[Home] [Groups] - Message: [Prev in Group] [Next in Group]
7236: [MUD-Dev] Re: [IDEAS] Starting from scratch
[Full Header] [Plain Text]
From: J C Lawrence <claw@under.engr.sgi.com>
Newsgroups: nu.kanga.list.mud-dev
Date: Tue, 15 Sep 1998 15:03:16 -0700
References: [1]
Organization: Kanga.Nu
On Thu, 06 Aug 1998 20:34:06 -0000
Ross Nicoll<rnicoll@calmar-mud.com> wrote:
> s001gmu wrote in a Mail about "[MUD-Dev] Re: [IDEAS] Starting from
> scratch":
>> I believe they also tied that in with the crash-recovery. In
>> case of a seg fault they catch the signal and do a gracefull
>> soft-boot. I dunno if they put a counter/timer in it, to catch
>> things like loading a certain p-file causing the seg-fault (which
>> would send the server into a lovely infinite reboot cycle).
> That wouldn't really suit my MUD, if anything is wrong, it almost
> definately means some of the data is corrupted, and so a database
> save would just crash it again. Anyway, we tend to be up for about
> 2 weeks at a time, which isn't bad...
This is why I roll the DB back to the last known logically correct
and consistent point upon restart. Of course there is a possibility
that the seeds of the prior crash already exist there, but I've
found that to be the rare case.
--
J C Lawrence Internet: claw@null.net
(Contractor) Internet: coder@ibm.net
---------(*) Internet: claw@under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...