Error 90000 & lag

[Raist]

how do i get them to change the rout?

i've asked AT&T and they said they cannot change how it connects, because they only provide access to the internet.... should i ask SE?

Demand escalation to someone who can. They have agreements with the big 3 top level ISP's...if nothing else they can get you to run through an alternate partner to see if things improve. TWC does it for me on a regular basis... but I've been dealing only with Tier3 technicians, not the people that answer the phone at the help desk. It also can't be done by the guys who come test your local setup either--this is something at a higher networking tier.

Just to test the idea out, you can sign up with a VPN service only for their free trial period... if things improve, it's a pretty solid sign that they need to address a routing issue between you and SE.

If after all that, they still refuse to budge.. you can contact the local regulatory authority. Details should be provided with your bill. They sometimes print it on the back of the stub you send in for payment... but it is supposed to be provided with your billing statement.

[reliquum]

I have an ATS (tier3 tech) assigned to my account and we have been trying everything over the last two weeks, he is even testing my line at night when he is at home. In what way do i need to post the question/statement to be rerouted elsewhere? In a way they will completely understand what I am saying, and actually do it and not ignore me. He HAS suggested I try a VPN, but they are all pay monthly. Guess I could try the trial and tell the Tech on Monday when he is back at work if it works. Raist, do you suggest one specifically? On Tuesday or Wednesday they will be moving me to a lesser used node. It worked in my old place when there was issues with games, I am hoping it will here. It is only SE, as I have no issues what so ever with WoW, Terra, Rift, Neverwinter, LoTR, LoL and D3..... no lag, no DCing, nothing. They all work perfectly. When there is lag so bad I am kicked off of FF14, i will log into WoW and see if that works, and it does. I'll even turn on netflix on my xbox, download crap on another computer and have WoW, no lag. I'll turn all that off, have only my computer online and log back into FF14 and it is so lagy. My computer is hard wired. With a new cat5 cable.



Also, asking if someone has rebooted their router/modem/computer is as insulting as asking if their computer is even turned on. ^_~

[Raist]

Traceroutes are a big help... especially if they clearly show a bottleneck happening with someone like Level3, Cogent, TATA, etc. (their routing partner that is taking you from AT&T's network across country and into Montreal to SE's ISP). Seeing dramatic improvement via a VPN is another good clue that there is a routing issue... they will need to trace your specific route so they get a better idea of where you may be bogging down in route (again, having tracert samples can be a big help).

Mileage may vary depending on who you are dealing with. IDK why I was able to get through with less effort than some... I basically just told them I was seeing high packet loss and lag spikes along specific routes and insisted I talk to Tier3 so I can get the data to them and they eventually put me through after I played their game of rebooting things and sending technicians to verify cabling and such was good. I basically forwarded my modems logs along with ping/trace data to my server, as well as some samples for other services like google/youtube and such that would also show issues at times. They've also been catching a lot of flack via their forums and twitter over similar issues as well. We were also dumping data on them through their posted help email address. IDK if I just timed it right to coincide with them eventually getting enough data to prompt them to take action, or if they were just tired of me harassing them each week (sometimes more often). Now I just send them the data when I have issues and they fix it a few days later. Sometimes, they catch it before me and change it automatically (modem will reboot on it's own, as it sometimes calls for it to make some changes). And by fixing it, I mean they change the route I take. A while back, they took me off the gateway for Florence and put me on Hilton Head's gateway. Other than that, they change the routing partner that gets me out of either Atlanta, Charlotte, or Raleigh-Durham and carries me up to Toronto/Montreal area to hit Ormuco. The route changes every couple of weeks amongst Level3, TATA, and Cogent (sometimes I get on PSI-Net, who is part of Cogent and has a strong peering link with Ormuco... but it unfortunately gets congested frequently, which sucks because the response time is amazing when I first get switched there).

Oh, and that's why people ask about rebooting the network devices by the way. Sometimes it is needed to do things like acquire new channel alignments and such. A quick reboot can sometimes wind up putting you on another gateway or a cleaner primary channel, potentially clearing up a lot of issues.

As for the VPN's, some that have decent free options don't work too well with games (will filter out things that aren't typical website traffic and such). Tunnel Bear worked well for me with the game. You get 500MB a month for free, can get extra data if you promote them through social media. To subscribe it can run between $4 and $5 a month depending on how you buy. GhostVPN worked too when I tried it last year. Can get some add pop-ups with their free version, and it's like a 3 hour limit per week or something like that. Hotspotshield was another that worked.. but it was ad supported for the free version as well, which could cause some issues. There are more out there, but those were the ones I tinkered with the most last year. Of those 3, Tunnel Bear seemed to be the more consistent for me on speed/latency when I tested them, so I stuck with their free version in case I needed it again. Still works today... but don't use it for the game (haven't needed VPN tweak for almost a year now). I just use it to secure my line for some other things I do from time to time, and 500MB is plenty for me to do that stuff.

[reliquum]

Hope this comes out ok. I have been doing that and sending it to them, this is a copy of the ones I did last week. Aaaaand, noticing an awful lot of level3's...... I can physically fix computers. build them .tear them apart and rebuild. software and programming and anything not physical and I am lost. My modem is rebooted every few days along with my computer. Aside from the level 3 pings.... anything i can point out to them i need to be routed away from?

And Raist, THANK YOU. so far over the last few weeks, this has been going on for close to 2 months. you've been the only person who has actually pointed me in a direction i have hope for fixing this issue!



C:\Users\USER>tracert 199.91.189.40

Tracing route to 199.91.189.40 over a maximum of 30 hops

1 1 ms 1 ms 1 ms homeportal
2 * * * Request timed out.
3 * * * Request timed out.
4 24 ms 26 ms 23 ms 12.83.80.189
5 26 ms 27 ms 26 ms gar26.dlstx.ip.att.net [12.123.16.85]
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 67 ms 68 ms 67 ms ae-14-14.ebr2.Chicago2.Level3.net [4.69.151.117]

10 * * * Request timed out.
11 * * * Request timed out.
12 68 ms 69 ms 69 ms ae-10-10.car2.Montreal2.Level3.net [4.69.153.86]

13 82 ms 71 ms 68 ms ORMUCO-COMM.car2.Montreal2.Level3.net [4.59.178.
74]
14 * * * Request timed out.
15 156 ms 141 ms 141 ms 192.34.76.2
16 147 ms 152 ms 145 ms 199.91.189.234
17 144 ms 143 ms 144 ms 199.91.189.40

Trace complete.

C:\Users\USER>tracert 199.91.189.40

Tracing route to 199.91.189.40 over a maximum of 30 hops

1 3 ms 1 ms <1 ms homeportal
2 24 ms 24 ms 22 ms 108-207-104-2.lightspeed.rcsntx.sbcglobal.net [1
08.207.104.2]
3 * * * Request timed out.
4 24 ms 26 ms 26 ms 12.83.80.189
5 25 ms 27 ms 27 ms gar26.dlstx.ip.att.net [12.123.16.85]
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 72 ms 74 ms 73 ms ae-14-14.ebr2.Chicago2.Level3.net [4.69.151.117]

10 * * 68 ms ae-1-100.ebr1.Chicago2.Level3.net [4.69.132.113]

11 * * * Request timed out.
12 68 ms 67 ms 67 ms ae-10-10.car2.Montreal2.Level3.net [4.69.153.86]

13 68 ms 69 ms 68 ms ORMUCO-COMM.car2.Montreal2.Level3.net [4.59.178.
74]
14 * * * Request timed out.
15 151 ms 149 ms 139 ms 192.34.76.2
16 143 ms 128 ms 123 ms 199.91.189.234
17 137 ms 145 ms 147 ms 199.91.189.40

Trace complete.

Tracing route to 199.91.189.40 over a maximum of 30 hops

1 1 ms 1 ms 1 ms homeportal
2 * * * Request timed out.
3 * * * Request timed out.
4 24 ms 26 ms 23 ms 12.83.80.189
5 26 ms 27 ms 26 ms gar26.dlstx.ip.att.net [12.123.16.85]
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 67 ms 68 ms 67 ms ae-14-14.ebr2.Chicago2.Level3.net [4.69.151.117]

10 * * * Request timed out.
11 * * * Request timed out.
12 68 ms 69 ms 69 ms ae-10-10.car2.Montreal2.Level3.net [4.69.153.86]

13 82 ms 71 ms 68 ms ORMUCO-COMM.car2.Montreal2.Level3.net [4.59.178.
74]
14 * * * Request timed out.
15 156 ms 141 ms 141 ms 192.34.76.2
16 147 ms 152 ms 145 ms 199.91.189.234
17 144 ms 143 ms 144 ms 199.91.189.40

Trace complete.

============================================
Trace : 199.91.189.40
------------------------------------------------------------------------------------------------------
local -> IP= RESERVED
TTL= 1 IP= 24 ms RESERVED
TTL= 2 TimedOut
TTL= 3 TimedOut
TTL= 4 IP= 12.83.80.189 26 ms UNITED STATES
TTL= 5 IP= 12.123.16.85 29 ms UNITED STATES
TTL= 6 TimedOut
TTL= 7 TimedOut
TTL= 8 TimedOut
ICMP Errors
=============================================


Microsoft Windows [Version 6.3.9600]
(c) 2013 Microsoft Corporation. All rights reserved.

C:\Users\USER>pathping 199.91.189.40

Tracing route to 199.91.189.40 over a maximum of 30 hops


1 homeportal
2 108-207-104-2.lightspeed.rcsntx.sbcglobal.net [108.207.104.2]
3 * * *
Computing statistics for 50 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address

0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% homeportal
0/ 100 = 0% |
2 25ms 0/ 100 = 0% 0/ 100 = 0% 108-207-104-2.lightspeed.rcsntx.sb
cglobal.net [108.207.104.2]

Trace complete.

C:\Users\USER>pathping 199.91.189.40

Tracing route to 199.91.189.40 over a maximum of 30 hops


1 homeportal
2 108-207-104-2.lightspeed.rcsntx.sbcglobal.net [108.207.104.2]
3 * * 70.143.193.134
4 12.83.80.189
5 gar26.dlstx.ip.att.net [12.123.16.85]
6 * * *
Computing statistics for 125 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address

0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% homeportal
0/ 100 = 0% |
2 25ms 0/ 100 = 0% 0/ 100 = 0% 108-207-104-2.lightspeed.rcsntx.sb
cglobal.net [108.207.104.2]
0/ 100 = 0% |
3 25ms 0/ 100 = 0% 0/ 100 = 0% 70.143.193.134
22/ 100 = 22% |
4 28ms 22/ 100 = 22% 0/ 100 = 0% 12.83.80.189
78/ 100 = 78% |
5 --- 100/ 100 =100% 0/ 100 = 0% gar26.dlstx.ip.att.net [12.123.16.
85]

Trace complete.

Microsoft Windows [Version 6.3.9600]
(c) 2013 Microsoft Corporation. All rights reserved.

C:\Users\USER>ping 199.91.189.40

Pinging 199.91.189.40 with 32 bytes of data:
Reply from 199.91.189.40: bytes=32 time=136ms TTL=241
Reply from 199.91.189.40: bytes=32 time=141ms TTL=241
Request timed out.
Reply from 199.91.189.40: bytes=32 time=134ms TTL=241

Ping statistics for 199.91.189.40:
Packets: Sent = 4, Received = 3, Lost = 1 (25% loss),
Approximate round trip times in milli-seconds:
Minimum = 134ms, Maximum = 141ms, Average = 137ms


C:\Users\USER>ping 199.91.189.40

Pinging 199.91.189.40 with 32 bytes of data:
Reply from 199.91.189.40: bytes=32 time=146ms TTL=241
Reply from 199.91.189.40: bytes=32 time=142ms TTL=241
Reply from 199.91.189.40: bytes=32 time=140ms TTL=241
Request timed out.

Ping statistics for 199.91.189.40:
Packets: Sent = 4, Received = 3, Lost = 1 (25% loss),
Approximate round trip times in milli-seconds:
Minimum = 140ms, Maximum = 146ms, Average = 142ms

[Raist]

EEP! Texas. There's been some weird stuff going on over there for a while now. A LOT of chatter from Austin area, a little around Dallas. So, there could be some localized issues at play. On the hopeful side though, you guys are getting a lot of new fiber rolled out around the state that will (eventually) pay off once they can get it in play to thin out some of the congestion.

You have a lot of triple star events, so it's hard to get a clear picture off the tracerts. Doesn't necessarily mean those hops are having issues... they are most likely just set to not respond to ICMP ECHO requests. But, on the other hand... that election could have been made because of congestion problems and they are just dropping the request to negate that potential overhead. What is of concern is when you get a * and a response at the same hop--like you did at a common problem area:

Chicago, with Level3

That definitely sends a flag on the field there. It shows up quite often with issues for Level3, as well as a few other corridors. Chicago is an exchange we are stuck with too because of the NSA crud... it often turns into a craps game trying to find which peer/transit partner is going to get along better throughout the week. AT&T peers with multiple companies, so that might be something they ook into--getting you guys handed off to either Cogent or TATA for a bit to see if things improve. How they orchestrate that is up to them.. but it's something they could be looking into. It may just happen as a side effect of them breaking you off to a different node. It's funny how sometimes it's the simplest fix that works....and sadly it can take way too much effort getting them to split a node up.

Hopefully they've taken note that you are getting intermittent time outs at your early hops (108.207.104.2 and 70.143.193.134 at hops 2 nd 3), and they'll get right on that and split you to a different gateway soon. Something wigging out on you right there in the first three hops from your door just compounds itself and can make things really ugly when you hit another hiccup down the line.

You may want to try testing against a couple other services as well. If you can capture issues elsewhere, it tends to prod them along a bit more. Could try things like YouTube or Google, twitchtv.com, hulu.com, etc... other common services and not just for this particular game. IF you play other online games and have server addresses to them, they could prove useful too though, so test them as well if you can. The more ways you can demonstrate the pattern, the more they have to work with and hopefully the more inspired they will be to do something about it.

[stellar_force]

reliquum, I had att while playing ffxiv. Same lag around same time. I can save you all this technical crap and time and tell you to just switch your isp (preferably cable) I switched to my alt ISP and it completely fixed the lag. Att aint fixing crap anytime soon and u can enjoy the game lag free now by switching. If you don't have an alt ISP at your location... Your screwed, and cancel your sub. Case closed.

[reliquum]

Stellar, AT&T is the only choice I have. They even bumped me from 18/1.5 to 30/2mb to see if that is the issue. They gave me a 1gb modem. Someone who is assigned to the account for the issue. They are trying, but as Raist said, they are not doing *all* they can. That is what I will be discussing with them on Monday.


Raist, I will prolly link this thread to him so he can see what you have said and look at the links you also provided. Plus you put things WAY more articulate than I could. In the mean time, I will be pinging/trace routing other things when there are, and when there are no issues to see if there is a common theme. Can't really thank you enough =D

[Brightwizard]

Try VPN service. I know it's expensive but it's worth mentioning so you can troubleshoot effectively.

[VoltaAsura]

VPN isn't all that expensive....hell there are a few out there that offer trials before you buy/sub.

I have used WTFast since Titan HM and have never looked back.

I subscribe to blast internet from Comcast, at 105mb/s - fast as s#!t, but not with ARR.

To reiterate what others have said on this post and many other threads on this forum....the issue is how your connection is connected to the game server. It has nothign to do with bandwidth - as this game does not require a lot (most MMOs don't). Also, there is honestly nothing that your ISP can do, as it is out of their hands.

Using a VPN can really help and give you a more direct connection and ultimately lowers your latency. For instance, when I use WTFast, I utilize the server located in Toronto, Canada - the same city the NA server is located and it truly is night/day for me ingame. Sure you shouldn't have to go to these lengths to have a responsive connection for a game you pay to play, but there's only such much that one can do. For me, I've decided that I don't care anymore and I will use WTFast for as long as I need to, as it is the one and only thing that has helped. I wish I knew of this when 1.0 was out >.>

I think I pay 5.99 a month...and I will admit, I was HIGHLY skeptical about trying it at first. I wasn't a big fan of having to pay for something like this to eb able to play and not have lag all the time, however the free trial got my curiosity. So I gave it a go. I pick the game it want it to load, and select a server around the world that I want it to connect to - and it is just the best thing ever. I want to point out that I am not a rep for them, nor am I trying to sell their products to anyone...I just know it helped for me, and surprised me at how well it works. It wasn't even hard to set up.

[Raist]

There is actually much your ISP can do... it's more an issue of them not wanting to do it. Not for the past decade or so at least... but some are finally warming up to the idea after the cat was let out of the bag over how bad they've neglecting their international routing agreements.

It's an issue of oversold markets and undermaintained networks and peering/transit/IPX agreemants amongst the various ISP's hat interconnect to form the world wide web. By not acquiring additional ports/bandwidth at the exchange points to better manage their increases in usage over the years, we are now having more congestion issues then ever before. Until they revisit these arrangements to stem the flooding of their ports, we will continue to have delayed/dropped packets.

Oh.. and the servers are in Montreal, not Toronto. Many people get routed there normally on the way to Montreal, so mileage may vary on whether the Toronto VPN exit point benefits them, depending on how their route behaves normally. There are some services with exit points in Montreal. Some may find they fair well with exiting at Seattle if it's an option (can ride the Vancouver exchange instead of running through the north east corridor). Other's may fair better exiting somewhere in a more centralized area of the US. It all depends on how your route normally runs as to which would be better for you. Try different points in the VPN's list to see how it does. Can connect and just run some extended pings (like with a -n 25 switch or something instead of just the default 4 pings) to your server and note which seems to be the most consistent, then play it for a while to see how it "feels". I was surprised to find one out of Hungary was actually pretty steady once back around launch... which was quite bizarre to say the least.