| Pages: [1] :: one page |
| Author |
Thread Statistics | Show CCP posts - 0 post(s) |
The Wretch
   |
Posted - 2003.08.04 14:30:00 -
[1]
Qball and I noticed a relationship between CTD's and the amount of data the server was sending to the client while monitoring it.
Just before each crash occured, there was an obscene amount of data that the server tried to send to the client - as high as 3-6 megabytes.
We both have very high bandwidth connections so the load time should not of been a problem. But still this ALWAYS caused a crash. Not only did we CTD when this happend, but several others on the server did at the same time.
Also, on the server stats you can see several time points where there are sudden drops and rebounds in player population the correlate to when I CTD myself (and others in the corp).
Don't know if anyone else has seen this relationship. This CTD is definately not hardware related.
The Wretch
Cyberdyne Systems CEO
|
StealthNet
|
Posted - 2003.08.04 15:51:00 -
[2]
Nooooooooo
It's your drivers!!! 
_______________________________________________
|
Erty
|
Posted - 2003.08.04 15:59:00 -
[3]
The CTDs that has occured this weekend is not hardware related, but others are.
Normally the CTDs are hardware related, but the last days something strange has been going on :)
This is my signature. |
The Wretch
|
Posted - 2003.08.04 16:11:00 -
[4]
I definately know its not drivers or computer hardware related on my system. I have only crashed 2-4x since I have had retail until last week. Now I have crashed more times then I can count and the only unusual thing I see relating to this is the huge amount of data the server tries to send.
The Wretch
Cyberdyne Systems CEO
|
Dragon Emperor
|
Posted - 2003.08.04 16:45:00 -
[5]
It's defenitely NOT hardware related for me, I didn't even remember CTD before patch 1128, after 1128. guaranteed CTD or Full SYSTEM REBOOT after switch display mode. and this weekend, CTD like crazy for NO REASON. Not only me, all my corp mates are SAME. And happen at SAME TIME, Don't use those silly excuse like "hardware problem. driver problem", CCP, that only makes you look silly.
|
StealthNet
|
Posted - 2003.08.04 17:57:00 -
[6]
When I CTD, I often see almost all corp members dropping too. This can't be hardware related.
Most of the time, the EVE-I monitor states that the server is down. Some seconds later, up.
_______________________________________________
|
Riddari
|
Posted - 2003.08.04 18:26:00 -
[7]
I got a blue screen (in WinXP this is a simple restart until I flip a switch which will show the actual blue screen) today, the first time for weeks.
I was trying to dock and after a 2 minute interval of nothing happening my PC rebooted.
¼©¼
a history |
solivagus
|
Posted - 2003.08.04 18:50:00 -
[8]
 Quote:
Qball and I noticed a relationship between CTD's and the amount of data the server was sending to the client while monitoring it.
Just before each crash occured, there was an obscene amount of data that the server tried to send to the client - as high as 3-6 megabytes.
How can you see how much data the server tries to send to the client?
Just curious. |
Beseb
|
Posted - 2003.08.04 19:01:00 -
[9]
I agree with Wretch on this as I've observed the same behaviour.
I haven't gone so far to capture the network traffic, but my "educated" guess definitly points to a traffic spike followed by a CTD. And like Wretch, I basically *never* CTDd before this recent rash of them.
|
BSOD
|
Posted - 2003.08.04 19:43:00 -
[10]
Quote:
Quote:
Qball and I noticed a relationship between CTD's and the amount of data the server was sending to the client while monitoring it.
Just before each crash occured, there was an obscene amount of data that the server tried to send to the client - as high as 3-6 megabytes.
How can you see how much data the server tries to send to the client?
Just curious.
Control-alt-F12
Or if you have a rather advanced router, many of them have some sweet traffic monitoring features. Probably is possible to monitor/log network traffic in Windows too. (Def. possible in Linux.)
CTDing upon getting sent large amounts of data (A severe variant of load lag) is why the Fountain Alliance has been having trouble running m0o/RUS out of the region. Last time they tried, 3/4 of their fleet CTDed at the end of warp. Needless to say, everyone who CTDed logged back in to find themselves cloned.
----------------
Blue Screen of Death
CEO
Exodus Enterprises |
slothe
|
Posted - 2003.08.04 20:05:00 -
[11]
i agree in the "strange things in the last week" theory.
i have v good computer as do other corp m8s.
we have never ctd before.
however in the last week it has been happening often, sometimes up to 3 times per log.
also it happens for us at the same time - even us / uk players go down at same time - strange.
Say hello on our forum @www.aserea.com
or join our public channel ingame "MLM Public"
http://www.khainestar.com/eve |
Raven DeBlade
|
Posted - 2003.08.04 21:57:00 -
[12]
it all started after the prolonged server maintanance Wednesday last week
"To hunt pirates you need time and patience, because even monkeys fall from the trees"
"Any statements made above this line are my persona" |
QBall
|
Posted - 2003.08.04 22:07:00 -
[13]
Edited by: QBall on 04/08/2003 22:10:32
w00t 4 MOTHER F***in restarts in the past 3 hours. I mean seriously can't you fix this, it something your server are sending to the client that is causing crashed BSOD(blue screen of deaths) and random restarts.
So basically, WTH. Ive been watching your packet o' meter thing (ctrl alt f12) and it usually gives some wierd stuff right before it crashes.
For instance Kilobyte's In tagged upto 2048 in the last 5seconds slot about the time of the crash, then it stopped recieving all data from the server both in and out shortly after that a CTD followed by a restart.
I believe the reboots are because you have some bad error handlers or something and when it tries to bring up the spash screen again it usually dies (and takes windows xp with it).
I'm almost possitive that the reboots are acually caused by EVE not terminating right due to an unforseen error. I suggest you look into the matter as a top priority because it's only a matter of time until you corrupt windows with all these damn restarts.
Oh and on the monitoring for linux all you need to do is run socks on your windows machine dump it to the socks server (linux machine) then run TCPDUMP on the linux machine to watch packet traffic. <--- havn't done that in awhile.
otherwise you can watch ingame stuff with ctrl alt f12.
In windows I think there is a win32 version of TCPDUMP but I'm not sure, and if I remember right you have to pay for it.
--------
"OMG IT'S TRAMMEL 2.0!!!!"
-QBall
And
QQ is QQ |
Rouen
|
Posted - 2003.08.04 22:08:00 -
[14]
Have had a lot of them the last day's
If you look at : http://www.eve-i.de/tool_serverstat.php
you can see that about 1000 players drops at the same time, so it's not a HW/driver client problem
|
QBall
|
Posted - 2003.08.04 22:20:00 -
[15]
It's called shabby programming. I've made a couple such problem in stuff I've worked on, but it shouldn't be that hard to fix.
--------
"OMG IT'S TRAMMEL 2.0!!!!"
-QBall
And
QQ is QQ |
Henka
|
Posted - 2005.01.26 12:12:00 -
[16]
and why is this topic brought up?
its AGES OLD!
2003-08
:)
|
Valar CCP
|
Posted - 2005.01.27 10:35:00 -
[17]
This is a bit old
------
Valar
Quality assurance department
CCP games |
| |
|
| Pages: [1] :: one page |
| First page | Previous page | Next page | Last page |