Do I have a routing issue?

Hello, I hope I'm posting in the correct forum.

My broadband is provided by Pipex and I've been using them since they provided the UKs internet backbone, but things change. My problem is that I am only able to access a veryfew sites and surprisingly this is one of them (but www.thinkbroadband.com is unreachable.) The other sites I can get to are pipex's own ones such as pipexuk.com and their webmail site, netmail.pipex.net. I can also send and receive email for which I use Outlook 2003 as my client.

Clearly, as I can access some websites and my mail client is working my link to the ISP is working. I've also ruled out name resolution issues by using a selection of DNS servers and any valid name you care to mention resolves.

My next step was to try to work out how routing might be working and this is where the clues are. Using tracert commands I can easily find the routes to the pipex sites, mostly in five hops. If, however, I try other addresses such as bbc.co.uk, google.co.uk and so on the trace starts timing out after three hops. Sometimes after a few timeouts another hop succeds followed by more timeouts and occasional successful hops but often the third hop is the last one to succeed. The trace to this forum is three hops followed by 6 timeouts and then this site for example.

On every occasion the third hop is 10.72.4.49 - the timeouts all happen after this device. i think the problem is either with this router or the one it should be handing off to. Can anyone suggest anything that might help me to resolve this issue? Pipex support are supposed to be calling me later this evening but the people I spoke to last night seem unable to deal with anything more complex than DNS issues.

Alistair

thinkbroadband.com was down earlier so you might have been unlucky with the timing.

Is the failure to get to web sites a recent problem? Any chance you are being throttled?

John,

we were away for a bit over a week and returned on Sunday morning to find this situation so it could have stopped working any time in the days before that.

Doesn't throttling take place at the exchange level?

Alistair

I'm using throttling in the sense of the ISP limiting your downloads because you have exceeded some download amount they impose. Two many users sharing a connection from an exchange would normally be termed congestion.

In your case it's seems more likely that there are routing issues which Pipex need to fix.

Very puzzled 10.72.4.49 is a reserved IP and probably a router IP.

Could you please post your tracert to say 85.92.82.57, which I hope will fail, and a tracert to an address that succeeds?

Many thanks

Hatari,

I agree, it has to be a router. I'll try to get the tracerts when I get home this evening.

Alistair

Some ISP's have hops in their network that appear with LAN IP addressing, but should still successfully route stuff around the network

One advantage to an ISP of this is that you cannot directly ping the box I guess

Hatari,

here are three tracerts which show success to a pipex site and two failures, one to the address you gave and one to the BBC. Hopefully this is what you wanted. Funnily enough pipex didn't call last night and there's still no update on the fault call.

Alistair

C:\>tracert netmail.pipex.net

Tracing route to netmail.pipex.net [212.74.100.20]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 37 ms 37 ms 38 ms lo98.sc-acc-sip-2.as9105.net [212.74.102.15]
3 38 ms 37 ms 36 ms 10.72.4.49
4 37 ms 39 ms 38 ms g3-1.switch2-mk.as9105.net [80.40.119.86]
5 39 ms 38 ms 38 ms netmail.pipex.net [212.74.100.20]

Trace complete.

C:\>tracert 85.92.82.57

Tracing route to eta.securesslhost.net [85.92.82.57]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 38 ms 36 ms 37 ms lo98.sc-acc-sip-2.as9105.net [212.74.102.15]
3 36 ms 36 ms 37 ms 10.72.4.49
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 46 ms 45 ms 45 ms host-78-144-0-253.as13285.net [78.144.0.253]
8 * * * Request timed out.
9 46 ms 45 ms 46 ms ae2.core-1.lhc.lon.uk.as29017.net [195.66.224.141]
10 48 ms 47 ms 46 ms ge-1-3-0.core-2.centro.hml.uk.as29017.net [89.145.125.38]
11 46 ms 47 ms 47 ms rt01b.cto-1.hml.as34282.net [89.145.72.70]
12 * * * Request timed out.
13 * * * Request timed out.
14 * * * Request timed out.
<edit for length>
29 * * * Request timed out.
30 * * * Request timed out.

Trace complete.

C:\>tracert bbc.co.uk

Tracing route to bbc.co.uk [212.58.224.138]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 38 ms 36 ms 37 ms lo98.sc-acc-sip-2.as9105.net [212.74.102.15]
3 36 ms 37 ms 37 ms 10.72.4.49
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 45 ms 44 ms 44 ms 212.58.238.133
10 * * * Request timed out.
11 * * * Request timed out.
12 * * * Request timed out.
<edit for length>
29 * * * Request timed out.
30 * * * Request timed out.

Trace complete.

Thank you all for your attention. A nice man from Pipex has just managed to resolve my problem by kicking a router somewhere. My tracerts now take a completely different path as soon as they leave my adsl router.

I wonder if mentioning a MAC request earlier this evening brought a proper engineer to the phone.

Alistair

That's good news. Pity it took such a roundabout route to the solution though!