Odd intermittency problems, crazy Speedtest results

[AshleyG]

Since the mid-December snow/ice storms that hit East TN, something odd has been happening with my home network. I have had multiple occasions where accessing websites will grind to a halt, file transfers will stop, and games will hitch and jerk as if there has been a huge lag spike. This has happened on both of the family laptops as well as the desktop. We are currently AT&T DSL customers with the 6 Mbps "Extreme" (hah!) package because there is no non-satellite alternative.

One of the more interesting / confusing things in the midst of this has been the results I've gotten on Speedtest.net. Below is one screen grab from a test this afternoon; I forgot to grab the link for the test that came up at 122.5 Mb/s downstream.

When file transfers don't give out, there will be a period of no movement followed by a sudden download completion. That's the way it worked with SpeedTest as well - the dial sat at 0, then spiked into the red zone and the test finished with an impossibly high result. It is affecting my wife's ability to work from home for her business and is annoying me and my guild on game night.

Tests from computer to computer and among other devices show no internal issue, although power-cycling the router does seem to help just a little bit. At no point are there LAN disconnects or errors. I'm listing the network topography below. All computer systems are running unmodified Windows 7 installations. This network had been running without reset or reconfiguration for more than 4 months.

Thanks for your help and advice in diagnosing this horribly annoying situation.

Phone line --> DLink DSL Modem (bridge mode) --> WNDR3700 (1.0.4.68 firmware)

WNDR3700 -wired-> Desktop (Intel Gigabit CT PCIe NIC), DLink Vonage adapter
WNDR3700 -wireless-> HP laptop, Sager laptop, ASUS NT13 router configured as repeater

ASUS NT13 - wired-> LG BD560, DirecTV HD/DVR

 

[gimp]

issue occurs on both wired and wireless devices, correct?

have you done a trace route during the time when you're having the issue?
or tried doing a constant ping to a known-good IP/domain? ie, google.com

 

[AshleyG]

Thanks for the suggestion. I feel like a fool having not thought to do a tracert already. Anyway, here's the log. :

Tracing route to google.com [74.125.65.105]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  READYSHARE [192.168.1.1]
  2     9 ms     9 ms     9 ms  adsl-86-28-1.tys.bellsouth.net [98.86.28.1]
  3    16 ms    16 ms    16 ms  12.81.60.62
  4    16 ms    16 ms    16 ms  12.81.60.58
  5    16 ms    15 ms    16 ms  12.81.60.18
  6    16 ms    15 ms    16 ms  ixc01cha-pos-7-0-0.bellsouth.net [65.83.239.103]

  7    16 ms    16 ms    16 ms  12.81.101.4
  8    16 ms    26 ms    16 ms  12.81.32.92
  9    16 ms    15 ms    16 ms  12.81.32.101
 10    16 ms    16 ms    15 ms  74.175.192.150
 11    25 ms    24 ms    24 ms  cr1.nsvtn.ip.att.net [12.122.148.14]
 12    24 ms    23 ms    23 ms  cr2.attga.ip.att.net [12.122.28.105]
 13    23 ms    23 ms    23 ms  12.123.22.129
 14    24 ms    23 ms    23 ms  12.88.97.6
 15    25 ms    25 ms    24 ms  72.14.233.54
 16    29 ms    24 ms    24 ms  72.14.239.127
 17    24 ms    35 ms    35 ms  209.85.253.209
 18    24 ms    24 ms    24 ms  gx-in-f105.1e100.net [74.125.65.105]

Trace complete.

Several of the 12.*.*.* IPs hung for 4 or 5 seconds despite the low response time. The numbers would appear, then I'd wait and wait and wait for the server name/IP address.

edit: Five minutes later, running a tracert to bellsouth.net....

Tracing route to bellsouth.net [216.77.188.73]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  READYSHARE [192.168.1.1]
  2     9 ms     9 ms    10 ms  adsl-86-28-1.tys.bellsouth.net [98.86.28.1]
  3    16 ms    16 ms    16 ms  12.81.60.62
  4    16 ms    16 ms    16 ms  12.81.60.58
  5    16 ms    16 ms    16 ms  12.81.60.20
  6    16 ms    17 ms    18 ms  ixc01cha-pos-7-0-0.bellsouth.net [65.83.239.103]

  7    17 ms    16 ms    17 ms  12.81.104.66
  8    17 ms    16 ms    17 ms  12.81.16.10
  9    22 ms    32 ms    17 ms  12.83.2.209
 10    16 ms    16 ms    16 ms  205.152.152.90
 11     *        *        *     Request timed out.
 12     *        *        *     Request timed out.
 13     *        *        *     Request timed out.
 14     *        *        *     Request timed out.
 15     *        *        *     Request timed out.
 16     *        *        *     Request timed out.
 17     *        *        *     Request timed out.
 18     *        *        *     Request timed out.
 19     *        *        *     Request timed out.
 20     *        *        *     Request timed out.
 21     *        *        *     Request timed out.
 22     *        *        *     Request timed out.
 23     *        *        *     Request timed out.
 24     *        *        *     Request timed out.
 25     *        *        *     Request timed out.
 26     *        *        *     Request timed out.
 27     *        *        *     Request timed out.
 28     *        *        *     Request timed out.
 29     *        *        *     Request timed out.
 30     *        *        *     Request timed out.

Trace complete.

It looks like this might be on AT&T's end, ergh. If that is the case it will probably not be fixed for a while.

 

[gimp]

Several of the 12.*.*.* IPs hung for 4 or 5 seconds despite the low response time. The numbers would appear, then I'd wait and wait and wait for the server name/IP address.

if it was "stalling" on displaying the IP/hostname, then there was no issue with that hop. Traceroute is just trying to resolve the IP to a hostname, when there is no hostname associated with the IP. It can just take a bit for that to timeout.

and yeah, it definitely looks to be an issue on AT&T's end. You can always call and complain to your ISP, and hope enough other people have that something will get done.

 

[AshleyG]

Well, this has proved fruitless.

I started with an AT&T chat session. That person was helpful, knowledgeable, and listened. I copied/pasted several tracert logs and provided what other information she asked for. She ran what diagnostics she could and told me that there appeared to be a problem on their end and elevated me up the chain so I could get the ball rolling on getting them to fix it. That's where it went south.

The next person I spoke to couldn't give a rat's ass what I knew. They immediately told me that it was my "non-AT&T hardware" that was causing the issue and would I please pay for ConnectTech support and a new AT&T dsl modem/router combo in order to resolve the issue. My second phone call was just as pointless.

Ergh. Time to beg Charter to run a line across the street.