Re: your mail

• Date: Wed, 26 Feb 1997 15:49:24 PST
• From:
  Chris Williams <psion@mv.mv.com>
• CC: moo-cows@parc.xerox.com
• Content-Transfer-Encoding: 7bit
• Content-Type: text/plain; charset=us-ascii
• Organization: Unorganized
• References: <Pine.LNX.3.95.970226155744.1163A-100000@redbeard>

> My question is, will I be better off scrapping the work I've done so far
> and loading up the new DB, or are there any big changes?  Also, I'm
I was in a similar situation, I've been working on a new MOO since mid-Jan but 
wanted to upgrade to the new Core. Although it took a bit of work, I managed 
to basicly port everything from my Dec95-based DB into a fresh new Core. 
Although there is 1 more object in the new core, if you recycle Mr Spell 
(which is tricky in itself, but doable), and then use #32 for whatever your 
#94 in the old DB was (for me it was Guest1), then all the objects after that 
case have exactly the same numbers, and you just @dump them over (modify @dump 
to @chown and @move them, and you can almost exactly duplicate your old DB). 
If you've mucked around with Core code a lot, that needs re-doing, but most of 
numerous changes to the Core are relatively minor, so it's not too big a 
problem. I whipped up a simple system which allows one MOO-server/DB to 
@compare objects in it to those in another (which hasta run a simple 'Compare 
Daemon' using listen()/:do_login_command). Then I could run my old DB, a 
generic Dec95 DB, and my new DB simultaneously, and look at what things I'd 
changed and what things in the core had changed. It ended up only taking me 
about 3 days to get everything transfer, and everything seems to be working 
fine now.
If you want I could send you a copy of the code I used, and/or a list of 
changes between the Dec95 and Feb cores.

> running under Linux kernel 2.0.27 on a DX4-120/16 megs/16 meg swap
> partition.  Aside from needing more RAM (Like maybe 32 to 64 megs) would
> this be sufficient for a MOO that has maybe 15 players logged in at once,
> that includes a combat system (forked tasks like combat make hella lag)?
> I've found that the only time I ever get any lag with Linux, regardless of
> how many or how big of tasks I'm running, is when it has to fetch and
> retrieve pages from swap.  Just for a point of reference..
I'd think that machine would be fine, up to a DB size of maybe 20 MB, assuming 
you stick ~40+ MB RAM in it. If you write forked-tasks (especailly NPCs can be 
a bit CPU hit) efficiently, you could prob'ly get away with somewhat more.
In my experience, much more critical is your network connection. 1 or 2 
seconds server lag doesn't matter much if you're getting 6-8 seconds network 
lag. This doesn't nessecarily depend on the bandwidth to your machine (you can 
get a long way on a 28.8 modem) as much as your ISP/network provider having 
reliable, non-overworked connections.

-- 
                                  -Chris^`~'*.,_,.*'~`^P$iON-
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS/CM/IT/M/S d-- s+: a15 C++++$ UL+ P L-- E? W++(+++) N+ o? K? w++++@
O--- M-- V- PS++ PE- Y+ PGP- t+ 5++ X+ R+ tv+ b++ DI+++ D+
G++ e* h! r y- 
------END GEEK CODE BLOCK------

• Re: your mail
• From: redbeard*NOJUNKEMAIL* <redbeard@cdc.net>

• Prev: Re: your mail
• Index: MOO-cows
• Thread: MOO-cows