-
New 90002 error
Despite all the 90k related posts flooding the boards after the last patch, this is not about that.
I've had my fair share of 90k's in the past, but i seem to be having a new problem now. I just moved into my new apartment yesterday, signed up with Time Warner Cable and the service started this morning. Getting 55-57mbs download, 5mbps upload, way faster then i've ever had before. The problem im having is since the service started ( about 12 hours ago) I have been getting kicked off FFXIV every couple of hours with a 90002 error message. Internet stays connected just fine and nothing else is disturbed but FFXIV. I have never seen a 90002 before today and i have been playing since launch.
I've done some reading, and it was suggested that while 90k error comes with d/c from the server's end, 90002 error comes from the clients side. I'm not sure how correct this is.
I am wired into a Arris DG1670 wirless modem/router provided to me from TWC.
Does anyone have any idea what might be causing this brand new 90002 error im getting?
-
Time Warner has been having issues off and on for certain regions. Here's a map image that updates peiodically throughout the day:
http://allestoringen.nl/maps/20096/latest.jpg
The source is available off the downdetector company page here:
http://downdetector.com/status/time-warner-cable
They track a lot of networks, including PSN, XBL, and some other ISP's and services like Netflix and Hulu:
http://downdetector.com/companies
Ultimately, you may have to get the IP your client is using for your specific server (netstat, resmon, or from your router's management pages if it has a full traffic monitor suite) and run some tracert/traceroute tests to find where the connection is lagging or dropping packets and forward the results to TWCable.Help@twcable.com with a brief explanation of what you are experiencing.
We've been hammering them for a while over the issues we've found for a while now... it shouldn't be something entirely new for them.
-
Im sorry but I am not entirely sure I understand what your saying here. I'm not having any problems with outages or net being down at all. I live in Austin, Texas, and I had Time Warner Cable before i moved. I only moved about 10 miles south from where i was before. I've never even seen a 90002 error before i moved, so i figured it had to be a routing problem or something like that.
-
Texas is an area that has been having routing issues. There's been complaints in various forums about it (including TWC's own forums). You may be on a different gateway than you were before, ultimately changing your pathing to your game server. This has happened to us in Florence, SC---they've got us running all the way to Summerville now (like 95 miles away or something... 1.5-2hr drive depending on which way you go). We used to go to either Columbia or Myrtle Beach depending on which side of town we were in. That simple change in our first hop has been shown to inject lag spikes of up to 1 second or more during times of heavy load or really foul weather--when that combines with screwiness at the Canadian border as well, it makes the game unplayable until the congestion subsides or they put us on a different route.
Really wish they would just give Florence it's own localized gateway... been running cable internet in this town for almost 20 years (1M/256k Beta was in 1996). They've made their share of money off of us... time for them to invest in some decent @#$@#$ hardware.
Code:
C:\Windows\System32>tracert 199.91.189.25
Tracing route to 199.91.189.25 over a maximum of 30 hops
1 1 ms <1 ms <1 ms LPTSRV [10.10.100.1]
2 49 ms 23 ms 253 ms cpe-098-025-064-001.sc.res.rr.com [98.25.64.1]
3 28 ms 11 ms 15 ms cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
4 25 ms 26 ms 28 ms xe-7-0-0.rlghncpop-rtr1.southeast.rr.com [24.93.64.40]
5 43 ms 30 ms 32 ms 107.14.19.44
6 31 ms 32 ms 34 ms ae1.pr1.dca10.tbone.rr.com [107.14.17.202]
7 31 ms 31 ms 32 ms ix-17-0.tcore2.AEQ-Ashburn.as6453.net [216.6.87.149]
8 71 ms 54 ms 65 ms if-2-2.tcore1.AEQ-Ashburn.as6453.net [216.6.87.2]
9 60 ms 55 ms 57 ms 64.86.85.1
10 54 ms 54 ms 57 ms if-10-2.tcore1.TTT-Toronto.as6453.net [64.86.32.33]
11 53 ms 53 ms 54 ms if-9-9.tcore1.TNK-Toronto.as6453.net [64.86.33.25]
12 55 ms 64 ms 55 ms if-7-2.tcore1.W6C-Montreal.as6453.net [66.198.96.61]
13 60 ms 52 ms 54 ms 66.198.96.50
14 55 ms 52 ms 57 ms 192.34.76.2
15 54 ms 57 ms 56 ms 199.91.189.234
16 51 ms 54 ms 56 ms 199.91.189.25
Trace complete.
Just look at that lovely response time from the Summerville gateway (98.25.64.1).... 23ms to 253ms. That's .023 to .253 seconds just at that first hop. Wreaks havoc for DNS lookups sometimes (which my PS3 REALLY does not like). It isn't always like that... just got lucky and caught it on the first trace. I can ping it 100 times and only see 6 or 7 break the 100ms most the time.... but that is the freaking gateway, not a router halfway across the country. Under heavy traffic times, it will even time-out some. It's the first hop on the other side of my modem, and SHOULD be the fastest and most reliable one I get past my local network....but it's worse than the hops to Raleigh, NC and Ashburn, VA,... or even Montreal for that matter..
-
Repeating 90002 errors every 2 hours or so. Connection is fine and no other games are dropping but FFXIV. Is there anything i can do to remedy this? Does anyone even know what 90002 really is?
-
Here is what i get pinging my server (Gilgamesh)
C:\Users\Ruiz>tracert 199.91.189.57
Tracing route to 199.91.189.57 over a maximum of 30 hops
1 3 ms 2 ms 5 ms 192.168.0.1
2 25 ms * 28 ms cpe-70-112-88-1.austin.res.rr.com [70.112.88.1]
3 25 ms 31 ms 41 ms tge2-1.ausutxla02h.texas.rr.com [66.68.5.89]
4 23 ms 29 ms 19 ms tge0-15-0-3.ausutxla01r.texas.rr.com [24.175.42.
234]
5 23 ms 30 ms 28 ms agg22.dllatxl301r.texas.rr.com [24.175.41.46]
6 23 ms 38 ms 38 ms ae-8-0.cr0.dfw10.tbone.rr.com [66.109.6.52]
7 35 ms 38 ms 29 ms 107.14.19.97
8 33 ms 28 ms 39 ms ix-23-0.tcore2.DT8-Dallas.as6453.net [66.110.57.
97]
9 * 78 ms 79 ms if-2-2.tcore1.DT8-Dallas.as6453.net [66.110.56.5
]
10 66 ms 77 ms 74 ms 64.86.78.5
11 63 ms 79 ms 89 ms if-22-2.tcore2.CT8-Chicago.as6453.net [64.86.79.
1]
12 65 ms 69 ms 79 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96
.45]
13 85 ms 89 ms 87 ms 66.198.96.50
14 82 ms 129 ms 79 ms 192.34.76.2
15 75 ms 82 ms 74 ms 199.91.189.234
16 88 ms 89 ms 149 ms 199.91.189.57
Trace complete.
-
Looks like you're suffering a similar fate to us in the Carolinas... gateway isn't as reliable as it should be so when it starts flaking out during peak times you loose the ability to recover from any bad glitches further down the route. So you wind up with lag and packet loss at your gateway compounding what is also coming through the TATA network that takes you into Canada (AS6453 ID's).
The same nightmares are plaguing at least a half dozen key areas in not only the RR/TWC coverage map, but also PSN, Netflix, and others. It's all because our last-mile ISP's (TWC, Comcast, etc.) have oversold their markets and under-maintained their networks and peering arrangements.
Forward some good and bad traces and signal/event logs to their help email address. If you point your browser at either 192.168.0.1 or 192.168.100.1, you should be able to get at some event logs and signal pages that you can copy/paste into emails--helps to print them out for the techs when they visit as well. It might take a few dumps before they actually start taking action, but they can take steps to dramatically reduce it. If you watch things enough, after a while you will start to identify the usual good/bad times so you can plan things a little better to avoid the really bad windows. I'm down to roughly about a 1-2 hour window each night on the weekends where it becomes unmanageable for critical things now.
I've basically gotten in the habit of running some traces before I launch so I can be more sure if it's worth doing the important stuff or I should just fart around instead. Tonight it was bad when I got on, so I killed time for a bit until it felt smoother. Verified some clean traces so I queued Stone Vigil, and aside from a few user errors it ran flawlessly. It isn't a perfect situation by any means, but once you learn your neighborhood's usage patterns you will "know" when to expect it to be decent or not so you can plan a little better.
Hopefully things will stabilize soon over there with the new gigabit fiber plans coming to that region. The competition will hopefully thin out some of the congestion for the current ISP's out there as people switch providers... who knows, you might even find a better deal yourself and kick TWC to the curb. I plan to if we ever get a viable alternative faster than 6MBit DSL.
-
Well after playing around with my modem/router for a bit, i turned off the modem firewall. After that all the 90002's turned back into 90k's. So i get no more 90002's but now get 90k every hour or so. I'm able to log right back in, not having to wait for my character to be logged out to be able to get back in like usual. Still no internet droppage just FFXIV seems to be affected. Dont know if what that means, if anything.
-
hang in there Gofur... there may be some changes for the better coming our way. Haven't gotten confirmation or anything... can't seem to get a straight answer these days, but a long letter I sent to the guys in Herndon and an intense back and forth I had with T3/T4 back at the first of the year may FINALLY be paying off. Basically, was discussing the potential of TWC opening a dialogue with Ormuco about peering with them. Basically, that means instead of hopping on TATA to get from the US into Canada and then get routed to Ormuco (SE's ISP), we go directly from TWC/RR to Ormuco in the US and ride their network into Montreal.
After bouncing my modem a few times last night trying to correct some new geolocation snafu's (were suddenly off by over 100 miles, potentially knackering up route analytics), my routing has made a big change. The new route I'm using via PsiNet/Cogent has an interesting link to Ormuco. It's one of those "3 degrees of separation from Kevin Bacon" kind of things. Ormuco has a massive peering arrangement with PsiNet for connecting to the US, and PsiNet is part of Cogent. Not uncommon for there to be a hand-off to Ormuco through a PsiNet interconnect, but that always happened at the last minute in the past when using Cogent. I was always using hops purely branded to Cogent and was slung all over the US trying to get up there.
Right now, I'm hopping on PsiNet in Atlanta and riding their lines in almost a straight line all the way to Ormuco's door step (there is a quick detour to Chicago for NSA scan, but otherwise it's straight up there in just 4 hops). The entire time it stays on the PSiNet lines. It is possible these lines may be within the Ormuco/PsiNet sphere of influence. Again, haven't been able to confirm that yet... but if it is indeed laying the pathway for TWC peering with Ormuco like they did with JustinTV and Twitch last year, this could be the beginning of a lot of relief for players using TWC (and potentially Comcast depending on what happens with that mess).
Whatever is going on, I am getting a really sweet latency level with this new route. It was rocking a solid 50-60ms during prime time...at times it flirted with numbers in the 40's even. That's nearly half the response times I usually see from South Carolina to my Midgard server in Montreal, Canada (virtually lag free--insane, right?). Hope this is indeed a sign of better things to come from Time Warner.
It may be worth doing some more powering down soon when you can afford the offline time to see if you get new routing as well.
-
Raist, you sir, have gone above and beyond, not only trying to find a solution to this problem hundreds of us have been experiencing, but helping loads of people on these forums understand and give advice to better our situation, and telling us what information to forward to these companies to help it further down the line. For this man, I really have to applaud and thank you. I have sent TWC event logs and tracert's like you advised.
I hope that your new info means hope for us all, and here in the next few mins i will reset, see if anything has changed and let you know. It sounds like you have been on top of this for awhile now, and I really hope your persistence has paid off!
-
Microsoft Windows [Version 6.3.9600]
(c) 2013 Microsoft Corporation. All rights reserved.
C:\Users\Ruiz>tracert 199.91.189.57
Tracing route to 199.91.189.57 over a maximum of 30 hops
1 9 ms 8 ms 9 ms 192.168.0.1
2 22 ms 28 ms 28 ms cpe-70-112-88-1.austin.res.rr.com [70.112.88.1]
3 30 ms 28 ms 28 ms tge2-1.ausutxla02h.texas.rr.com [66.68.5.89]
4 27 ms 28 ms 19 ms tge0-15-0-3.ausutxla01r.texas.rr.com [24.175.42.
234]
5 23 ms 39 ms 38 ms agg22.dllatxl301r.texas.rr.com [24.175.41.46]
6 40 ms 28 ms 28 ms ae-8-0.cr0.dfw10.tbone.rr.com [66.109.6.52]
7 26 ms 28 ms 58 ms 107.14.19.97
8 22 ms 28 ms 29 ms ix-23-0.tcore2.DT8-Dallas.as6453.net [66.110.57.
97]
9 67 ms 66 ms 79 ms if-2-2.tcore1.DT8-Dallas.as6453.net [66.110.56.5
]
10 71 ms 68 ms 80 ms 64.86.78.5
11 69 ms 78 ms 78 ms if-22-2.tcore2.CT8-Chicago.as6453.net [64.86.79.
1]
12 120 ms 78 ms 58 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96
.45]
13 62 ms 86 ms 81 ms 66.198.96.50
14 66 ms 68 ms 68 ms 192.34.76.2
15 91 ms 78 ms 75 ms 199.91.189.234
16 68 ms 68 ms 78 ms 199.91.189.57
Trace complete.
Sadly not much of a change on this end. Not as many spikes and time outs as before. Then again im running this at 2am rather then evening like before so who knows. What struck me as odd was this time I had to turn off my firewall to even get any kind of response. Otherwise it would just time out the whole way down.
-
Microsoft Windows [Version 6.3.9600]
(c) 2013 Microsoft Corporation. All rights reserved.
C:\Users\Ruiz>tracert 199.91.189.57
Tracing route to 199.91.189.57 over a maximum of 30 hops
1 <1 ms <1 ms <1 ms 192.168.1.1
2 4 ms 8 ms 3 ms 192.168.0.1
3 * 28 ms 24 ms cpe-70-112-88-1.austin.res.rr.com [70.112.88.1]
4 19 ms 20 ms 19 ms tge2-1.ausutxla02h.texas.rr.com [66.68.5.89]
5 26 ms 23 ms 28 ms tge0-15-0-2.ausutxla01r.texas.rr.com [24.175.42.
232]
6 24 ms 18 ms 28 ms agg10.ausutxla02r.texas.rr.com [24.175.41.205]
7 26 ms 26 ms 23 ms agg11.ausxtxir01r.texas.rr.com [24.175.41.206]
8 30 ms 33 ms 28 ms agg10.ausxtxir02r.texas.rr.com [24.175.41.209]
9 34 ms 28 ms 28 ms agg22.hstqtxl301r.texas.rr.com [24.175.41.48]
10 34 ms 38 ms 48 ms ae-2-0.cr0.hou30.tbone.rr.com [66.109.6.108]
11 30 ms 37 ms 32 ms ae-0-0.cr0.dfw10.tbone.rr.com [66.109.6.39]
12 29 ms 38 ms 38 ms 107.14.19.97
13 63 ms 69 ms 67 ms ix-23-0.tcore2.DT8-Dallas.as6453.net [66.110.57.
97]
14 72 ms 78 ms 78 ms if-2-2.tcore1.DT8-Dallas.as6453.net [66.110.56.5
]
15 74 ms 80 ms 77 ms 64.86.78.5
16 74 ms 74 ms 78 ms if-22-2.tcore2.CT8-Chicago.as6453.net [64.86.79.
1]
17 103 ms 78 ms 78 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96
.45]
18 73 ms 78 ms 73 ms 66.198.96.50
19 83 ms 68 ms 79 ms 192.34.76.2
20 75 ms 78 ms 78 ms 199.91.189.234
21 81 ms 74 ms 78 ms 199.91.189.57
Trace complete.
So i got some new developments here. A few extra hops compared to traces a few days ago. I also might add that i went a whole 24+ hours w/o a d/c (hope im not jinxing myself here). Any idea what this might mean Raist?
*Edit* - Forgot to add that since the arris DG1670 is a modem/router combo, i turned off all its wireless routing and added a Netgear wireless router between it and my desktop. I did this a few days ago and would still get d/c's, but it did seem to help with the speed.
-
Saw where TWC was looking into some issues in Austin, TX and several spots around Ohio. Some of it was addressing "circular routing" issues that were causing unwanted delays in transit (like how you are stuck in the Austin/Dallas area for a while--some were seeing twice as many hops as you are... was pretty ugly). It's possible that whatever fixes they've been implementing elsewhere are starting to roll out to more markets. They've started rolling out more speed upgrades in select areas as well. With any luck, congestion will gradually improve from here out. Their outage map is slowly improving in the middle of the US at least. Northeast and Southern Cali are still a mess though....but there is progress being made for most other markets.
ooh... just noticed a buggy vlan up in Montreal may finally be out of the picture now too. Was harassing Ormuco and SE about that a while back--was right there at the hand-off into Ormuco's address that starts with 192. It was throwing 200ms+ spikes periodically. Nice to see what you're getting with those segments is staying under 100ms. Looks like it's trying to smoothen out up there.