| Pages: [1] :: one page |
| Author |
Thread Statistics | Show CCP posts - 3 post(s) |
Jaabaa Prime
   |
Posted - 2005.09.22 23:42:00 -
[1]
I've been playing EVE since the last beta phase and we used to be proud to get 4000+ players online at the same time after gemini was released.
We went through the last HW update (over a year ago) with a few issues and a few unplanned downtimes and we kept on trucking.
Now I see a massive influx of players to EVE and it would appear that the servers can't keep up.
Oveur has already said that there is new server hardware in planning, but the question is when can we expect to see these new nodes enter the cluster ?
I was personally only hit once by the extreme lag reported by others here when I went to pick up a spare ship, and god knows that 2 minutes between giving a command and the ship moving is a mighty long time. And that's without a fleet battle and only 10 in local, and fitting the ship was a nightmare to says the least.
To you that are new to EVE I can say but 1 thing, the current lag situation is not normal and that is why I feel compelled to ask CCP what will happen in the immediate future to resolve it.
Don't get me wrong, I'm very happy that EVE is gaining in popularity but I'm wondering when the server cluster will be able to cope with the new found "load" caused by the new members of the EVE universe.
--
Intergalactic Teeth Pullers
"Only two things are infinite, the universe and human stupidity, and I'm not sure about the former."
Albert Einstein |
DigitalCommunist
|
Posted - 2005.09.23 03:48:00 -
[2]
Todays problem wasnt the result of so many people online because we get 12k almost daily now with no probs.. it was probably something else that made a few nodes fail.
Perpetually driven, your end is our beginning.
|
Sorja
|
Posted - 2005.09.23 04:39:00 -
[3]
Last battle we had against BoB (read: should have had), everybody went home after hearty handshaking because we managed to crash the node with 75 people in local so we all agreed it was pointless engaging combat.
Retarded.
Kill mails |
infused
|
Posted - 2005.09.23 04:48:00 -
[4]
I'd say it is the influx of people. Why else are they changing load balancing?
Something is wrong somewhere.
|
K'thang
|
Posted - 2005.09.23 06:41:00 -
[5]
 Originally by: infused
I'd say it is the influx of people. Why else are they changing load balancing?
Something is wrong somewhere.
It's probably a mix of things, but aye I'd say more people must play a role in this. I played in beta too, but I've had some breaks along the way. But when I reactivated my account a few weeks back it was bad from the start and got worse each day. Yesterday was "a laugh" though  |
F4ze
|
Posted - 2005.09.23 09:22:00 -
[6]
At least one of the things I'm seeing is that the hourly "ticks" on the server that keep my POS going are getting later and later.
A batch of moon materials comes in every hour. It used to be at xx:03h every time but has been getting later the last few weeks. At the moment my POS ticks at xx:18h.
|
Avon
|
Posted - 2005.09.23 09:40:00 -
[7]
Maybe the problem is that the new players start in the same few systems, causing those and the other systems on the same node to become unstable?
The influx of players may have something to do with the problem, not due purely to numbers on the cluster, but the distribution of those numbers.
______________________________________________
The Battleships is not and should not be a solo pwnmobile - Oveur |
turnschuh
|
Posted - 2005.09.23 10:11:00 -
[8]
no, thats bs
see:
http://www.gameindustry.com/interview/item.asp?id=39
Petursson: Everyone will be able to play in the same universe. We can easily double the capacity of our server cluster without major technological changes. Above that we will need to make some underlying changes, but in theory there is really no hard limit on how many players can play together in one world if it is done correctly.
|
Matthew
|
Posted - 2005.09.23 10:28:00 -
[9]
Originally by: turnschuh
no, thats bs
see:
http://www.gameindustry.com/interview/item.asp?id=39
Petursson: Everyone will be able to play in the same universe. We can easily double the capacity of our server cluster without major technological changes. Above that we will need to make some underlying changes, but in theory there is really no hard limit on how many players can play together in one world if it is done correctly.
Important things to note about that quote:
1) It doesn't say that it can be done with the current hardware, just that it can be done. Hence we may have some pains right now until the new kit gets delivered.
2) He's talking about the number of people in the universe. This is a very different issue from the number of people in a system. If you think of the universe as a hotel, and a system as a room within that hotel, it's obvious that you can get more people into the hotel by building an extension, but the extension won't let you get any more people into the same room. The problems at the moment seem to mostly be caused by the huge blob systems overwhelming nodes. As already mentioned, new players are more likely to reside in high-sec empire systems - the very area that is already overcrowded. The server would probably be quite happy and lag-free if we got everyone to spread out.
You can do anything. But you can't do everything. |
Bring Cranker
|
Posted - 2005.09.23 11:56:00 -
[10]
Originally by: Matthew
Originally by: turnschuh
no, thats bs
see:
http://www.gameindustry.com/interview/item.asp?id=39
Petursson: Everyone will be able to play in the same universe. We can easily double the capacity of our server cluster without major technological changes. Above that we will need to make some underlying changes, but in theory there is really no hard limit on how many players can play together in one world if it is done correctly.
Important things to note about that quote:
Well put Matthew, and he even states "Above that we will need to make some underlying changes.." as we all know that extending the hardware is not the end all be all.
Yes, there has been lag problems lately - severe yesterday - but thats to some extend to be expected in MMORPGs where they cant 100% control the load of systems.
That being said, I hope they are working on solutions as we speak. Because its not going to get better by the apparently accelerating influx atm.
|
|
Valar
|
Posted - 2005.09.23 12:51:00 -
[11]
Edited by: Valar on 23/09/2005 12:52:00
The changes to loadbalancing that were made yesterday were to work around a bug in the loadbalancing that caused nodes to register like they had negitive load, causing the cluster to allocate too many systems on the affected nodes. That caused *extreme* lag on the affected nodes.
------
Valar
Quality assurance department
CCP games
How to write a good bugreport |
|
|
Oveur
|
Posted - 2005.09.23 13:17:00 -
[12]
Originally by: Valar
Edited by: Valar on 23/09/2005 12:52:00
The changes to loadbalancing that were made yesterday were to work around a bug in the loadbalancing that caused nodes to register like they had negitive load, causing the cluster to allocate too many systems on the affected nodes. That caused *extreme* lag on the affected nodes.
Yup, about 4000 of the players online were affected by this in one way or the other.
We have different hardware upgrades planned, one being specialized solid state hardware. However, that's coming from Texas so Rita is putting a dent in those plans 
We're also doing a some optimizations these days, some of which we are able to deploy as server hotfixes if QA signs off on them.
_____________________________
"There's no reason to become alarmed, and we hope you'll enjoy the rest of your flight. By the way, is there anyone on board who knows how to fly a plane?" |
|
Fortior
|
Posted - 2005.09.23 13:30:00 -
[13]
Originally by: Oveur
Originally by: Valar
Edited by: Valar on 23/09/2005 12:52:00
The changes to loadbalancing that were made yesterday were to work around a bug in the loadbalancing that caused nodes to register like they had negitive load, causing the cluster to allocate too many systems on the affected nodes. That caused *extreme* lag on the affected nodes.
Yup, about 4000 of the players online were affected by this in one way or the other.
We have different hardware upgrades planned, one being specialized solid state hardware. However, that's coming from Texas so Rita is putting a dent in those plans
We're also doing a some optimizations these days, some of which we are able to deploy as server hotfixes if QA signs off on them.
And isn't there a very recent blog by someone (porkbelly was it?) talking about moving from 32-bit architecture to 64-bit or somesuch?
|
|
kieron
|
Posted - 2005.09.23 14:24:00 -
[14]
Yes, Porkbelly posted a blog on the migration to 64 bit and some other things. You can see a recap of three blogs touching on lag and load balancing in this forum post.
kieron
Community Manager,
EVE Online
|
|
Andrue
|
Posted - 2005.09.23 14:31:00 -
[15]
Edited by: Andrue on 23/09/2005 14:31:41
Originally by: Oveur
Originally by: Valar
Edited by: Valar on 23/09/2005 12:52:00
The changes to loadbalancing that were made yesterday were to work around a bug in the loadbalancing that caused nodes to register like they had negitive load, causing the cluster to allocate too many systems on the affected nodes. That caused *extreme* lag on the affected nodes.
Yup, about 4000 of the players online were affected by this in one way or the other.
Yeah I was smirking at the occasional complaint around 2300. I had a good evenings mining in ANOther system not so far from that. Before logging off I decided to fly to Amarr for a little shopping. Left Irnal for Amarr and...
...I..b.e.g.a.n...t.o...s.e.e...w.h.a.t...t.h.e.y...m.e.a.n.t.
Luckily by the time I'd tidied the kitchen up and put the budgie to bed my ship had entered warp toward the station 
I just need to be able to get back out of Amarr before it all goes wobbly tonight :)
--
(Battle hardened miner)
[Brackley, UK]
WARNING:This post may contain large doses of reality. |
GouldFish
|
Posted - 2005.09.23 16:03:00 -
[16]
It's great to hear the from the dev's telling us in there own words the issues, and not trying to hide it under the carpet.
You guy's are adoing a great job, keep up the good work and have your self's a nice cold beer :)
|
Karash Amerius
|
Posted - 2005.09.23 16:07:00 -
[17]
Dev posting like this goes a long ways...
Thanks for the heads up guys.
|
Katamarino
|
Posted - 2005.09.23 17:18:00 -
[18]
We love you Devs 
|
Taketa De
|
Posted - 2005.09.23 17:50:00 -
[19]
Any chance you can enhance your software so systems on nodes can be redistributed on the fly?
(In a way like VMWare can move an active, running system from one VM on one piece of hardware to another VM on another piece hadware with no dowtime)
Especially with an increase in players it would seem that the chances of sudden player "concentraion" happening goes up and systems that never get emptied except during downtime would prevent any rebalancing of them except at downtime.
|
| |
|
| Pages: [1] :: one page |
| First page | Previous page | Next page | Last page |