one route to employer's network taking a scenic route
Posted: Sat Mar 07, 2020 1:52 pm
So I've been troubleshooting a performance issue with one model of my employer's teleworker vpn devices (its likely affecting more than just these devices though). This device's controllers are located in our inbound dmz. The other model we use goes through our outbound dmz.
The problem seems to be that traffic from my house to our inbound dmz is being routed through Australia by Cogent. Why is this a problem? Because I'm in San Francisco and my employer's data center is in Sacramento. There is no reason for this traffic to be routed out of country, let alone freaking Australia (nothing against Aussies... its just a bit out of the way).
I emailed Cogent and asked them about the latency jump and they said that is normal for the hop to Australia. I then asked why is traffic between SF and Sacramento going through Australia, but I think they already expended as much effort as they are willing to exert on my problem.
Is there anything that can be done by Sonic to change the route so its not handed off to Cogent for a scenic sunday drive?
This is the bad route:
9 4 ms 4 ms 4 ms 100.ae1.nrd1.equinix-sj.sonic.net [75.101.33.185]
10 4 ms 4 ms 4 ms 61.ae1.nrd1.pao1.sonic.net [157.131.209.178]
11 4 ms 4 ms 5 ms hu0-3-0-2.ccr31.sjc04.atlas.cogentco.com [38.104.141.81]
12 6 ms 13 ms 13 ms be2016.ccr22.sfo01.atlas.cogentco.com [154.54.0.177]
13 21 ms 21 ms 23 ms be3694.ccr21.pdx01.atlas.cogentco.com [154.54.84.30]
14 21 ms 22 ms 21 ms be2216.ccr51.pdx02.atlas.cogentco.com [154.54.31.158] <--- US up to this point
15 154 ms 155 ms 156 ms be2237.ccr51.syd01.atlas.cogentco.com [154.54.45.122] <-- AUS
16 166 ms 166 ms 165 ms level3.syd01.atlas.cogentco.com [154.54.64.2] <--- AUS
17 222 ms 223 ms 222 ms 4.69.218.102 <--- US
This route to our outbound dmz works fine:
9 58 ms 39 ms 52 ms 0.ae0.cr1.equinix-sj.sonic.net [75.101.36.253]
10 5 ms 5 ms 119 ms 100.ae1.nrd1.equinix-sj.sonic.net [75.101.33.185]
11 4 ms 4 ms 5 ms ae17.cr6-sjc1.ip4.gtt.net [69.22.130.109]
12 5 ms 5 ms 4 ms et-0-0-13.cr5-sjc1.ip4.gtt.net [89.149.180.26]
13 11 ms 9 ms 5 ms 63-235-42-1.dia.static.qwest.net [63.235.42.1]
14 5 ms 9 ms 13 ms svl-edge-24.inet.qwest.net [67.14.43.26]
15 14 ms 8 ms 11 ms 65.113.42.90
16 * * * Request timed out. <-- none of the hops respond to tracert from here on
The problem seems to be that traffic from my house to our inbound dmz is being routed through Australia by Cogent. Why is this a problem? Because I'm in San Francisco and my employer's data center is in Sacramento. There is no reason for this traffic to be routed out of country, let alone freaking Australia (nothing against Aussies... its just a bit out of the way).
I emailed Cogent and asked them about the latency jump and they said that is normal for the hop to Australia. I then asked why is traffic between SF and Sacramento going through Australia, but I think they already expended as much effort as they are willing to exert on my problem.
Is there anything that can be done by Sonic to change the route so its not handed off to Cogent for a scenic sunday drive?
This is the bad route:
9 4 ms 4 ms 4 ms 100.ae1.nrd1.equinix-sj.sonic.net [75.101.33.185]
10 4 ms 4 ms 4 ms 61.ae1.nrd1.pao1.sonic.net [157.131.209.178]
11 4 ms 4 ms 5 ms hu0-3-0-2.ccr31.sjc04.atlas.cogentco.com [38.104.141.81]
12 6 ms 13 ms 13 ms be2016.ccr22.sfo01.atlas.cogentco.com [154.54.0.177]
13 21 ms 21 ms 23 ms be3694.ccr21.pdx01.atlas.cogentco.com [154.54.84.30]
14 21 ms 22 ms 21 ms be2216.ccr51.pdx02.atlas.cogentco.com [154.54.31.158] <--- US up to this point
15 154 ms 155 ms 156 ms be2237.ccr51.syd01.atlas.cogentco.com [154.54.45.122] <-- AUS
16 166 ms 166 ms 165 ms level3.syd01.atlas.cogentco.com [154.54.64.2] <--- AUS
17 222 ms 223 ms 222 ms 4.69.218.102 <--- US
This route to our outbound dmz works fine:
9 58 ms 39 ms 52 ms 0.ae0.cr1.equinix-sj.sonic.net [75.101.36.253]
10 5 ms 5 ms 119 ms 100.ae1.nrd1.equinix-sj.sonic.net [75.101.33.185]
11 4 ms 4 ms 5 ms ae17.cr6-sjc1.ip4.gtt.net [69.22.130.109]
12 5 ms 5 ms 4 ms et-0-0-13.cr5-sjc1.ip4.gtt.net [89.149.180.26]
13 11 ms 9 ms 5 ms 63-235-42-1.dia.static.qwest.net [63.235.42.1]
14 5 ms 9 ms 13 ms svl-edge-24.inet.qwest.net [67.14.43.26]
15 14 ms 8 ms 11 ms 65.113.42.90
16 * * * Request timed out. <-- none of the hops respond to tracert from here on