This past few months every now and then I get huge ping spikes on Dota 2 (as high as 2000ms) and I have to reboot Dota to try and fix the issue. It makes the game unplayable, I’m on Ubuntu 16.10 and I was wondering if there are any tools I can use to isolate the issue, if its ISP related I guess there isn’t much I can do.
traceroute [IP address or dns address]
Ex.
luke@Mercury ~> traceroute ddns.net
traceroute to ddns.net (8.23.224.108), 30 hops max, 60 byte packets
1 gateway (192.168.0.1) 0.444 ms 0.442 ms 2.024 ms
2 * * *
3 omitted_for_security 25.109 ms 24.260 ms 25.379 ms
4 omitted_for_security 27.586 ms 28.539 ms 28.831 ms
5 ae-17-17.car2.LosAngeles1.Level3.net (4.69.201.14) 84.150 ms ae-27-27.car2.LosAngeles1.Level3.net (4.69.201.18) 83.372 ms 84.390 ms
6 VITALWERKS.car2.LosAngeles1.Level3.net (4.59.60.174) 84.689 ms 71.420 ms 80.422 ms
7 * * *
8 freedns.no-ip.com (8.23.224.108) 79.126 ms 80.050 ms 79.411 ms
Stuff from your ISP shows the latency going from there network. Your LAN usually has <1 ms of latency so its almost negligible.
I omitted two lines but only one of them was from the ISP, the other was some location not indexed so I left it anonmyous.
I only spend ~25ms of latency within my IPS’s network, but YRMV.
Note: this only helps you get a rough idea of the situation and should not be used as a deliminator, as a traceroute is an echo pings, and some ISP’s either block it entirely (in which case it will time out), or throttle it which skew results.
The most accurate way to test is to set up monitoring from the applications side and see what there results are. You get that in the latency response graph.
Furthermore, not all routes are the same every time. Once you send a packet to the game servers, your next packet will probably take a different path, as the path that data travels through nodes in a networked system varies a lot.
Your ISP maybe doing shady shit during peak hours and throttling different kinds of traffic. But it also possible that the new ‘shortest’ path for your data succumbed to a buffer bloat issue and the node has to discard packets to keep up with traffic.
I’ll try this but basically there’s nothing I can directly do to actually fix the issue other than monitor it I guess.
I’ve actually had the steam app stutter the last few days. I’m questioning if the build in arch is stable anymore and I’m a little worried about it.