[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Traceroute
- To: Sean <scase@email7.starnetinc.com>
- Subject: Re: Traceroute
- From: Robert Dinse <nanook@eskimo.com>
- Date: Wed, 22 Sep 1999 12:35:38 -0700 (PDT)
- cc: outages-list@eskimo.com
- In-Reply-To: <v04003a01b40ecce07ecd@[209.224.139.144]>
- Resent-Date: Wed, 22 Sep 1999 12:38:10 -0700 (PDT)
- Resent-From: outages-list@eskimo.com
- Resent-Message-ID: <"nFDy13.0.da.R0Jwt"@mx2>
- Resent-Sender: outages-list-request@eskimo.com
On Wed, 22 Sep 1999, Sean wrote:
>
> Mr.. Dinse,
>
> Judging by the traceroute, everything seems smooth until you get to the
> modem on the other end of your trace.
>
> the 09-016.009.popsite.net is the reverse of one of our assigned addresses.
> Looking at hop 13, it seems the modem could use some tweeking. I would
> like to try some tests with the particular modem that your traced to in the
> list below and see if we can pick up that number.
>
> See if you can call me today or respond with a call back number of which
> directs me to you.
>
> Thanks,
> Sean (StarNet Technical Support Team)
At the time I did this trace yes, but if, during peak hours, you ping each
hop along the way, the last hop in sprint will be less than 2% packet loss, but
the first hop in NapNet will be 40% packet loss. One of your people did the
same in the opposite direction, and it's good up to NapNet and then goes to
hell in Sprint.
There is a peering problem between NapNet and Sprint, part of that problem
is an overloaded DS3 going into the MAE-West peering point from Sprint. They
are installing two OC12's to alleviate this bottleneck, but I would like to
know that once this is in place there won't also be problems on the NapNet end.
Since I am, at present, not a NapNet customer, I need for you to work this
issue from your side and see where they are at with this issue.
You can ping my router port or traceroute to it, at 204.122.16.1 from a
location in NapNet. For reasons that escape me, friday night evening,
8-10pm'ish pacific, but also sometimes bad during monday and tuesday business
days, seem to be the worse times, even though it is not the times of heaviest
traffic for us. Apparently our utilization pattern here does not mirror that
of the Internet in general, or there is some other large customer that eats up
a lot of bandwidth at these times.