All the lag, network issues are server side [CONFIRMED]

[Nekodar]

It is if the server is specifically timing out certain IPs to improve service going to others. I think it's kind of funny that an SE rep specifically told me it was the server, and you guys are still acing like that isn't the problem.

1) Lag issues are from congestion
2) A small number of 90k's can be from the server
3) As can be proven by plenty of people, not all 90k's are SE's fault

Don't read more into the reply you got than is needed.

[Appleh4x]

1) Lag issues are from congestion
2) A small number of 90k's can be from the server
3) As can be proven by plenty of people, not all 90k's are SE's fault

Don't read more into the reply you got than is needed.

I'm not reading more into it, the SE guys didn't have me run any tests, this is what they flat out told me was the problem. My own traceroutes I have run here on the forums are fine and the ones you are referring to about other people "proving" some aren't SE's fault is BS. A 200 ms ping is 1/5 of a second. That is what cogent is giving those people, 1/5 a sec of lag. What everyone is complaining about is actually several full seconds of lag, timeouts, etc that are the fault of the server. You guys have been wrong from the start and know next to nothing about how a traceroute actually even works. I saw someone blame a lone * (timeout) as the problem in someone's traceroute, not a full timeout mind you, but 1 lone asterisk.

[Nekodar]

I'm not reading more into it, the SE guys didn't have me run any tests, this is what they flat out told me was the problem. My own traceroutes I have run here on the forums are fine and the ones you are referring to about other people "proving" some aren't SE's fault is BS. A 200 ms ping is 1/5 of a second. That is what cogent is giving those people, 1/5 a sec of lag. What everyone is complaining about is actually several full seconds of lag, timeouts, etc that are the fault of the server. You guys have been wrong from the start and know next to nothing about how a traceroute actually even works. I saw someone blame a lone * (timeout) as the problem in someone's traceroute, not a full timeout mind you, but 1 lone asterisk.

90000 disconnect is not the same as the lag issues. They are two separate problems.
Lag issues come from two main points: 1) Server congestion (affects most people on at the time), 2) Packet shaping (affects only people whose path is being shaped)

I'm not saying the 90000 error is never on SE's side, but in general it's not, and it's also not the same thing as the lag issues... Please understand the difference between these issues and that there are more ways than one that things can go wrong.

[Gonzothegreat198]

90000 disconnect is not the same as the lag issues. They are two separate problems.
Lag issues come from two main points: 1) Server congestion (affects most people on at the time), 2) Packet shaping (affects only people whose path is being shaped)

I'm not saying the 90000 error is never on SE's side, but in general it's not, and it's also not the same thing as the lag issues... Please understand the difference between these issues and that there are more ways than one that things can go wrong.

I'm really starting to wonder, when they did that upgrade, did they just add servers or did they also increase the bandwidth and capabilities of the networking hardware and infrastucture to handle the increased amount of incoming connections as a result of the increased server capacity.

[Cynthis]

I'm really starting to wonder, when they did that upgrade, did they just add servers or did they also increase the bandwidth and capabilities of the networking hardware and infrastucture to handle the increased amount of incoming connections as a result of the increased server capacity.

That might be a very good point. I've been able to get on twice since they did that so they screwed something up.

[ntall1]

I'm really starting to wonder, when they did that upgrade, did they just add servers or did they also increase the bandwidth and capabilities of the networking hardware and infrastucture to handle the increased amount of incoming connections as a result of the increased server capacity.

This is the reason that every mmo ever has had all their servers in many different data centers around the world, and not in one single location.

[Gonzothegreat198]

This is the reason that every mmo ever has had all their servers in many different data centers around the world, and not in one single location.

Hell doesn't blizzard not only have separate NA and EU Centers, but also East and West coast centers? Dang it, now i'm curious and actually have to look it up...crap...wow...double crap....

[Gonzothegreat198]

Hell doesn't blizzard not only have separate NA and EU Centers, but also East and West coast centers? Dang it, now i'm curious and actually have to look it up...crap...wow...double crap....

Hah I was right they do, heck they added a few servers since I last played, holy crap...http://www.wowwiki.com/US_realm_list_by_datacenter