Did the duct tape go undone again? (Community)

Did the duct tape go undone again? // Community

1  |  

nornny

Dec 30, 2000, 1:27am
Someone must've spilled some leftover egg nog or something but the AWCOM
building worlds aren't up again. :/ Yet again, the invincible AWGate and
unweilding Junodome manage to survive this. :) I wonder what's wrong this
time and if we have to wait until the next millenium to log into Alphaworld.
Well, at least it's something for AW employees to put in their New Year's
resolution list. :)

Nornny

agent1

Dec 30, 2000, 2:12am
Apparently the Uniserver still isn't fixed... "a $40,000 server should spontaneously reboot..." I believe that was the quote from
Roland :/

-Agent1


[View Quote]

captain mad mike

Dec 30, 2000, 2:23am
No no no, you're all wrong. Enzo guzzled a bottle of bubbli in the
pre-new-year's party and stumbled in the server room with a few six packs.
After getting Roland and the rest of the technicians drunk he thought the
server was a lady who wanted a drink so he poured beer down the fan hole in
the power supply.

Either that or this was "another bribe from juno" but I like the above idea
better :)


--
/ |\ /| |\ /|
/ | \ / | | \ / |
\ | \/ | | \/ |
\_ | | | |

crazy glue il

Dec 30, 2000, 2:52am
Plus this time AWTeen is down :(( It sux can't build in any of my normal
places. and I have the terrible erge to BUILD
~Crazy
http://www.crazyglue.8m.com

[View Quote]

twisted horizon

Dec 30, 2000, 11:04am
LOL.... ever considered doing stand-up comedy in AW?
-Twist
[View Quote]

kah

Dec 30, 2000, 11:35am
The trial I'm having is DOWN! And trials only last for 30 DAYS! This makes
me MAD!

KAH
PS: Juno ain't brinbing AWCI but they're the ones that might sue AWCI, so
the keep the worlds up even when all the others that have payed are let
down, but they don't care after buying a world you don't have money to sue
them...
[View Quote]

ananas

Dec 30, 2000, 11:39am
Dies ist eine mehrteilige Nachricht im MIME-Format.
--------------D38B60E7AC27D022FBE8E5B0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Last time when it happened they added the days that were lost (I think
even a few more), maybe they will do it again

kah schrieb:
>
> The trial I'm having is DOWN! And trials only last for 30 DAYS! This makes
> me MAD!
--------------D38B60E7AC27D022FBE8E5B0
Content-Type: text/x-vcard; charset=us-ascii;
name="vha.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Visitenkarte für Volker Hatzenberger
Content-Disposition: attachment;
filename="vha.vcf"

begin:vcard
n:Hatzenberger;Volker
x-mozilla-html:FALSE
url:oct31.de
adr:;;Bornheimer Strasse 15;Bonn;;53111;Germany
version:2.1
email;internet:vha at oct31.de
end:vcard

--------------D38B60E7AC27D022FBE8E5B0--

kah

Dec 30, 2000, 2:08pm
That was cuzz it was down a week. Other times it has been down, they didn't
do it.

KAH
[View Quote]

captain mad mike

Dec 30, 2000, 5:50pm
No but maybe I will now ;)
[View Quote]

mauz

Dec 30, 2000, 5:54pm
[View Quote] TechTalk was cancelled on Friday, all Roland said then was:
"
Roland: server rebooted (again)...even though Sun said all problems were fixed
Roland: I have to clean up the mess
Roland: I don't know what the problem is...all I know is that $40,000 sun servers should not be spontaneously rebooting
Roland: 3.1 beta is reaaaal close...got the SDK docs done this week, check 'em out if you are interseted
Roland: I all set for next week...but I dunno...if the server reboots again all bets are off
Roland: ok...I'm off...I need to see if there's any hope of recovering alphaworld this time around...
"

The Network Status http://www.activeworlds.com/status_inworld.html says now
"
The main Active Worlds server has experienced a relapse of the hardware problem
that occurred three weeks ago, despite the fact that Sun assured us at the time that
everything was fixed. We have moved the universe and many of our worlds to a
backup server until the main server can be fixed. During this time some worlds
(including AlphaWorld) will be down. We apologize for the inconvenience.
Please direct any inquiries to support at activeworlds.com
"

Another information I got stated that the universe will be running
on the backup server until at least Tuesday. The new (temporary?) address
for auth.activeworlds.com is 166.90.181.71, so some world hosters,
especially on Linux, may have to add it to world.ini:
[universe]
host=166.90.181.71

The old hardware problem they're referring to is the one where the primary
Active Worlds server kept rebooting itself, until Sun Microsystems finally
replaced the whole motherboard. They also experienced power outages at the time,
despite of UPS at the network site. That in turn delayed the AlphaWorld
database rebuild, which takes at least 10 hours even under good circumstances.

If somebody has more info, please post it here. (AWCOM employees? Anyone...?)

--
Mauz
http://tnlc.com/mauz/

rolu

Dec 30, 2000, 9:23pm
if the database needs a rebuild every time something goes wrong, they use a
crappy database, and it's time to move on to something else. Especially if
it takes 10 hours every time something goes wrong. And if their server keeps
doing weird things they should kick Sun. Hard.

Rolu

[View Quote]

agent1

Dec 30, 2000, 10:07pm
The Alphaworld "database" could be corrupted. That's why they need to check it out. I don't know how a large database like the one(s) AW uses for its worlds, etc. don't use things like fflush() to send the buffered data to disk after "writing" each time... Something to look into, maybe...

-Agent1


[View Quote]

rolu

Dec 31, 2000, 8:28am
[View Quote] I know that. But there are databases that can't go corrupted when you flip
the switch. And there are filesystems that can't go corrupted when you flip
the switch. You'll loose the transactions that were done at the moment the
system went down, but as soon as it is up you can start using it again. If
they shell out $40k for a server, maybe they should look into that kind of
technology too. If the server AW is on goes down in an uncontrolled way now,
it will take at least 10 hours before it is up again.

Rolu

>
> -Agent1
>
>
[View Quote]

eep

Jan 1, 2001, 12:50am
Hey, if AW itself requires redownloading the cell database each time it crashes, is it that surprising to find the world server programmed just as badly?

[View Quote] > if the database needs a rebuild every time something goes wrong, they use a
> crappy database, and it's time to move on to something else. Especially if
> it takes 10 hours every time something goes wrong. And if their server keeps
> doing weird things they should kick Sun. Hard.

kah

Jan 1, 2001, 1:59pm
If they're 40 000-dollar-server doesn't work correctly, why don't they just
ask for a refund or something and move on to something that works?????????
They can do that or lower the prices for worlds and citizenships, because
why should one pay for something that has that much downtime??

KAH
[View Quote]

macb

Jan 1, 2001, 3:20pm
Yeah, makes you wonder doesn't it?

Well, we rarely get the full scoop here from AWLD, so we have to make
educated guesses. Which of course has tremendous PR value to the company.

My guess is that they don't even own the server. If you have shopped for
web hosting services you will know that you are often given the option of
hosting your web page on a dedicated server. Someone may still remember
the name of the company AWLD uses for this (the ISP), I don't. With that
companies name, you might be able to see what dedicated services they offer
and how much they cost.

Then I would try and find some info. on the health of that ISP. Maybe they
are struggling to survive and can't afford a lot of backup. Maybe
communications are strained between this company and AWLD over some issue,
billing perhaps?

One of the status messages indicated that the backup server was actually in
a different location, I'm guessing this would be the actual AWLD
headquarters, bolstering my opinion that the normal AW main server is at the
ISP location. The statement that it is a $40,000 server does not imply
that AWLD purchased the machine, only that it was an expensive machine, and
one would expect good support from Sun on it.

Where I work, we buy machines that cost that much and more from Data
General, Compaq, Dell, and Hewlett Packard on a fairly regular basis.
Having bought ONE of these machines does not make you a significant customer
to any of these companies, so you do not get anything special in the way of
support. If you buy 50 of them on the other hand, your sales rep who
collected a nice commission check, will make sure you remain a happy
customer.

The ISP in this case may have been an important customer to Sun at one time,
but if they are not doing too well, i.e. haven't bought anything lately, the
level of support tapers off rather rapidly.

Anyway, my point is that the info from AWLD is sketchy at best probably
because its 3rd hand information by the time we get it.

I'm sure that the technical people involved at AWLD are making the best of a
bad situation. They are smart, honest and dedicated as far as I have been
able to tell.

Well that's for the technical people anyway.

[View Quote]

builderz

Jan 3, 2001, 10:13am
Are you referring to http://aw2.mpl.net/ and/or http://www.mpl.net/? :)

-Builderz

[View Quote]

1  |  
Awportals.com is a privately held community resource website dedicated to Active Worlds.
Copyright (c) Mark Randall 2006 - 2024. All Rights Reserved.
Awportals.com   ·   ProLibraries Live   ·   Twitter   ·   LinkedIn