Why do the servers go down daily?
Jroot
loop.
But does anyone really know?
Duke
Jroot
retard
but what needs to maintained daily
Duke wrote in message <9HSX1.1017$p4.27...@newsgate.direct.ca>...
Latrobe Valley Computers
Jroot wrote in message ...
around, like furs ect...
Kind of like a mud rebooting, I guess....
I do notice that there is less lag after a reboot than before one, though I
haven't tried enough to be sure.
Any comments anyone ?
MoZ @ Pacific
Mehcroh
server crash the most you would lose is one whole days worth of play time.
Imagine how much you would lose if it was only a weakly backup. I work
in a large office building, on the evening shift and we do partial
server backups nightly and full backups weekly.
In article <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>, "Jroot"
Martin Tykesson
Jroot skrev i meddelandet <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>...
>The only logical reason is that theres a serious memory leak in the game
>loop.
Ever heard of backup?
diver
is a half hour playtime?
greetz,
diver
Mehcroh <Meh...@hotmail.com> wrote in message 70paii$oi5$1...@supernews.com...
>My understanding is it is mainly for backup purposes, so if there is a
major
>server crash the most you would lose is one whole days worth of play time.
>Imagine how much you would lose if it was only a weakly backup. I work
>in a large office building, on the evening shift and we do partial
>server backups nightly and full backups weekly.
>
>
>In article <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>, "Jroot"
><jr...@mediaone.net> wrote:
>>The only logical reason is that theres a serious memory leak in the game
>>loop.
>>
Garion, GL
time. At last report they were done hourly (approx)
On Fri, 23 Oct 1998 09:25:30 +0200, "Martin Tykesson"
<martin....@helsingborg.nitator.se> wrote:
>
>Jroot skrev i meddelandet <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>...
>>The only logical reason is that theres a serious memory leak in the game
>>loop.
>
>
>Ever heard of backup?
Matthew Burgos
machine, then i guess it resets because of memory holes....i know from
runnning my own servers on Windows machines....you do hvae to reset every
once in a while depending on traffic. any other i have no idea.
Latrobe Valley Computers <l...@latrobe.net.au> wrote in message
MXSX1.50$Jy4.2...@nsw.nnrp.telstra.net...
>
>Jroot wrote in message ...
>>duh....
>>retard
>>but what needs to maintained daily
>>
>>
>>Duke wrote in message <9HSX1.1017$p4.27...@newsgate.direct.ca>...
>>>Daily maintenance.
>>>>The only logical reason is that theres a serious memory leak in the game
>>>>loop.
>>>>
>>>>
>>>>
>>>
>>>
>>
>>
Jroot
Martin Tykesson wrote in message <36303...@d2o67.telia.com>...
>
>Jroot skrev i meddelandet <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>...
>>The only logical reason is that theres a serious memory leak in the game
>>loop.
>
>
>Ever heard of backup?
Dopester
>around, like furs ect...
nope... if at the time it does the last save, there are furs, empty packs
etc lying around, they will still be there when it comes back up...
Mehcroh
separate drive or media like a tape drive or optical disk. If there is a major
server crash data stored on a hard drive may be lost so the last save could be
damaged, incomplete, or completely wiped. In my experience it always seems that
a server crashes while you are trying to save the report you just spent the
last 5 hours proof reading and correcting. But as long as you have a reliable
backup stored on disk at least you don't lose the whole report you spent a
week writing.
In article <363072fa...@news.cyberstreet.com>, nospam...@usa.net
wrote:
>Nope, Backup's not the answer. Backups are run during regular game
>time. At last report they were done hourly (approx)
>
>On Fri, 23 Oct 1998 09:25:30 +0200, "Martin Tykesson"
><martin....@helsingborg.nitator.se> wrote:
>
>>
Oracle of Occlo
fragmentation usually is caused by the system more that the application.
With all the spawned tasks of people logining on and off, there is many
opertunities for consectutive free memory to be split and re-split.
If Origin had significant memory leaks the game couldn't stay up for 20+
hours.
my vote is to do the re-boot at 8-12 am Server Local Time.
foamy
>my vote is to do the re-boot at 8-12 am Server Local Time.
Why should a 24/7 service conduct disruptions at a time convenient to the
staff ? How about 2-5 am local ?
foamy/napa
Mortis
>ever heard of thinking before speaking?
>
>Martin Tykesson wrote in message <36303...@d2o67.telia.com>...
>>
>>Jroot skrev i meddelandet <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>...
>>>The only logical reason is that theres a serious memory leak in the game
>>>loop.
>>
>>
>>Ever heard of backup?
>>
>>>But does anyone really know?
They do do a backup (i think somehow a more permanent backup than the
half-hour or hour ones) but they also do physical checkups on the
servers and do some stuff with the software on the servers, to stop
big probs from happening while the servers are up.
Mortis
Master of the Unknown, KPS
"Carpe Noctum"
"Contra Malum Mortis Non Est Medicamen In Hortis"
"Ego Sum Lux Mundi. Qui Sequitur Me Non Ambulat In Tenebris."
Xalorous
once per hour that a backup was being performed. They do run hourly
backups. Sometimes though they have to go back to the previous
maintenance backup (time warps).
diver wrote in message <70ptqj$lrs$1...@news.news-service.com>...
>I thought that there is a backup every half hour; so the most you
will lose
>is a half hour playtime?
>
>greetz,
>diver
>
>Mehcroh <Meh...@hotmail.com> wrote in message
70paii$oi5$1...@supernews.com...
>>My understanding is it is mainly for backup purposes, so if there is
a
>major
>>server crash the most you would lose is one whole days worth of play
time.
>>Imagine how much you would lose if it was only a weakly backup. I
work
>>in a large office building, on the evening shift and we do partial
>>server backups nightly and full backups weekly.
>>
>>
>>In article <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>, "Jroot"
>><jr...@mediaone.net> wrote:
>>>The only logical reason is that theres a serious memory leak in the
game
>>>loop.
>>>
Xalorous
Matthew Burgos wrote in message ...
>well first off it depends on the OS. If it rns ona windowsx or NT
based
>machine, then i guess it resets because of memory holes....i know
from
>runnning my own servers on Windows machines....you do hvae to reset
every
>once in a while depending on traffic. any other i have no idea.
Something tells me that this game doesn't run on any sort of Windows.
More likely multiple workstations. Running Unix or some other similar
OS.
The daily down time is for 'maintenance'. Probably defragmenting the
harddrive and so forth.
Xalorous
foamy wrote in message ...
They're done at 11am server time.
The time is based on server load. Less people play during the day
than at any other time. Work, school, and so forth. And it doesn't
take 3 hours. Normally about an hour. Average maybe 1.5...unless
there's a patch.
Confidential
times a day last few days. And it's every DAMN time I manage to do
something difficult -- finally manage to get a newbie transported to the
guild stone after much coddling -- get her in the guild -- start to work up
her stats -- and BLAMMO -- a surprise server downing and we are BACKWARDS
over an hour -- what's that CRAP? Save the state at least. And happened
more than once. I finally get over 94 in sword and blammo -- back to 93.9.
Thanks a lot. When I get PK'd by a lamo who is blue and turns on you after
a large battle with multiple monsters -- does the server go down then --
NEVER -- @#$ @#$$@# UO auto checks for that and delays the downing.
I know some of that is just raving from bitterness -- but part of that IS a
legitimate complaint.
Friggin have scheduled downtimes and stop timewarping us 3 times a day.
Valiant. GL
Located somewhere in UO -- he never knows when he gets time warped into the
middle of wilderness -- at least due to graphic differences he can tell if
he's in old world or T2A. Geez.
Oracle of Occlo wrote in message <7104fb$62p$3...@news-2.news.gte.net>...
>Main reason is not memory leaks, but memory fragmentation. The
>fragmentation usually is caused by the system more that the application.
>With all the spawned tasks of people logining on and off, there is many
>opertunities for consectutive free memory to be split and re-split.
>
>If Origin had significant memory leaks the game couldn't stay up for 20+
>hours.
>
Oreh Ta Egral (GM Lagger)
>I don't mind the scheduled down time -- I just mind that GL does down like
3
>times a day last few days. And it's every DAMN time I manage to do
>something difficult -- finally manage to get a newbie transported to the
>guild stone after much coddling -- get her in the guild -- start to work up
>her stats -- and BLAMMO -- a surprise server downing and we are BACKWARDS
>over an hour -- what's that CRAP? Save the state at least. And happened
>more than once. I finally get over 94 in sword and blammo -- back to 93.9.
>Thanks a lot. When I get PK'd by a lamo who is blue and turns on you after
>a large battle with multiple monsters -- does the server go down then --
>NEVER -- @#$ @#$$@# UO auto checks for that and delays the downing.
>
>I know some of that is just raving from bitterness -- but part of that IS a
>legitimate complaint.
>
>Friggin have scheduled downtimes and stop timewarping us 3 times a day.
>
>Valiant. GL
I really get tired of people slamming Great Lakes. You just leave it
alone!!!!! Great Lakes is the most consistant of all servers. It is alway
1st down for maintenance, and the last to come back up. It is #1 also for
crashes and timewarps, no other server can match the frequency. Everyone
should aspire to be the best at something. 8-)
--
Oreh Ta Egral (CoP)
GM Lagger, GM Connection Lost, GM Dead Guy
Great Lakes
>
Xalorous
Oreh Ta Egral (GM Lagger) wrote in message
<71djtj$k...@news.nfinity.com>...
Umm, Lag Superior is probably in close competition.
IROC
Its to IPL and run backups (full tape Backups of everything not just
"snapshots")
the IPL cleans all the Garbage from memory and reboots for a clean start.
Granted my servers only handles orders from vendors and customers,and not
anything as "important" as Ultima online, but I'm sure the reasons are the
same.
Jroot wrote in message <7wSX1.24$Y73.8...@elnws01.ce.mediaone.net>...