Hi David,

Tom and I looked at this issue today, and eventually discovered that there was an interfering station that "recently" (>1mo ago) came on the air and was causing the drama on the Tukwila-Baldi link.  We shifted frequencies slightly and the link is feeling much better now.  If you re-run your tests, you should see better results.

The QA-Westin and QA-CP links still need fixing, but they shouldn't affect you since you're routing through Baldi to hit the Internet.

--Bart


On 4/5/2016 10:09 AM, Tom Hayward wrote:
On Tue, Apr 5, 2016 at 9:48 AM, David Giuliani <David@giuliani.org> wrote:
Where can I find out what is located at each 44…. address? Otherwise it’s just a bunch of numbers.  I wonder if you have a detailed system map somewhere on-line?

David,

The Mikrotik traceroute tool has an option called "use-dns=yes" that will resolve the names for each of the IP addresses for you. Here's an example with and without:

[tom@WA6PXX-MercerIs] > /tool traceroute 8.8.8.8
 # ADDRESS                          LOSS SENT    LAST     AVG    BEST
 1 44.24.240.33                       0%    5   6.6ms       6       3
 2 44.24.240.7                        0%    5   4.2ms     4.2     4.1
 3 44.24.242.37                       0%    5   8.3ms     9.7     7.8
 4 44.24.241.82                       0%    5   8.3ms      12     7.8
 5 44.24.242.34                       0%    5  16.9ms    16.3      12
 6 44.24.241.97                       0%    5  11.7ms    15.2    11.7
 7 209.189.196.65                     0%    5    20ms      18    12.3
 8 209.189.202.199                    0%    5  20.6ms    18.2    11.7
 9 206.81.80.17                       0%    4  21.3ms    17.5    12.6
10 74.125.37.91                       0%    4  16.8ms    15.2    11.7
11 209.85.249.245                     0%    4  23.5ms      19    11.8
12 8.8.8.8                            0%    4  19.9ms    16.1      12

[tom@WA6PXX-MercerIs] > /tool traceroute 8.8.8.8 use-dns=yes
 # ADDRESS                          LOSS SENT    LAST     AVG    BEST
 1 wlan1.capitolpark-s2.hamwan.net    0%    8     5ms     7.7     3.7
 2 ether1.queenanne.capitolpark....   0%    8   4.1ms     6.6       4
 3 wlan1-cppaine.capitolpark.que...   0%    8  12.6ms    12.2     7.8
 4 ether1.seattle.queenanne.hamw...   0%    8  12.2ms    11.8     8.4
 5 wlan1.queenanne.seattle.hamwa...   0%    8  12.4ms    18.3      12
 6 switch.er1.seattle.hamwan.net      0%    8  12.4ms    16.9      12
 7 209.189.196.65                     0%    8  11.9ms    21.7    11.9
 8 ge0-0-0-1-thbr06.threshinc.net     0%    8  12.4ms    17.3      12
 9 six.sea01.google.com               0%    8  16.7ms    17.4    12.5
10 74.125.37.91                       0%    8    16ms    16.7    12.3
11 209.85.249.245                     0%    8  16.6ms    18.3    12.3
12 google-public-dns-a.google.com     0%    8  12.4ms    17.4    11.9


The naming convention we try to follow is INTERFACE.TASK.LOCATION.hamwan.net. For example, the second hop from you is ether1.queenanne.capitolpark.hamwan.net. That is a router at the Capitol Park site pointed at Queen Anne, and you're getting a response from its ethernet port.

The physical locations of the sites can be found on the map: https://hamwan.org/t/Coverage+Map

A full list of HamWAN routers can be found at http://portal.hamwan.org/host/all

Tom KD7LXL