69 replies to this topic

[Kaspirikay]

Ed Steele, on 11 September 2014 - 08:17 PM, said:

Good information, but data does not quite travel at the speed of light, even through fiber, because of the transducers required to convert the light to electricity and back.

Hell, if you want to get technical, you'll have to take into account the chipset clock speeds that are used in the router's circuits to do the conversion.

[SirLankyIII]

Mine has settled around 300ms +/- 10ms depending on time of day, which is still 40ms higher than it was mind you. That said it spikes as high as 340 for very short periods which it didn't do near as often before hand. I can't help but feel it's the jump to the telstraglobal servers in hong kong that is causing the slow down. Why go via hong kong when you can just cross the pacific? Is it just that much cheaper?

[lpmagic]

Well, I live on the west coast, and the move has cost me about 50 ms or so, I used to live at between 75-90 now I run up to the 130-140 range, I will run the trace route tonight when I get home.

[Clit Beastwood]

I'm in Southern California. My ping went from 75-85 to 92. Not a dramatic increase, but my hitreg is dramatically worse. Used to be pretty consistent 800+ dmg regardless of mech. Lucky to break 200 now. Last night every single drop, twf, scr, even my uber reliable Ember- 200dmg, 236 dmg, 210 dmg. It's been bad since the rejoin patch, but it's been *miserable* since the server move. This game is not fun when all your weapons sparkle ineffectually on enemy armor. I contacted support about it and the reply, summed up, was "yeah hitreg is bad, especially lasers".

Edited by Fierostetz, 12 September 2014 - 06:07 AM.

[Krapsack]

I was at about 42 ping before the migration, 156 after.

Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 172.16.0.1
2 1 ms 1 ms 1 ms 192.168.0.1
3 26 ms 24 ms 25 ms mpls-dsl-gw57.mpls.qwest.net [207.225.140.57]
4 22 ms 22 ms 22 ms mpls-agw1.inet.qwest.net [75.168.229.193]
5 33 ms 32 ms 32 ms chp-brdr-03.inet.qwest.net [67.14.8.194]
6 60 ms 58 ms 58 ms 62.157.250.1
7 139 ms 139 ms 138 ms 217.239.40.234
8 * * 177 ms fra-1-6k.de.eu [94.23.122.153]
9 142 ms 142 ms 142 ms sbg-g2-a9.fr.eu [91.121.128.125]
10 143 ms 143 ms 142 ms 37.187.36.155
11 * * * Request timed out.
12 * * * Request timed out.
13 141 ms 142 ms 142 ms mwomercs.com [192.99.109.192]

Trace complete.

[Jon Cunningham]

DeeBeeP, on 12 September 2014 - 12:49 AM, said:

I don't know what my route before the data center move was but I had a consistent 180ms latency to the old data center which was pretty good for where I live.

YIKES! NTT in Japan is routing your traffic to Milano (!) before sending it back.

This is a good example of the kind of stuff we can fix. I've leaned on our partners, if you could possibly contact your ISP as well and show your traceroute to them, that would be great. Make sure you are talking to a competent person, if they don't understand network pathing, probably want to ask for someone else. Basically someone who will know that they need to take your traceroute and show it to their network engineers.

Krapsack, on 12 September 2014 - 07:10 AM, said:

I was at about 42 ping before the migration, 156 after.

Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 172.16.0.1
2 1 ms 1 ms 1 ms 192.168.0.1
3 26 ms 24 ms 25 ms mpls-dsl-gw57.mpls.qwest.net [207.225.140.57]
4 22 ms 22 ms 22 ms mpls-agw1.inet.qwest.net [75.168.229.193]
5 33 ms 32 ms 32 ms chp-brdr-03.inet.qwest.net [67.14.8.194]
6 60 ms 58 ms 58 ms 62.157.250.1
7 139 ms 139 ms 138 ms 217.239.40.234
8 * * 177 ms fra-1-6k.de.eu [94.23.122.153]
9 142 ms 142 ms 142 ms sbg-g2-a9.fr.eu [91.121.128.125]
10 143 ms 143 ms 142 ms 37.187.36.155
11 * * * Request timed out.
12 * * * Request timed out.
13 141 ms 142 ms 142 ms mwomercs.com [192.99.109.192]

Trace complete.

Yeah. Another Qwest customer having their traffic sent to Germany. Absurd. I've already been leaning on Qwest to get this fixed.

[Deathlike]

My only suggestion here is to sticky this thread for better visibility.

[Symbiodinium]

I'm in New Zealand, ISP is Vodafone. Vodafone bought Telstra's ISP service nationwide last year so consider this Telstra with another brand name (my bill payments still say Telstra). Ping before the DC migration was about 220, jumped to 320 after the move, sometimes 400+. Seems to be routing from NZ to Hong Kong, Italy, France, then Canada. Can we have this post bumped up somewhere? I had to search for it (even though I knew it existed) and it took a while.

Tracert output:

Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 * 1 ms 1 ms NAS [192.168.20.1]
2 * * * Request timed out.
3 15 ms 12 ms 13 ms lo0.internet.ivpn.pe19.telstraclear.net [218.101.61.100]
4 20 ms 21 ms 22 ms ie2-g-0-0-0.telstraclear.net [203.98.50.2]
5 22 ms 20 ms 22 ms ge-0-2-0-1.xcore1.acld.telstraclear.net [203.98.50.251]
6 25 ms 24 ms 23 ms unknown.telstraglobal.net [134.159.174.41]
7 149 ms 150 ms 148 ms i-0-0-4-1.tlot-core01.bx.telstraglobal.net [202.84.142.106]
8 146 ms 146 ms 144 ms i-0-5-0-2.eqla01.bi.telstraglobal.net [202.84.253.10]
9 245 ms 249 ms 245 ms tisparkle-peer.eqla01.pr.telstraglobal.net [134.159.61.26]
10 244 ms 244 ms 272 ms xe-2-1-0.paloalto2.pao.seabone.net [89.221.35.145]
11 245 ms 245 ms 245 ms pal-5-6k.ca.us [178.32.135.206]
12 * 293 ms * 198.27.73.177
13 302 ms 303 ms 302 ms 198.27.73.175
14 475 ms 322 ms 323 ms bhs-g1-6k.qc.ca [198.27.73.1]
15 523 ms 309 ms 309 ms 198.27.73.231
16 321 ms 320 ms 320 ms mwomercs.com [192.99.109.192]

Trace complete.

Edited by Symbiodinium, 12 September 2014 - 03:22 PM.

[Primetimex]

Thanks Jon for looking into these network issues for us, here is my tracert since I haven't posted here before:

Tracing route to www.mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms fritz.box [192.168.178.1]
2 7 ms 6 ms 6 ms Lo0.amnet.l1-45sgt-se800-01.wa.amnet.net.au [203
.161.65.253]
3 7 ms 7 ms 8 ms vl70.ten7-3.cr01.wa.amnet.net.au [203.161.65.41]

4 7 ms 7 ms 7 ms te3-2.br02.wa.amcom.net.au [203.161.65.65]
5 59 ms 59 ms 59 ms mail-o.mcms.com.au [61.88.8.41]
6 * * * Request timed out.
7 * * * Request timed out.
8 * * * Request timed out.
9 114 ms 114 ms 115 ms 59.154.142.236
10 266 ms 265 ms 270 ms 203.208.190.17
11 267 ms 269 ms 266 ms 203.208.173.130
12 * * * Request timed out.
13 * * * Request timed out.
14 335 ms 333 ms 334 ms 198.27.73.173
15 337 ms 336 ms 336 ms bhs-g1-6k.qc.ca [198.27.73.5]
16 331 ms 330 ms 330 ms 198.27.73.231
17 335 ms 334 ms 336 ms mwomercs.com [192.99.109.192]

Trace complete.

[BigBucket]

Thanks for the info. This explains why my ping went from 29-31 to 39-41. I am in Wisconsin and the sever moved a little further away.
I do not mind losing a few pings if it benefits the whole.
I do still experience some desyncing where I ping 70+ and have other strange symptoms which are usualy fixed by exiting and rejoining the match. I can tell when I have the desyncs when I cannot see what I myself type in chat. Sometimes it seems my teammates can see what I type and sometimes not.
Do we have any info on this type of minor desyncing? If you would like tracerts or any other info please let me know.

Edit: words hard phone buttons weak!

Edited by BigBucket, 12 September 2014 - 04:52 PM.

[elsie]

From Oregon:

https://www.dropbox....4-9-12.jpg?dl=0



Seems like a lot of dropped packets out there today.

Edited by elsie, 12 September 2014 - 05:13 PM.

[WM Wraith]

Central Texas. My pings went double after the move - much more latency on my end.

Before data center move - constant 45 ms ping at all times.

After: 93ms low up to average of 110ms ping times.

best traceroute time is attached - 93 ms



C:\Windows\system32>tracert 192.99.109.192

Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 192.168.1.1
2 2 ms 2 ms 2 ms 10.50.152.1
3 * * * Request timed out.
4 2 ms 2 ms 2 ms rrcs-24-173-24-221.sw.biz.rr.com [24.173.255.221
]
5 3 ms 3 ms 3 ms xe-1-2-0-0.CNLKTX271ZW.tx.twcbiz.com [24.73.242.
160]
6 8 ms 8 ms 8 ms xe-0-2-0-0.NBRNTXLU1CW.tx.twcbiz.com [97.77.2.20
0]
7 8 ms 9 ms 9 ms tge0-8-0-8.lvoktxad01r.texas.rr.com [24.175.33.5
6]
8 9 ms 9 ms 9 ms agg21.snavtxuu02r.texas.rr.com [24.175.32.144]
9 10 ms 11 ms 7 ms agg23.hstqtxl301r.texas.rr.com [24.175.32.156]
10 8 ms 11 ms 11 ms 107.14.19.94
11 13 ms 15 ms 16 ms ae-0-0.cr0.dfw10.tbone.rr.com [66.109.6.39]
12 14 ms 14 ms 14 ms ae0.pr1.dfw10.tbone.rr.com [107.14.17.232]
13 12 ms 12 ms 12 ms ix-23-0.tcore2.DT8-Dallas.as6453.net [66.110.57.
97]
14 48 ms 47 ms 47 ms if-2-2.tcore1.DT8-Dallas.as6453.net [66.110.56.5
]
15 * 47 ms * if-10-3.tcore1.AEQ-Ashburn.as6453.net [66.198.15
4.117]
16 50 ms 48 ms 48 ms if-2-2.tcore2.AEQ-Ashburn.as6453.net [216.6.87.1
]
17 46 ms 47 ms 47 ms if-11-2.tcore2.NJY-Newark.as6453.net [216.6.87.1
38]
18 48 ms 48 ms 47 ms if-2-2.tcore1.NJY-Newark.as6453.net [66.198.70.1
]
19 * 90 ms * 192.99.146.48
20 92 ms 91 ms 92 ms bhs-g2-6k.qc.ca [198.27.73.207]
21 91 ms 91 ms 91 ms 198.27.73.233
22 92 ms 93 ms 93 ms mwomercs.com [192.99.109.192]

Trace complete.

C:\Windows\system32>

[M a y h e m]

Hi, Jon. This is a TR from Vancouver BC, overall ping is OK but a hop in Quebec is encountering 100% PL..

[Jon Cunningham]

Alaric VISIGOTH, on 12 September 2014 - 05:28 PM, said:

Hi, Jon. This is a TR from Vancouver BC, overall ping is OK but a hop in Quebec is encountering 100% PL..

That's fine - many routers start dropping pings, especially if they are under load, because they're focusing on delivering the other content. If that router dropped "100%" of all packets traveling through it, you wouldn't even be able to reach the subsequent steps of the path, let alone our website!

So long as *your* ping time is fine (it is; 85 ms is slightly better than what we get in the office, which is 87) and your actual packets aren't getting dropped / gameplay isn't being affected, you're good!

[100 Tonne]

I am from Australia. The ping seems to have settled down to 320-330. this is up from 290-300 from before the move, but better than just over 500 which it has been last week.

[Violent Nick]

Hi there..

Just wanted to register the fact that my PING is all over the place. The standard level for me is 102 being in the UK and uding Virgin Media 50mb/s fibre optic broadband (soon to be upgraded..) but over the past week it has spiked to 300 and even 600 Ping. This has only happened in the past week since the Data Center move. I had suffered massive desyncs before with a lot of people, but never a Ping spike like these. It's getting worse also it seems. Today my ping reached 645.

Thanks..

[Mike Forst]

For anyone who is curious about how data travels across (well...under) the major oceans there's a pretty cool website that shows all the current and future submarine fibre lines:

http://www.submarinecablemap.com

With that it's pretty easy to get a visual example of how one wrong hop at an internet backbone router can send your packets in the wrong direction.

[Ultimax]

This thread is not only amazing, watching these folks work out this issue - but also informative on a subject of which I have zero knowledge.


5/5.

[SirLankyIII]

Mike Forst, on 12 September 2014 - 08:58 PM, said:

For anyone who is curious about how data travels across (well...under) the major oceans there's a pretty cool website that shows all the current and future submarine fibre lines:

http://www.submarinecablemap.com

With that it's pretty easy to get a visual example of how one wrong hop at an internet backbone router can send your packets in the wrong direction.

That maps just shows how ridiculous it is for a connection from Melbourne to go to Hong Kong first and then to Los Angeles. You end up travelling almost the same distance along the cable from Melbourne to Hong Kong as you do from Sydney to LA. There really is no reason to go via Hong Kong unless that route is simply cheaper.

What's even worse is just going by eye, it looks like the route from Hong Kong to LA is longer than it is from Sydney to LA.

[ice trey]

Location: Fukushima prefecture, Japan

My internet is going over the wrong ocean.

Before this, a ping of 180-200 was normal. Not GOOD, but normal, especially when my old Ottawa ping was about 18-45

C:\Users\Michael>tracert mwomercs.com
 
Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:
 
  1	 6 ms	 6 ms	 6 ms  nas813.p-fukushima.nttpc.ne.jp [210.153.247.99]
 
  2	13 ms	13 ms	13 ms  210.165.255.161
  3	18 ms	17 ms	17 ms  e1-1-n-otemachi-core13.sphere.ad.jp [210.165.252.201]
  4	13 ms	13 ms	13 ms  xg1-1-n-otemachi-gw1.sphere.ad.jp [210.153.254.157]
  5	15 ms	15 ms	15 ms  61.207.54.117
  6	17 ms	15 ms	15 ms  122.28.169.137
  7	15 ms	15 ms	15 ms  60.37.27.153
  8	22 ms	23 ms	29 ms  ae-5.r21.tokyjp05.jp.bb.gin.ntt.net [129.250.11.53]
  9	58 ms	25 ms	25 ms  ae-0.r24.tokyjp05.jp.bb.gin.ntt.net [129.250.6.96]
10	99 ms	99 ms	99 ms  ae-3.r21.sngpsi05.sg.bb.gin.ntt.net [129.250.7.35]
11   100 ms   100 ms   100 ms  ae-13.r00.sngpsi02.sg.bb.gin.ntt.net [129.250.4.74]
12   112 ms   104 ms	98 ms  ae0.singapore2.sin.seabone.net [93.186.133.12]
13   306 ms   306 ms   306 ms  xe-1-1-0.milano51.mil.seabone.net [195.22.205.243]
14	 *		*		*	 Request timed out.
15   296 ms	 *	  296 ms  mil-5-6k.it.eu [94.23.122.251]
16   299 ms	 *	  299 ms  fra-5-6k.fr.eu [91.121.128.35]
17   301 ms   301 ms   301 ms  sbg-g2-a9.fr.eu [91.121.128.125]
18   297 ms   297 ms   297 ms  37.187.36.155
19	 *		*		*	 Request timed out.
20	 *		*		*	 Request timed out.
21   301 ms   296 ms   296 ms  mwomercs.com [192.99.109.192]

Edited by ice trey, 12 September 2014 - 09:51 PM.