We were trying to diagnose an issue we had around 1 PM EDST, and were looking at net flow data. The data indicated a significant change in our traffic patterns, all coming from Akamai address space. The Akamai utilization graphs show a near doubling of retail traffic in the same time period that we had traffic spikes. Does anybody have any idea what caused such a major surge in traffic? http://www.akamai.com/html/technology/nui/retail/charts.html Robert D. Scott Robert@ufl.edu Senior Network Engineer 352-273-0113 Phone CNS - Network Services 352-392-2061 CNS Phone Tree University of Florida 352-392-9440 FAX Florida Lambda Rail 352-294-3571 FLR NOC Gainesville, FL 32611 321-663-0421 Cell 3216630421@messaging.sprintpcs.com
On Oct 4, 2010, at 3:50 PM, Scott, Robert D. wrote:
We were trying to diagnose an issue we had around 1 PM EDST, and were looking at net flow data. The data indicated a significant change in our traffic patterns, all coming from Akamai address space. The Akamai utilization graphs show a near doubling of retail traffic in the same time period that we had traffic spikes. Does anybody have any idea what caused such a major surge in traffic?
http://www.akamai.com/html/technology/nui/retail/charts.html
Akamai is happy to discuss traffic to individual ASes with those ASes. Please be sure to send e-mail from a verifiable address in that AS if you want information about that AS. If you are a peer, the standard peering@akamai.com address works. If you have Akamai boxes on your network, you can open a ticket with the Network Support group at NetSupport-tix@akamai.com. Our 24/7 NOC, noc@akamai.com, can help with emergency problems, such as a congested link. However, you will have to reach one of the other groups for in-depth, historical traffic investigation. Or you can find one of the Akamai people at NANOG. :) As for this specific problem, I haven't an idea what happened. I can tell you Akamai's global traffic at 1300 EDT / 1700 UTC today was actually lower than yesterday's traffic at the same time. Not sure if that means anything, though. -- TTFN, patrick
participants (2)
-
Patrick W. Gilmore
-
Scott, Robert D.