Re: NAP/ISP Saturation WAS: Re: Exchanges that matter...
Subject: Re: NAP/ISP Saturation WAS: Re: Exchanges that matter... To: doleary@cisco.com (dave o'leary) Date: Mon, 16 Dec 1996 16:00:09 -0500 (EST) Cc: mleber@he.net, forrestc@iMach.com, nathan@netrail.net, joe@Navigist.Com, nanog@merit.edu From: dvv@sprint.net (Dima Volodin)
More of that - at least one site uses one of our hosts as a target of pings for measuring "Internet health". We consider this to be an abuse of our resources
How many ping per second do you see? How many pings per second can your host handle? Your network?
and, most probably, we will disable any external traffic we find inappropriate for this or other hosts.
So, do I understand you correctly as saying that you have no interest in helping your customers understand the state of your service? -tjs "Those darn customers, things would work just fine, if they didn't keep using the network..."
Tim Salo writes:
How many ping per second do you see?
How many pings per second can your host handle? Your network?
We never stress-tested this host with pings as serving the pings is not the primary reason of existence of this particular host and the LAN it's on. The LAN the host (for historical reasons) is on is a rather busy 10M ethernet, the host itself experiences the UDP load of approx. 500 datagrams/sec, so using it as a ping reflector for any purposes other than checking if it's alive is an absolutely misleading activity.
and, most probably, we will disable any external traffic we find inappropriate for this or other hosts.
So, do I understand you correctly as saying that you have no interest in helping your customers understand the state of your service?
No, I neither have said nor intend to say anything like that.
-tjs
Dima
participants (2)
-
dvv＠sprint.net -
salo＠msc.edu