Emergency!! Kurt Needs Help!!

Meat_

Banned
Location
Lehi
Yep same as cruiseroutfit exactly. Very strange. The only common denominator that I can think of is that we are both in Utah, and we are both going to micfo.com. Very odd.

I'm in Utah, on Quest dsl and the only problems I've had in recent weeks have been intermittent that I wrote off to Herzog upgrading
Maybe you guys are hitting different routers


Tracing route to rockymountainextreme.com [216.246.1.218]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 3 ms 1 ms 1 ms dslmodem.domain.actdsltmp [192.168.0.1]
3 43 ms 42 ms 41 ms slkc-dsl-gw10-202.slkc.qwest.net [67.41.239.202]
4 42 ms 42 ms 42 ms slkc-agw1.inet.qwest.net [67.41.239.93]
5 56 ms 42 ms 42 ms slk-core-01.inet.qwest.net [205.171.131.21]
6 59 ms 59 ms 58 ms svx-core-02.inet.qwest.net [67.14.1.58]
7 59 ms 58 ms 58 ms sjp-brdr-01.inet.qwest.net [205.171.214.138]
8 73 ms 59 ms 58 ms sjo-bb1-link.telia.net [213.248.86.25]
9 133 ms 91 ms 95 ms chi-bb1-pos6-1-0-0.telia.net [213.248.80.25]
10 91 ms 91 ms 90 ms so2-2-0.j1.ord.scnet.net [216.246.12.93]
11 90 ms 90 ms 91 ms so2-1-0.j1.ord.scnet.net [205.234.158.81]
12 90 ms 96 ms 92 ms unknown.ord.micfo.com [216.246.1.218]

Trace complete.


Kurt, you said you were on Quest? I hate to sound like one of those dumAs help guys but have you power cycled your modem and router recently?

If you are on Quest, using an Actiontec modem and haven't done this you WILL eventually have DNS issues.
 

Rusted

Let's Ride!
Supporting Member
Location
Sandy
I will work on a google route, but here is my tracert to RME from work

1 216.21.254.3 0 msec 0 msec 4 msec
2 216.21.240.5 4 msec 4 msec 4 msec
3 * * *
4 * * *

Just like Kurt it dies leaving my ISP.
 

Rusted

Let's Ride!
Supporting Member
Location
Sandy
Yes but that is a tracert from my internet router that is on the outside of the firewall. That is the route that all the clients behind the firewall would take once they made it to the router.
 

78mitsu

Registered User
best guess this is the problem:
sh ip bgp 67.14.1.58
% Network not in table
All the other subnets are advertised through BGP

and looks like it's supposed to be routing to 205.171.0.244 and 205.171.0.243
not 67.14.1.58 so Best guess is they've got a static route entry that is routing a particluar subnet or set of subnets to 67.14.1.58 that is probabbly dropping them in the bit bucket.
 
But Kurt could access one of the 3 sites that I posted earlier. All 3 are on the same subnet as RME. It would seem a general routing problem would affect all 4 sites the same way. Curiouser and curiouser.
 

Herzog

somewhat damaged
Admin
Location
Wydaho
But Kurt could access one of the 3 sites that I posted earlier. All 3 are on the same subnet as RME. It would seem a general routing problem would affect all 4 sites the same way. Curiouser and curiouser.

The weird thing is... we could only access the same one he could, and not the other two. :eek: :spork:
 

Rusted

Let's Ride!
Supporting Member
Location
Sandy
Yes weird, but at least it's consistent. Can someone with the problem post the routing info from you to rme and the 3 sites I posted?


This is all I have, I have not got a trace from Kurt that I can post yet. This is a post from the internet router to RME

1 216.21.254.3 0 msec 0 msec 4 msec
2 216.21.240.5 4 msec 4 msec 4 msec
3 * * *
4 * * *

thats all folks. Kurt did get further, about 6-8 hops and they when it left the qwest network it was dead.
 

78mitsu

Registered User
It's a BGP Problem, I dug into it last night a little, Looks like Qwest only advertises one path through BGP to the real world, an alternate path is available to the internal network, but not to the rest of the world. They lost a router in their SunnyValley location which functions as a network tandom, so packets going out work fine, but there is a bottleneck for the packets coming back. So depending on the traffic sometimes it will work sometimes it won't. get DNS involved, and you're getting some -caching on different DNS servers causing the problem to appear more consistent. They have 5 OC3's leaving SLC and one OC48, which is 3.2 Gbps. but because of the way their routing is built they only route out to SVC then to Chicago which is an OC48.
 

Rusted

Let's Ride!
Supporting Member
Location
Sandy
I am thinking that it is a BGP issues as well. The qwest half makes sense, but why did a few days later my connection through broadwing go down as well? Any why can I get to 216.246.1.219, but not 218 or 217 or 216? I can get to 220, 221, 222. Seems like a BPG issue would route that whole subnet and I would not be seeing the pattern of only getting to some IP's and not to others.

And if it was a qwest BGP issue how come one customer is blocked but others are getting through? Especially if there was only one route out? I am thinking that it is more of a BGP issue on the web hosting end.
 
Last edited:

78mitsu

Registered User
Qwest re-sells their bandwidth to other providers, So other providers could be affected by a router outage, but Broadwing isn't one of them at last I knew, but they are leasing dark fiber from Qwest and in some cases are using portions of their oc-3s I know they share the SLC->Boise->Portland->Seattle route and the SLC->RockSprings->Laramie->Cheyenne->Denver route and several others, so if it was a fiber cut or a broken terminal the symptomology woudl be the same for both providers, difference is that Broadwind actually knows how to manage their network. I agree it should be a more intermittent failure but, it may be as simple as the service is flapping. I don't beleive it is interweb wide, None of our customers have complained about being able to get to a partiular site on the interweb.
 

Rusted

Let's Ride!
Supporting Member
Location
Sandy
Here is a copy of Kurts trace, looks like he is going through eli.net and not the qwest route. who is scnet.net, and does the "tagged" mean anything. It also hits the 57 IP twice like it is getting bounced around scnet.net.


1 <1 ms <1 ms <1 ms 10.0.0.1
2 55 ms 60 ms 54 ms 67.50.206.1
3 54 ms 53 ms 53 ms fe-4-0-0--0.gw02.slkc.eli.net [209.210.119.17]
4 53 ms 53 ms 52 ms ge-3-1-0--0.cr01.slkc.eli.net [207.173.115.245]
5 52 ms 84 ms 85 ms so-1-0-0--0.cr02.chcg.il.frontiernet.net [74.4.4.238]
6 85 ms 83 ms 85 ms ge-2-0-0--0.br01.chcg.il.frontiernet.net [74.4.4.142]
7 83 ms 85 ms 88 ms ge0-0-1.j2.ord.scnet.net [206.223.119.57]
8 84 ms 85 ms 85 ms ge0-0-1.j2.ord.scnet.net [206.223.119.57]
9 85 ms 85 ms 85 ms so2-1-0.j1.ord.scnet.net [64.202.110.13]
10 84 ms * * tagged.b2.ord.scnet.net [64.202.111.114]
11 * * * Request timed out.
12 * * * Request timed out.
13 * * * Request timed out.
 
Top