IP routing/ AT&T contact info help?

[Merusk]

For the last 5 days I've had problems connecting to my Guild's vent.  It's not a problem with vent (and I'm going to strangle the 5 fucktards who keep saying it is and to enter the ip#.)  but the routing to the server itself.   

A tracert travels to ATT land at 12.122.87.62 then bounces around 4 times before dying.   

9 -> cr1.attga.ip.att.net [12.122.87.62]
10 -> cr2.s19no.ip.att.net [12.122.2.61]
11 -> cr2.kc9mo.ip.att.net [12.122.22.89]
12 -> gar3.kc9mo.ip.att.net [122.123.130.89]
13 -> 12.88.210.110
14 -> Request timed out... Destination host unreachable.

I'm trying to reach 64.254.38.226 but that's what I wind up with.  Since my ISP can't do anything, and nationvoice claims they're good I need some AT&T numbers so I can ask wtf they're doing. (All I find are the web contacts that need an account.)  If anyone else has any ideas on what I can do to resolve this I'd appreciate it. 

[Trippy]

Can you ping the Vent server and what's the IP of that server?

[Merusk]

Nope, can't ping the vent server.  The IP is in the last bit above: 64.254.38.226.

[SurfD]

I'm Located in Toronto.

I did a tracert on that address you said you were trying to reach:  64.254.38.226 got the following

  9    28 ms    28 ms    28 ms  64.212.107.98
 10    63 ms    75 ms    61 ms  cr1.n54ny.ip.att.net [12.122.81.58]
 11    59 ms    59 ms    59 ms  cr2.cgcil.ip.att.net [12.122.1.2]
 12    61 ms    59 ms    61 ms  cr2.sl9mo.ip.att.net [12.122.2.22]
 13    60 ms    59 ms    59 ms  cr2.kc9mo.ip.att.net [12.122.28.89]
 14    61 ms    59 ms    59 ms  gar3.kc9mo.ip.att.net [12.123.130.89]
 15    67 ms    67 ms    71 ms  12.88.210.110
 16  12.88.210.110  reports: Destination host unreachable.

Trace complete.

a ping to the same returned:

Pinging 64.254.38.226 with 32 bytes of data:

Request timed out.
Request timed out.
Request timed out.
Request timed out.

Ping statistics for 64.254.38.226:
    Packets: Sent = 4, Received = 0, Lost = 4 (100% loss),

Assuming other people can connect to that vent, something is definately screwy for anyone on the East coast maybe?

[Merusk]

Yeah the west coast folks have no problems. I'm not sure how many east coasters there are, but I know that I and one other person can't connect at all.  Several others can't do the fatal.nationvoice.com address and have to enter the IP manually.

[Reg]

I'm also in Toronto and get the same results as SurfD.

[sanctuary]

In Western Australia

Tracing route to port064254038226.valnet.net [64.254.38.226]
over a maximum of 30 hops:


  6    79 ms    80 ms    79 ms  i-10-0.sydp-core02.bi.reach.com [202.84.220.149]
  7   228 ms   227 ms   228 ms  i-2-1-1.wil-core02.bx.reach.com [202.84.140.38]
  8   229 ms   229 ms   229 ms  i-1-2.tlot03.bi.reach.com [202.84.251.238]
  9   261 ms   336 ms   423 ms  208.175.201.5
 10   264 ms   263 ms   264 ms  cr2-tengig-0-7-3-0.losangeles.savvis.net [204.70.193.49]
 11   263 ms   263 ms   263 ms  er2-tengig-3-1.lay.savvis.net [204.70.198.10]
 12   264 ms   263 ms   264 ms  208.174.194.66
 13   307 ms   307 ms   307 ms  cr2.la2ca.ip.att.net [12.123.30.134]
 14   304 ms   304 ms   304 ms  cr1.slkut.ip.att.net [12.122.30.29]
 15   308 ms   307 ms   307 ms  cr2.dvmco.ip.att.net [12.122.30.26]
 16   307 ms   306 ms   306 ms  cr1.dvmco.ip.att.net [12.122.31.21]
 17   308 ms   307 ms   307 ms  cr1.kc9mo.ip.att.net [12.122.3.46]
 18   307 ms   307 ms   306 ms  gar3.kc9mo.ip.att.net [12.123.130.37]
 19   315 ms   314 ms   314 ms  12.88.210.110
 20  12.88.210.110  reports: Destination host unreachable.

Trace complete.

[JWIV]

Hrm.  Fun.    Both a west and east coast traceroute off of different networks are coming back with host unreachable.

Probably going to need someone to do a traceroute that's working to see where the path differs (or if the vent host is dual-homed and maybe the people working are using the other pipe).



East Coast Trace off of the AT&T route server
route-server>traceroute 64.254.38.226

Type escape sequence to abort.
Tracing the route to port064254038226.valnet.net (64.254.38.226)

  1 white-dwarf.cbbtier3.att.net (12.0.1.1) [AS 7018] 0 msec 0 msec 0 msec
  2 12.89.5.13 [AS 7018] 4 msec 4 msec 4 msec
  3 cr1.n54ny.ip.att.net (12.123.2.6) [MPLS: Label 16748 Exp 0] 36 msec 36 msec         36 msec
  4 cr2.cgcil.ip.att.net (12.122.1.2) [MPLS: Labels 20486/16021 Exp 0] 36 msec 3        6 msec 36 msec
  5 cr2.sl9mo.ip.att.net (12.122.2.22) [MPLS: Labels 20105/16021 Exp 0] 36 msec         36 msec 36 msec
  6 cr2.kc9mo.ip.att.net (12.122.28.89) [MPLS: Labels 0/16021 Exp 0] 36 msec 36         msec 36 msec
  7 gar3.kc9mo.ip.att.net (12.123.130.89) 36 msec 36 msec 36 msec
  8 12.88.210.110 [AS 7018] 44 msec 44 msec 44 msec
  9  *  !H  *
route-server>

West Coast Trace off the Oregon-IX route server

route-views.oregon-ix.net>traceroute 64.254.38.226

Type escape sequence to abort.
Tracing the route to port064254038226.valnet.net (64.254.38.226)

  1 vl-51.uonet1-gw.uoregon.edu (128.223.51.2) [AS 3582] 0 msec 200 msec 200 msec
  2 vl-3.uonet9-gw.uoregon.edu (128.223.3.9) [AS 3582] 0 msec 0 msec 0 msec
  3 eugn-car1-gw.nero.net (207.98.68.181) [AS 3701] 0 msec 0 msec 0 msec
  4 eugn-core2-gw.nero.net (207.98.64.162) [AS 3701] 0 msec 0 msec 0 msec
  5 ptck-core2-gw.nero.net (207.98.64.10) [AS 3701] 200 msec 200 msec 200 msec
  6 ptck-core1-gw.nero.net (207.98.64.137) [AS 3701] 200 msec 212 msec 200 msec
  7 so-6-1.hsa2.Seattle1.Level3.net (63.211.200.245) [AS 3356] 200 msec 204 msec 200 msec
  8 ge-2-0-0-52.gar1.Seattle1.Level3.net (4.68.105.41) [AS 3356] 200 msec 208 msec 200 msec
  9 att-level3-oc48.seattle1.level3.net (4.68.127.110) [AS 3356] 204 msec 200 msec
    192.205.33.145 [AS 7018] 332 msec
 10 cr2.st6wa.ip.att.net (12.122.146.178) [AS 7018] [MPLS: Label 16852 Exp 0] 200 msec 204 msec 200 msec
 11 cr2.cgcil.ip.att.net (12.122.31.129) [AS 7018] [MPLS: Labels 20458/16021 Exp 0] 204 msec 204 msec 200 msec
 12 cr2.sl9mo.ip.att.net (12.122.2.22) [AS 7018] [MPLS: Labels 20422/16021 Exp 0] 200 msec 200 msec 200 msec
 13 cr2.kc9mo.ip.att.net (12.122.28.89) [AS 7018] [MPLS: Labels 0/16021 Exp 0] 200 msec 64 msec 224 msec
 14 gar3.kc9mo.ip.att.net (12.123.130.89) [AS 7018] 200 msec 208 msec 200 msec
 15 12.88.210.110 [AS 7018] 200 msec 200 msec 200 msec
 16  *  !H  *
route-views.oregon-ix.net>

[Morfiend]

Yeah the west coast folks have no problems. I'm not sure how many east coasters there are, but I know that I and one other person can't connect at all.  Several others can't do the fatal.nationvoice.com address and have to enter the IP manually.

I'm pretty much as west coast as you can get, and I cant hit that IP ether. I am in southern california.

Code:

traceroute to 64.254.38.226 (64.254.38.226), 64 hops max, 40 byte packets
 1  wsip-98-174-154-145 (98.174.154.145)  0.938 ms  0.779 ms  1.032 ms
 2  wsip-64-147-12-233 (64.147.12.233)  81.483 ms  36.431 ms  42.494 ms
 3  wsip-64-58-133-221 (64.58.133.221)  18.897 ms  43.036 ms  57.195 ms
 4  ip68-4-13-245 (68.4.13.245)  69.165 ms ip68-4-13-69 (68.4.13.69)  41.872 ms ip68-4-13-245 (68.4.13.245)  11.597 ms
 5  rsmtdsrj02-ge600.0.rd.oc.cox.net (68.4.14.213)  4.592 ms ip68-4-13-93 (68.4.13.93)  10.670 ms rsmtdsrj02-ge600.0.rd.oc.cox.net (68.4.14.213)  3.315 ms
 6  langbbr01-as0.r2.la.cox.net (68.1.0.230)  4.899 ms  9.012 ms  8.612 ms
 7  ae-1-60.edge1.LosAngeles1.Level3.net (4.69.144.7)  5.239 ms ae-4-90.edge1.LosAngeles1.Level3.net (4.69.144.199)  5.302 ms ae-1-60.edge1.LosAngeles1.Level3.net (4.69.144.7)  23.625 ms
 8  level3-gw.la2ca.ip.att.net (192.205.33.229)  7.422 ms 192.205.33.225 (192.205.33.225)  10.003 ms  5.484 ms
 9  cr2.la2ca.ip.att.net (12.122.129.14)  51.603 ms  51.744 ms  51.939 ms
10  cr1.slkut.ip.att.net (12.122.30.29)  51.309 ms  51.033 ms  119.017 ms
11  cr2.dvmco.ip.att.net (12.122.30.26)  83.871 ms  51.908 ms  111.983 ms
12  cr1.dvmco.ip.att.net (12.122.31.21)  79.028 ms  51.569 ms  110.001 ms
13  cr1.kc9mo.ip.att.net (12.122.3.46)  53.271 ms  51.256 ms  51.365 ms
14  gar3.kc9mo.ip.att.net (12.123.130.37)  51.293 ms  64.914 ms  112.770 ms
15  12.88.210.110 (12.88.210.110)  128.894 ms  58.440 ms  112.642 ms
16  * * *
17  * * *
18  * * *
19  * * *
20  * * *

[Lantyssa]

That doesn't necessarily tell you anything as the server could just be blocking ICMP.

If you can get a reply with traceroute and one of the servers shows the reply time spike, then you have a culprit.  Or if you can get a host some times, but not others.

DSLreports has some constant ping tools you may be able to find a problem with, but it's entirely dependent upon ICMP, which far too many machines block nowadays.

[Ookii]

It's really a moot point, whoever is in charge of 64.254.38.226 has fucked up. Tell the admin to fix his shit.

[Tmon]

Your chances of getting help from AT&T (or any other provider you don't have an account with) are pretty much 0.  They only care about paying customers and then only the big ones.  They especially don't care about individual users trying to push Voice or VPN tunnels through their networks.  If the guy hosting the server is a customer and has the right kind of data plan with them, he might get them to open a ticket and pursue it but it's a long shot.

[Merusk]

Well crap. Meh, thanks folks.

[Chimpy]

Your chances of getting help from AT&T (or any other provider you don't have an account with) are pretty much 0.  They only care about paying customers and then only the big ones.  They especially don't care about individual users trying to push Voice or VPN tunnels through their networks.  If the guy hosting the server is a customer and has the right kind of data plan with them, he might get them to open a ticket and pursue it but it's a long shot.

Shit, my company pays AT&T millions a year for circuits and we can't get them to fix shit in any reasonable amount of time.

And it really sounds like the vent hosting service has hosed something with their upstream provider or their equipment, if people can't get to it off of a canonical name then their DNS records are not propagating correctly either.

OrgName:    Terra World, Inc.
OrgID:      TERW
Address:    200 ARCO Place, Suite 252
City:       Independence
StateProv:  KS
PostalCode: 67301
Country:    US

NetRange:   64.254.32.0 - 64.254.63.255
CIDR:       64.254.32.0/19
NetName:    TERRAWORLD-INC
NetHandle:  NET-64-254-32-0-1
Parent:     NET-64-0-0-0-0
NetType:    Direct Allocation
NameServer: NS1-AUTH.TERRAWORLD.NET
NameServer: NS2-AUTH.TERRAWORLD.NET
Comment:    ADDRESSES WITHIN THIS BLOCK ARE NON-PORTABLE
RegDate:    2001-01-24
Updated:    2001-06-27

RTechHandle: DS1958-ARIN
RTechName:   Schaub, Duane
RTechPhone:  +1-316-332-1616
RTechEmail:  dschaub@terraworld.net

OrgTechHandle: DS1958-ARIN
OrgTechName:   Schaub, Duane
OrgTechPhone:  +1-316-332-1616
OrgTechEmail:  dschaub@terraworld.net


Looks like from everyone's traceroutes, it is dying pretty much where AT&T would hand off to this dude (last 2 registered hops are routers in Kansas City, and the hop after that is an IP that is in the 12.x.x.x range which is AT&T's class A).

www.terraworld.net redirects to http://www.valnet.net/

[fuser]

Shit, my company pays AT&T millions a year for circuits and we can't get them to fix shit in any reasonable amount of time.

And it really sounds like the vent hosting service has hosed something with their upstream provider or their equipment, if people can't get to it off of a canonical name then their DNS records are not propagating correctly either.

I'd say its some DDoS mitigation gone wrong. As everyone else has said start at the Vent provider and work back.

[Trippy]

That doesn't necessarily tell you anything as the server could just be blocking ICMP.

Even if it did make it all the way through it doesn't tell you much cause routing these days for paths of any decent length are almost always asymmetrical (unlike the good old days). In other words the path a packet takes *to* a destination is not always the same path as the packet takes *from* the destination. So the hops you see in a traceroute typically only show one half of the true route and problems along the way, if any, are not necessarily caused at the spot shown in traceroute cause it could be a hop along the return path that's causing the issue.

[Merusk]

Got the vent owner to talk to Nationvoice, since he started having problems too so suddenly he believed me. 

There's using a custom domain name for the vent server, and Nationvoice's custom server hosed its routing somewhere.  I got the 'true name' (vent servers are demons?) of the server from the owner when he sent in the service ticket and was able to connect fine.  They're doing reboots of the name server and our vent server to try and fix it now.  I blame an unnamed/ unknown intern as I always do in these situations.

Thanks folks.

[Lantyssa]

If it's hosted, then yes it's likely it's a daemon.  (I know next to nothing about Vent.)  The name simply points to a specific path/port on the server.  They can then host as many 'servers' as the machine has capacity for.  Or likely a few more.

It's more of a server in the software client-server model sense if that is the case, we just think of it as its own machine since it has a unique URL.

[Merusk]

If it's hosted, then yes it's likely it's a daemon. 

The joke was apparently too fantasy geek for you.  D-E-M-O-N not D-A-E-M-O-N.. unless you were punning on my geek joke about demons and true names.  If that, then well played.

[Lantyssa]

No, I was just too tired and skipping every other word when reading to get it.  Since it would be a daemon process, you questioned if it was a 'demon', and the "true name" might mean a real designator, it kind of went over my head.  Too geeky for my own good.

Though daemons and devils have true names, too, so it still works.