Hi, Bernie.

Do you see anything at the downstream hops? Do they exhibit similar spikes? What do you see in the final destination? Do these latency spikes change to packet loss at any point in your downstream tree?

I suspect that another constant in your connection is the phone line and hardware at the phone company office. Can you share with us who your phone company is, and your physical location (ie: City)? We've heard another report of something similar (although different as well) in the last several weeks, and it'd be interesting to correlate the facts in these two cases to see if there are any similarities we can figure out.

It would be interesting to see some longer pathed instances as well - using a known target (like www.google.com, or even one of our servers) to see what the downstream hops look like.

If you're *only* seeing this at hop 2 (unlikely or you'd not be bringing it up as an issue), make sure you check out our knowledgebase article on the topic of intermediate hops.

- Pete