[info] AT&T Connecticut-Los Angeles routing problem:mad: There's a routing loop affecting AT&T customers in the 18.104.22.168/19 block to Los Angeles via NTT, in my case, this affects my ability to access my accounts on DreamHost. Over 6 calls and about 4 hours on the phone they steadfastly refused to look at the traceroutes, instead running an insane numbers of diagnostics on the DSL; they finally moved us to a new subnet rather than fixing the bloody routing loop. Keeping an eye on the problem, it's now 2 days later and I can still see the routing loop:
traceroute to adsl-108-93-171-254.dsl.wlfrct.sbcglobal.net (22.214.171.124), 30 hops max, 60 byte packets
1 ip-75-119-220-1.dreamhost.com (126.96.36.199) 61.358 ms 61.342 ms 61.333 ms
2 router-0.hq.newdream.net (188.8.131.52) 41.634 ms 41.638 ms 41.630 ms
3 ip-66-33-201-213.dreamhost.com (184.108.40.206) 1.203 ms 1.463 ms 1.464 ms
4 xe-0-6-0-3.r04.lsanca03.us.bb.gin.ntt.net (220.127.116.11) 1.176 ms 1.169 ms 1.424 ms
5 xe-1.sbc.lsanca03.us.bb.gin.ntt.net (18.104.22.168) 1.138 ms 1.112 ms xe-0.sbc.lsanca03.us.bb.gin.ntt.net (22.214.171.124) 0.841 ms
6 ppp-151-164-39-101.rcsntx.swbell.net (126.96.36.199) 2.460 ms 2.007 ms 1.751 ms
7 188.8.131.52 (184.108.40.206) 20.846 ms 20.842 ms 20.833 ms
8 220.127.116.11 (18.104.22.168) 1.680 ms 1.483 ms 1.471 ms
9 22.214.171.124 (126.96.36.199) 20.574 ms 24.468 ms 24.016 ms
10 188.8.131.52 (184.108.40.206) 2.004 ms 1.991 ms 1.972 ms
11 220.127.116.11 (18.104.22.168) 23.923 ms 21.859 ms 21.822 ms
12 22.214.171.124 (126.96.36.199) 1.748 ms 1.795 ms 1.836 ms
13 188.8.131.52 (184.108.40.206) 20.084 ms 25.985 ms 25.973 ms
14 220.127.116.11 (18.104.22.168) 2.101 ms 2.097 ms 1.932 ms
15 22.214.171.124 (126.96.36.199) 23.763 ms 21.772 ms 25.776 ms
16 188.8.131.52 (184.108.40.206) 2.334 ms 2.329 ms 2.301 ms
Everyone makes mistakes, but their refusal to fix this in the face of clear evidence just shows them to be a bunch of twonks. I know customers are idiots but after an hour or two they could at least LOOK at the traceroutes I was offering. No. Loser company.
Hello, not sure if this is relevant or not but - seeing major ping increases all of a sudden hitting game servers all over the country. I have AT&T dsl service in CT, first hop is to MRDNCT.SBCglobal.net (220.127.116.11). Seeing about a 70 ms delay before I get out of the sbcglobal network, that wasn't there before. Sorry if I am off topic.
I'm experiencing the same problem. It's been a number of days since my connection has been functioning correctly. I can't visit any site on Blizzard's network, which is located in LA, among others. I put up a thread in direct support forum and I'm hoping to sort it there.
I've had to deal with their phone support and the outsourced tier 1 support makes my head want to explode in frustration. Not only do they absolutely insist on walking you through every troubleshooting tip you've already done but it's hard to understand them through the accent most of the time.
Then, at least everytime I've dealt with them, instead of forwarding the call to tier 2 support they take a call back number and you get to wait by the phone and hope they call, feeling like a cheap date after a 1 night stand.
reply to Perette
Same issue here in Ellington, CT AT&T DSL. Always had ping in the 20s on the NYC TF2 server I play on and suddenly a few weeks ago my ping is around 70 (according to net_graph in TF2) and still hasn't gone down. Any idea what I/we should do?
Not much experience with this, but here's my tracert to the server.
Tracing route to c-70-42-74-28.internap-nyc.nfoservers.com [18.104.22.168]
over a maximum of 30 hops:
1 1 ms 1 ms 1 ms DD-WRT [192.168.1.1]
2 1 ms 1 ms 1 ms 192.168.0.1
3 50 ms 51 ms 49 ms bras9-l0.mrdnct.sbcglobal.net [22.214.171.124]
4 47 ms 46 ms 47 ms dist1-vlan50.mrdnct.sbcglobal.net [126.96.36.199]
5 46 ms 47 ms 46 ms 188.8.131.52
6 57 ms 56 ms 56 ms ex2-p3-0.eqabva.sbcglobal.net [184.108.40.206]
7 55 ms 56 ms 55 ms gar7.wswdc.ip.att.net [220.127.116.11]
8 58 ms 57 ms 56 ms cr84.wswdc.ip.att.net [18.104.22.168]
9 59 ms 59 ms 59 ms cr2.wswdc.ip.att.net [22.214.171.124]
10 61 ms 59 ms 60 ms cr2.n54ny.ip.att.net [126.96.36.199]
11 56 ms 56 ms 57 ms gar14.n54ny.ip.att.net [188.8.131.52]
12 55 ms 56 ms 57 ms 184.108.40.206
13 57 ms 57 ms 58 ms border2.pc2bbnet2.ext1.nym.pnap.net [220.127.116.11]
14 57 ms 55 ms 55 ms c-70-42-74-28.internap-nyc.nfoservers.com [18.104.22.168]
Ping response time increases would be a different issue; the routing loop kills the traffic, not delays it. It's sort of the internet equivalent of a black hole.
@Jgr9: The differentials times between upstream hops looks okay. The big jump (1ms to 50ms=49ms) is the connection from your local network to the BRAS (probably actually to the DSLAM, but you can't see this with ping), so I'd log into your DSL modem and check the speed it has negotiated, I'm guessing you'll find it's crap. Try power cycling it, if you haven't, see if it improves. Have you noticed any change in phone line quality? (Suddenly scratchy now) If you've added any new phones/answering machines--take them off. Last resort, just take every other phone/answering device/etc off, power cycle the modem, see if it's better. If so, add things back one every few minutes until you identify the problematic device. If all that fails... maybe have SNET check the line.
Sorry, I meant that, as far as I can tell, I seem to have the same problem as Tom, the second post, who happens to live a couple towns away from me and whose connection also currently goes through Meriden, CT. My ping in the game TF2 is now around 70 when it used to always be in the 20s before a few weeks ago.
Power cycling the modem and router didn't help. One of our phones is dying, but it's pretty old. All the others work fine.
I also made a (private) topic in AT&T Direct about possibly having gotten switched from fast to interleaved or something.