, so what is the point in killing
> the server?
> 
> Losing a game due to a server that dies should only be acceptable if the
> player who is playing this game can go to us with the core file and get
> the bug fixed. Then he/she may feel that it has gotten better and the loss
> of the game was not so bad after all. However, when our response is only a
> shrug - then that is doubly aggrevating.

In the case of pubserver we have to do of course investigate these
problems. There are saves for every turn for the pubserer games. If
these doesn't provide enough information to track this done we should
change to code so that the necessary information are logged or saved.

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "The very concept of PNP is a lovely dream that simply does not translate to
  reality. The confusion of manually doing stuff is nothing compared to the
  confusion of computers trying to do stuff and getting it wrong, which they
  gleefully do with great enthusiasm." 
    -- Jinx Tigr in the SDM