Gaahh server people!!

[dlots]

Just nail thier behinds to the wall if you can prove its 480 byte payloads, the application is not fit for purpose in a WAN deployment.

I have a packet cap, they are doing database post it sends 1 480B packet (a post I assume) waits for the ACK, and repeat 100k times.

They wrote back that they want a per hop analysis of the delay. I wrote back that this wasn't really possable, and that I would need $40K of gear to do that (For GPS clocks) and for someone to install them.

I already explained to them how TCP works and they want to know why it takes so long to get the ACK back

Also it's 2ms round trip time for 30 miles over 2 WAN links (ISP?) (So 60 miles assuming it's not going back to some distribution hub somewhere along the line)... or say a ring somewhere.

Cool fact I learned that the SOL is ~30% slower in fiber, so that's 26% of the delay now.

[icecream-guy]

SOL?

are these fiber or traditional copper circuits,  pure ethernet fiber is much faster than copper or even  copper to CO, then fiber, then back to copper at the other CO.

[dlots]

No clue, our documentation sucks enough that I can't even tell you if they are ISP links or not.  I had to figure out the path via traceroute.