[All][Medium] Terrible latency during my playable hours

Samfucious

So, I've been troubleshooting latency issues that only seem to happen during the evenings, approximately 6pm to 10pm PST (2pm to 10pm GMT). These issues started in late October to Early November. After going back and forth with my ISP to see if they can fix the issue, and trying a VPN, tonight I installed Wireshark, reproduced my latency issues with the server and took a capture.

The output is a bit much. But taking the target address and performing tracert gave me this:

C:\Users\gyrep>tracert 51.195.47.134

Tracing route to vps-ba7c30aa.vps.ovh.net [51.195.47.134]
over a maximum of 30 hops:

1 2 ms 1 ms 1 ms modem [192.168.0.1]
2 5 ms 4 ms * tukw-dsl-gw69.tukw.qwest.net [63.231.10.69]
3 11 ms 8 ms 10 ms tukw-agw1.inet.qwest.net [63.226.198.33]
4 * * 4 ms 4.68.38.173
5 20 ms 21 ms 29 ms ae-2-3611.ear4.SanJose1.Level3.net [4.69.211.221]
6 23 ms 22 ms 24 ms sjo-sv5-bb1-a9.ca.us [142.44.208.133]
7 40 ms 45 ms 39 ms be100-1365.lax-la1-bb1-a9.ca.us [198.27.73.104]
8 85 ms 81 ms 89 ms ash-5-a9.va.us [178.32.135.161]
9 * * * Request timed out.
10 181 ms 167 ms 182 ms be102.sbg-g1-nc5.fr.eu [94.23.122.138]
11 183 ms 168 ms 170 ms fra-fr5-sbb1-nc5.de.eu [94.23.122.137]
12 178 ms 172 ms 173 ms be101.fra1-lim1-g1-nc5.de.eu [91.121.215.117]
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.
20 240 ms 295 ms 202 ms vps-ba7c30aa.vps.ovh.net [51.195.47.134]

In a nutshell, though there seems to be a blip in Tukwilla (just South of Seattle), the bulk of my problems seem to be on the European side.

Please, I've been struggling with this for more than a month now, and CAN'T play on the servers at the end of my work day. It's interfering with my participation with this lovely community. And if it doesn't get resolved soon, I'm going to have to find myself somewhere else.

Please help.

Samfucious

In game name is also "Samfucious"

Vaskel

Are you using the -na extension?

Samfucious

Vaskel I usually do, yes. Though I've tried both endpoints. It doesn't make a difference.

anoldshoe

I've had those issues with request timeouts. Its the ISP. I tried getting them to resolve it back at my old house, and they wouldn't recognize that it was their issue, and blame it on the server host. I would test it with other server hosts, I.E. google, with the same issue. This is at least my personal experience. I recommend switching ISP if that is possible

john01dav

If you can run MTR (lookup "winmtr" if you're on Windows, otherwise lookup "mtr"), it can give better data than trace route and this data is sufficient to determine, mostly with certainty, exactly which ISP is to blame (yours, the servers, or some ISP that is involved in the connection between the two). The fact that it happens with both the -na proxy and the main IP, though, makes it most likely that it's your ISP. MTR measures packet loss in addition to latency, so it provides a much more useful data point for each IP. Note that intermediete routers sometimes drop ICMP packets but pass along non-ICMP packets, so if a single node in the MTR output has significant packet loss, but the node after it doesn't, then that packet loss value is a false reading.