Why is there a sudden increase in ping latency in Seattle, WA, USA?

March 3, 2018 806 views
Networking Deployment Debian

I am in Portland, OR. Our server app.kartbites.com is in the sgp1 region. The traceroute shows a sudden increase in latency in the middle, but its somehow between two Seattle servers. Currently this is probably the cause of my very slow SSH connection to our server, but I am concerned more about our services that will be going into production soon. Would someone be kind enough to help interpret/fix/shed some light on why we are seeing these high latencies and what can be done about it?

The delay is between two Tata Communications routers in Seattle. There is no support on their page so I haven't been able to contact them. I can't tell if Tata is competing with DO or if DO is leasing data center resources from Tata Communications or if these just happen to be core routers on the path to Singapore. Let's just hope this isn't a competition-related issue...

Admins-MBP:go_server daku$ traceroute app.kartbites.com
traceroute to app.kartbites.com (159.89.209.0), 64 hops max, 52 byte packets
 1  archer (192.168.0.1)  2.326 ms  0.756 ms  0.658 ms
 2  96.120.60.85 (96.120.60.85)  11.498 ms  11.728 ms  12.914 ms
 3  ae-226-rur01.beaverton.or.bverton.comcast.net (68.85.151.225)  11.377 ms  11.810 ms  12.783 ms
 4  ae-2-rur02.beaverton.or.bverton.comcast.net (68.85.243.154)  11.398 ms  16.325 ms  13.921 ms
 5  ae-51-ar01.troutdale.or.bverton.comcast.net (68.87.216.105)  26.222 ms  13.969 ms  47.710 ms
 6  be-33490-cr01.seattle.wa.ibone.comcast.net (68.86.92.217)  19.791 ms  22.638 ms  19.892 ms
 7  be-10846-pe01.seattle.wa.ibone.comcast.net (68.86.86.90)  18.925 ms  16.956 ms  16.426 ms
 8  ix-ae-13-0.tcore2.00s-seattle.as6453.net (64.86.124.20)  15.247 ms  14.554 ms  16.008 ms
 9  if-ae-2-2.tcore1.00s-seattle.as6453.net (64.86.123.92)  260.701 ms  315.418 ms  309.192 ms
10  if-ae-20-2.tcore1.sv1-santa-clara.as6453.net (64.86.123.95)  305.320 ms  213.147 ms  210.129 ms
11  if-ae-8-2.tcore1.lvw-los-angeles.as6453.net (66.110.59.8)  212.224 ms  214.081 ms  212.717 ms
12  if-ae-2-2.tcore2.lvw-los-angeles.as6453.net (66.110.59.2)  211.690 ms  220.837 ms  215.563 ms
13  if-ae-7-2.tcore2.svw-singapore.as6453.net (180.87.15.25)  300.878 ms  307.085 ms  307.152 ms
14  if-ae-2-2.tcore1.svw-singapore.as6453.net (180.87.12.1)  307.258 ms  305.076 ms  307.428 ms
15  if-ae-11-2.thar1.svq-singapore.as6453.net (180.87.98.37)  210.802 ms  211.171 ms  300.048 ms
16  120.29.214.50 (120.29.214.50)  307.852 ms
    120.29.214.142 (120.29.214.142)  217.987 ms
    120.29.214.50 (120.29.214.50)  227.503 ms
17  * * *
18  * * *
19  * * *
20  * * *
21  * * *
22  * * *
23  * * *
24  * * *
1 Answer

It would appear that the slowdown you are seeing is in core routers included in the route between your ISP in seattle and our data center in Singapore. Without checking with our networking team directly, between your traceroute results and my own (Through Mediacom's network in the Midwest) it would appear that your Comcast connection as well as our data center are peering with Tata on this route. I would recommend opening a ticket with our support team so they can escalate this issue to our network engineers. I would also recommend reporting this issue to Comcast as they are directly peering with Tata as well.

Have another answer? Share your knowledge.