Is the issue outside the network?

dccsr2007

Honorable
Dec 3, 2012
1
0
10,510
Humdinger here if I've ever crossed one.

A short track of the network in place:

Modem - bridged into Sonicwall
Sonicwall - DHCP to Switch
Switch - Servers / client pc's
VPN piggybacked for direct connection to offsite accounting system

Sorry for non-detailed information as I'm only part time in the office and have some technical know how, but limited.

The issue:

Slow, crawling, unresponsive internet. Constant time-out errors when attempting to connect to any website. Accounting software (on VPN) slow as well.
ISP claims to have no issue.

What I have done?

Traced lines from network closet to locations across building - Good returns on my line diagnostic tool
Ping test from PC to PC, PC to Server, PC to Gateway across network - Good echos
Tracert from PC to PC and PC to Server and PC to Gateway - Good echos

Tracert from PC to Google, MSN, Yahoo, etc. - Here's where latency problems arise.

The first hop times out (IMCP? blocking echo request) - Gateway HOP receives good echo - 3 ISP Servers (small town ISP - DSL) 90% good echos (some latency in the 1200 range) - Next hops are to ATT servers, constant 2800+ latency on all 3 returns of all 3 servers - Rest of hops to servers are fine, no issue.

After two weeks of bickering, ISP switches the modem. Issue clears for roughly a day and a half, two days or so. Issue creeps back in.

Run Tracert - Issue persists at ATT servers.

My question:

Is this the network hardware or is there further testing that can be done to further rule out the network hardware on our side of the modem?
I have consulted with a few friends that have more knowledge than I, and both have said that I have done what they would to rule out our office hardware.

Thank you in advance for any and all help given. This issue has persisted for almost three weeks now, with no end in sight.
 
yeah, I'd say try direct connect to the modem with a single machine, if the issue goes away it's something on your system. But judging from traces, it's on the ATT servers

you can try flushing dns, but I'm doubtful that will solve the issue