see for yourself


Microsoft Windows XP [Version 5.1.2600]
(C) Copyright 1985-2001 Microsoft Corp.

C:\Documents and Settings\jason>ping httpd.net

Pinging httpd.net [216.180.171.215] with 32 bytes of data:

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

Ping statistics for 216.180.171.215:
Packets: Sent = 4, Received = 0, Lost = 4 (100% loss),

C:\Documents and Settings\jason>ping foonet.net

Pinging foonet.net [216.207.29.74] with 32 bytes of data:

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

Ping statistics for 216.207.29.74:
Packets: Sent = 4, Received = 0, Lost = 4 (100% loss),

C:\Documents and Settings\jason>ping snchosting.com
Ping request could not find host snchosting.com. Please check the name and try a
gain.

C:\Documents and Settings\jason>ping snchosting.net
Ping request could not find host snchosting.net. Please check the name and try a
gain.

C:\Documents and Settings\jason>ping monster-hosting.com
Ping request could not find host monster-hosting.com. Please check the name and
try again.

C:\Documents and Settings\jason>

HUH? What happened to that when you go down its only for 2-10 mins, its been over 20! No support, No emails Phone lines busy. Anyone else get through ?

There's tons of support, and we know they're not going any where. I'm sure something has happened that is out of their hands.

It looks like the upstream providor is messed up. No phones or anything telco related in the town where foonet is, i guess they are working as hard as they can to get it resolved. Not foonet's fault!

Originally posted by sysc
HUH? What happened to that when you go down its only for 2-10 mins, its been over 20! No support, No emails Phone lines busy. Anyone else get through ?

Do a traceroute,

Qwest is the culprit.

-Robert

I thought they had more than one provider?

Shouldn't they be working at least?

Originally posted by BlackAng
I thought they had more than one provider?

Shouldn't they be working at least?

It's called BGP and routing tables. Those don't change on the fly.

Carl

**UPDATE (from paul over cell phone):

All telco running into the entire city is down. all voice/data lines are down. Numerous trouble tickets open/active with telco facilities and providers.

will keep informed as it progresses...


-Robert

Sorry if i'm wrong here, but BGP doesn't that already load balance the transfers between multiple pipes? So I'm assuming if one pipe is closed, and the other open, it shouldn't take that long to figure out which pipe to send the data down?

Anyways, I was just asking a question, I don't need to hear your smug remarks.

Thanks for coming out Carl!

Originally posted by BlackAng
Sorry if i'm wrong here, but BGP doesn't that already load balance the transfers between multiple pipes? So I'm assuming if one pipe is closed, and the other open, it shouldn't take that long to figure out which pipe to send the data down?

Anyways, I was just asking a question, I don't need to hear your smug remarks.

Thanks for coming out Carl!

No, basically the Net is a collection of 'routing paths'. If one circuit between those paths drops, changes need to be made and propagated along a great deal of the network. This is what may take a while.
Of course, things are much much more complicated than this, but it's a simple summary.

Carl

I'm not flaming foonet I love them from the bottom of my heart so don't take this the wrong way.

However, Carl, we aren't talking about some Trans-Atlantic fibre that's been disconnected, which would require sending new gateway information all the way to Qwest's head office. We are talking about a data center. Hence the thread topic.

Thanks once again for your most elightening comments.

BlackAng,
Wherever the problem lies, you need to see it on a worldwide scale. Routing tables need to be propagated to a great deal of peers, wherever the outage may occur.
In this case, there simply isn't a workaround, since the complete city seems to be cut off, including all providers.

Carl

FOONET is still the best for shell hosting.

It's not their fault they only have a few T1 lines... or wait? Am I living in the past? Have they actually upgraded?

From the looks of it, although they may be multi-homed, they probably only have one fiber provider. If that's the telco, and everything else is down, chances are it's a problem with the telco.

I believe that they've upgraded a lot from a few T1s, don't know specifics though.

- Matt

Originally posted by BlackAng
Sorry if i'm wrong here, but BGP doesn't that already load balance the transfers between multiple pipes? So I'm assuming if one pipe is closed, and the other open, it shouldn't take that long to figure out which pipe to send the data down?

Anyways, I was just asking a question, I don't need to hear your smug remarks.

Thanks for coming out Carl!

It's called a "BGP Flap" typically when a route drops and you "flap" between the two. Usually it takes up to 2 minutes (usually around 30 - 60 seconds) for the route tables to update. This should not ever take more then 5 minutes, heck, ideally it should (almost) always be under 1 minute. Typically this only applies to incoming routes.

We can speculate for hours on this, but nobody can offer a real answer since Foonet/CIT hides their super top secret network configuration. Otherwise, we might be able to help them!

Fiber Lines were cut. They are back now.