FYI: Network Routing Experiment in Progress

This is something that was discussed on the EME Discord today, but in response to people having a bunch of routing problems with Zayo particularly in the last few weeks, they are experimenting with turning off Zayo at the moment, which is routing most people to Savvis again. If your latency is significantly worse as a result of the change (note: not talking about framerate), they are looking for feedback during this test -- especially during "prime time" (from about 3pm to 7pm Pacific, 6pm to 10pm Eastern).

How to do a traceroute on Windows:
Start -> type "cmd" -> press Enter -> type "tracert -d diag.enmasse.com" -> press Enter -> when complete, press Ctrl+M and highlight the output of the command -> right-click on what you highlighted -> Ctrl+V to paste it in a message or DM.

(You don't need to type the quotes in the commands listed above. You can remove the first few hops from the output.)

You can send your results to either me or @Rox via PM through Discord, or post it here. We're collecting all the data we get into a giant file to correlate where the bad routes are coming from. Obviously no guarantee that companies like Zayo or Savvis will listen, but we're at least trying to collect enough data to make a convincing case of where the problems are.
«13

Comments

  • ElinLoveElinLove ✭✭✭✭✭
    Well, even tho I just can't play at all until like mid December (thanks uni...), I can still contribute a bit.
    Test ran just now:

    Rastreando a rota para diag.enmasse.com [208.67.49.36]
    com no máximo 30 saltos:

    1 3 ms 1 ms 1 ms 192.168.0.1
    2 11 ms 10 ms 9 ms 10.111.192.1
    3 16 ms 25 ms 14 ms 201.21.192.1
    4 12 ms 17 ms 13 ms 201.21.192.10
    5 13 ms 12 ms 11 ms 201.21.192.1
    6 26 ms 21 ms 22 ms 189.86.58.69
    7 24 ms 21 ms 29 ms 200.244.214.95
    8 146 ms 160 ms 149 ms 200.230.230.28
    9 * * * Esgotado o tempo limite do pedido. <-maximum request time surpassed
    10 * * * Esgotado o tempo limite do pedido.
    11 207 ms 220 ms 174 ms 4.68.74.158
    12 * * * Esgotado o tempo limite do pedido.
    13 168 ms 170 ms 165 ms 216.33.94.118
    14 170 ms 172 ms 174 ms 64.37.205.6
    15 180 ms 180 ms 179 ms 208.67.48.14
    16 169 ms 172 ms 170 ms 208.67.48.14
    17 165 ms 167 ms 171 ms 208.67.49.36

    From Porto Alegre, Rio Grande do Sul, Brazil.
    My normal game ping was 180ms, jumping occasionally to 210 tops, never dipping much bellow 180ms (I think never)
  • DL7MMWLJ3WDL7MMWLJ3W ✭✭✭✭
    Here we go again with this.

    I remember back when Ruinous Manor first came out this issue happened that lasted half a year with a pinned topic that gotten nowhere but 100+ pages of worthless data.

    I wouldn't get anyone's hope with this. Best way is to deal with it, wait for it to clear, or not play and enjoy being with family/play other games.
  • erwin11erwin11 ✭✭
    edited November 2018
    zayo on: 160 ms
    zayo off: 60 ms

    west coast
  • Since i've changed ISP my ping has been muuuch lower and stable, don't even need VPN anymore, getting 150-155ms with Zayo, now i'm at 140-145ms from Curitiba, Brazil. Great ping and stability considering the distance, old VPN was 200-500 tho lmao

    1 <1 ms <1 ms <1 ms 192.168.100.1
    2 2 ms 2 ms 2 ms 100.127.252.74
    3 * * * Esgotado o tempo limite do pedido.
    4 3 ms 2 ms 2 ms 200.150.94.65
    5 7 ms 8 ms 7 ms 208.178.245.65
    6 113 ms 113 ms 113 ms 67.17.99.233
    7 * * * Esgotado o tempo limite do pedido.
    8 * * 113 ms 4.69.207.29
    9 113 ms 112 ms 112 ms 63.146.26.137
    10 141 ms 142 ms 142 ms 67.14.8.190
    11 143 ms 142 ms 142 ms 216.33.94.118
    12 143 ms 143 ms 143 ms 64.37.205.6
    13 143 ms 142 ms 143 ms 208.67.48.14
    14 144 ms 144 ms 144 ms 208.67.48.14
    15 143 ms 144 ms 143 ms 208.67.49.36
  • KillerPenguinsKillerPenguins ✭✭✭✭
    edited November 2018
    As I stated about 400 times last time they tried savvis, my ping is a joke through them, because it adds 1600+ miles to my route. I suspect this is the case for anyone carried on cogent who lives closer to chicago than dallas or atlanta, since cogent does not seem to peer effectively with savvis anywhere near chicago. The problem is even apparent on cogent's own looking glass - the connection jumps to atlanta and back if you test from their chicago router. For me, this effectively triples my ping at a minimum, much worse at prime time. On zayo my ping was 15-20, with savvis it's a consistent 50 at off hours and spikes to 100+ during peak hours. I'm certain this is a problem with carriers other than cogent as well, but i can't be bothered to do EME's research for them again, feel free to check through my past posts if you want 60 pages of rants about how terrible savvis and tera's datacenter both are.
    The real solution is for tera to get a quality isp in a quality data center instead of using centurylink in 2018...
  • sanj66sanj66 ✭✭✭✭
    edited November 2018
    what does it matter to do a trace route eme knows this happens, half the population gets screwed either way i guess, 40-45 higher ping since they switched.... and spiking at times to more. its really great they choose to screw people over the day before double drop.....after all the decent work they did leading up to this....just great.....my ping is literally the worst it has ever been in tera sincei began playing in whhm patch.
  • My old ping was 60-80ms, now hovers around 110-130. Feels good
  • TechGuyTechGuy ✭✭
    edited November 2018
    for me its really weird because I have 60ms on velika, but 150ms on kaiator??? does it makes any sense? arent both servers located on the same place physically?
  • sanj66sanj66 ✭✭✭✭
    now my ping is down to 90ms.....what is really going on here o.o
  • As noted above, there was some weird problem this morning where Kaiator was having a lot more server lag than Velika server, but I don't think it's related to this issue because they are both at the same DC. They appeared to fix that somewhere around 9:20am PST from the reports of people saying it was fixed. So it's possible that some of the bad ping times people were seeing originally may actually have been related to this, and now is a better test of how it should be. See how it goes...
  • sanj66sanj66 ✭✭✭✭
    As noted above, there was some weird problem this morning where Kaiator was having a lot more server lag than Velika server, but I don't think it's related to this issue because they are both at the same DC. They appeared to fix that somewhere around 9:20am PST from the reports of people saying it was fixed. So it's possible that some of the bad ping times people were seeing originally may actually have been related to this, and now is a better test of how it should be. See how it goes...

    no its the same thing again tonight lol 160-180 ms
  • tisnotmetisnotme ✭✭✭✭
    I see it has added an additional 100ms to what I had previous and that wasn't good to start with due to where I play from
  • sanj66sanj66 ✭✭✭✭
    the game is actually unplayable at this point idk why they all of a sudden decided hey lets screw people over, a few hours of unstable low ping and then 160-180 ms plus for the majority of the time. great job eme.......
  • It seems like the same thing happened to Kaiator again somehow. Velika is okay, but ping on Kaiator is high. Really weird, but still seems like some other reason.
  • Still low and stable ping on my end, 142-144ms at roughly 5300 miles from Chicago.
Sign In or Register to comment.