356 replies to this topic

[Oderint dum Metuant]

Batch1972, on 14 February 2016 - 02:19 AM, said:

Hey Guys,

Telstra user from Sydney. I've had ping increase from approx 250 to 550+ when connecting to the US server for the last two weeks or so. This is also happening for other games that I play with US servers. I spoke to Telstra and they have advised that there is an issue with the PPC-1 cable out of Guam.. not sure that I believe them

If you can run a traceroute. It'll show you where the ping increase is.
I wouldn't be surprised to see .fr hops in there and some core network structure ping increases.

Of course if you really want to throw the spotlight on Telstra, you could try a decent VPN service if your ping improves (which i suspect it would) then it's down to Telstra's routing/network.
There are some that do a free trial type thing.

Edited by DV McKenna, 14 February 2016 - 04:08 AM.

[Karamarka]

C:\Users\*******>ping 101.100.201.86

Pinging 101.100.201.86 with 32 bytes of data:
Reply from 101.100.201.86: bytes=32 time=375ms TTL=112
Reply from 101.100.201.86: bytes=32 time=369ms TTL=112
Reply from 101.100.201.86: bytes=32 time=358ms TTL=112
Reply from 101.100.201.86: bytes=32 time=371ms TTL=112

Ping statistics for 101.100.201.86:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 358ms, Maximum = 375ms, Average = 368ms

Amazing. This is from AUS. I get better ping to North America server, 270-280 which is higher then Normal too. (West coast should be 180, Mid is 230 and East is 250)

I wish they just had the NA server in the West Coast. Los Angeles is a direct link and only 180 ping for any AUS player.

Edited by Karamarka, 14 February 2016 - 04:52 AM.

[OmegaUltima]

Brisbane | iinet

|------------------------------------------------------------------------------------------|
| WinMTR statistics |
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| router.asus.com - 0 | 31 | 31 | 0 | 0 | 1 | 1 |
| lo0.bng1.bne4.on.ii.net - 0 | 31 | 31 | 20 | 23 | 31 | 22 |
| ae4.cr1.bne7.on.ii.net - 0 | 31 | 31 | 19 | 25 | 37 | 23 |
| ae6.br1.syd4.on.ii.net - 0 | 31 | 31 | 123 | 138 | 181 | 133 |
| ae16.cr1.adl2.on.ii.net - 0 | 31 | 31 | 124 | 127 | 130 | 129 |
| ae17.cr1.per1.on.ii.net - 0 | 31 | 31 | 77 | 83 | 88 | 80 |
| po0-3-0.bdr1.sin1.on.ii.net - 8 | 26 | 24 | 124 | 128 | 138 | 125 |
| gw-4657.starhub.com.sg - 0 | 31 | 31 | 127 | 131 | 142 | 127 |
| w118012041.w118012.starhub.net.sg - 4 | 28 | 27 | 125 | 127 | 137 | 127 |
| 203.118.2.25 - 0 | 31 | 31 | 127 | 130 | 138 | 129 |
| an-ats-int11.starhub.net.sg - 9 | 24 | 22 | 125 | 128 | 136 | 125 |
| sh2.nc1.sg.vodien.com - 0 | 31 | 31 | 125 | 128 | 139 | 127 |
| 86.201.100.101.in-addr.arpa - 0 | 31 | 31 | 126 | 132 | 144 | 126 |
|________________________________________________|______|______|______|______|______|______|
WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

[EasyPickings]

It looks like the PPC-1 submarine cable between Sydney and Guam went down on 5 Feb. The outage is causing all traffic to be re-routed through the Southern Cross link, which is what I normally am traversing (from NZ). All of the extra traffic is causing my latency issues and will until the PPC-1 is repaired, probably end of March.

Articles here, here and here.

[A Corpse]

I'm on fibre in NZ and I see 220-280 ping on all three regions. Disappointing, really.

[jjm1]

packet loss again on the "oceanic" server.

PGI seriously needs to drop this crappy second rate provider already.

|------------------------------------------------------------------------------------------|
|									  WinMTR statistics								   |
|					   Host			  -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|					   dsldevice.gateway -    0 |   93 |   93 |    0 |    0 |    1 |    0 |
|				   No response from host -  100 |   19 |    0 |    0 |    0 |    0 |    0 |
|						 x -    0 |   93 |   93 |   20 |   24 |   46 |   46 |
|bundle-ether7.wel-core3.perth.telstra.net -    0 |   93 |   93 |   19 |   25 |   44 |   44 |
|tengige0-8-0-5-1.pie-core1.perth.telstra.net -    0 |   93 |   93 |   19 |   25 |   48 |   48 |
|tengige0-5-1-0.pthp-core01.perth.net.reach.com -    0 |   93 |   93 |   19 |   24 |   48 |   48 |
|i-0-0-1-0.skdi-core01.bx.telstraglobal.net -    0 |   93 |   93 |   67 |   71 |   90 |   90 |
|   i-0-3-0-0.6ntp02.bi.telstraglobal.net -    0 |   93 |   93 |   68 |   72 |  103 |   91 |
|						   203.116.5.157 -    0 |   93 |   93 |  189 |  193 |  233 |  233 |
|						  203.118.15.186 -    0 |   93 |   93 |  187 |  189 |  226 |  226 |
|			 an-ats-int11.starhub.net.sg -    0 |   93 |   93 |  188 |  191 |  225 |  225 |
|				   sh2.nc1.sg.vodien.com -    0 |   93 |   93 |  163 |  168 |  209 |  209 |
|			 86.201.100.101.in-addr.arpa -   15 |   60 |   51 |  170 |  172 |  188 |  172 |
|________________________________________________|______|______|______|______|______|______|

and proof its not the pipes. its the service:

my ping to speedtest-sgp1.digitalocean.com

|------------------------------------------------------------------------------------------|
|									  WinMTR statistics								   |
|					   Host			  -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
|					   dsldevice.gateway -    0 |   42 |   42 |    0 |    0 |    0 |    0 |
|				   No response from host -  100 |    9 |    0 |    0 |    0 |    0 |    0 |
|						 x -    0 |   42 |   42 |   18 |   21 |   23 |   22 |
|bundle-ether7.wel-core3.perth.telstra.net -    0 |   42 |   42 |   20 |   23 |   28 |   23 |
|tengige0-7-0-0.pthw-core01.perth.net.reach.com -    0 |   42 |   42 |   21 |   25 |   61 |   22 |
|i-0-0-1-0.istt-core02.bx.telstraglobal.net -    0 |   42 |   42 |   67 |   71 |   82 |   67 |
|   i-0-0-0-0.istt02.bi.telstraglobal.net -    0 |   42 |   42 |   67 |   71 |   85 |   68 |
|			   unknown.telstraglobal.net -    0 |   42 |   42 |   66 |   69 |  125 |   67 |
|						 103.253.144.238 -    0 |   42 |   42 |   68 |   71 |   77 |   68 |
|						  128.199.90.252 -    0 |   42 |   42 |   67 |   69 |   78 |   67 |
|________________________________________________|______|______|______|______|______|______|

[Kairaku]

Quote

It looks like the PPC-1 submarine cable between Sydney and Guam went down on 5 Feb. The outage is causing all traffic to be re-routed through the Southern Cross link, which is what I normally am traversing (from NZ). All of the extra traffic is causing my latency issues and will until the PPC-1 is repaired, probably end of March.

That doesn't explain what is happening on my traceroutes. The Southern Cross goes from Auckland, Sydney, Hawaii, US west Coast, Hawaii, Auckland. As you will see below Telstra in their infinite wisdom are routing through Bahrain (Bx), Burundi (Bi), and France before crossing to Canada.

Is anyone in New Zealand getting different routing? What ISP are you with?



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 * * * Request timed out.
3 136 ms 134 ms 128 ms lo0.internet.ivpn.pe25.telstraclear.net [218.101
.61.124]
4 111 ms 66 ms 48 ms ie2-g-0-0-0.telstraclear.net [203.98.50.2]
5 55 ms 19 ms 19 ms ge-0-2-0-1.xcore1.acld.telstraclear.net [203.98.
50.251]
6 22 ms 24 ms 22 ms unknown.telstraglobal.net [134.159.174.37]
7 154 ms 152 ms 161 ms i-0-0-4-1.1wlt-core01.bx.telstraglobal.net [202.
84.142.110]
8 213 ms 215 ms 211 ms i-0-4-0-4.tnrt-core01.bi.telstraglobal.net [202.
84.141.130]
9 297 ms 296 ms 295 ms i-0-3-0-2.ulhc-core01.bx.telstraglobal.net [202.
84.143.230]
10 286 ms 286 ms 286 ms i-0-0-0-5.ulco01.bi.telstraglobal.net [202.84.17
8.10]
11 * * * Request timed out.
12 314 ms 304 ms 305 ms sbg-g1-a9.fr.eu [37.187.232.86]
13 304 ms 303 ms 308 ms sbg-g5-a9.fr.eu [37.187.36.153]
14 * * * Request timed out.
15 394 ms 392 ms 393 ms mwomercs.com [192.99.109.192]

Trace complete.

[EasyPickings]

It probably depends on time of day. I use Vodafone from NZ and this was taken at 1:30pm local:
Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 3 ms <1 ms <1 ms 192.168.90.1
2 29 ms 21 ms 21 ms 210-246-47-254.dsl.dyn.ihug.co.nz [210.246.47.254]
3 22 ms 21 ms 21 ms bvi-400.bgnzwtc02.wlg.vf.net.nz [203.109.180.210]
4 22 ms 21 ms 22 ms bvi-188.bgnzwtc02.wlg.vf.net.nz [203.109.180.209]
5 156 ms 155 ms 155 ms vodafone-new-zealand.as9500.any2ix.coresite.com [206.72.210.227]
6 159 ms 159 ms 158 ms ovh.as16276.any2ix.coresite.com [206.72.210.214]
7 223 ms 224 ms 223 ms be100-1015.nwk-5-a9.nj.us [198.27.73.193]
8 232 ms 232 ms 232 ms bhs-g2-a9.qc.ca [192.99.146.101]
9 231 ms 231 ms 232 ms bhs-g5-a9.qc.ca [198.27.73.233]
10 231 ms 231 ms 231 ms mwomercs.com [192.99.109.192]

Trace complete.

As you can see, it goes straight from NZ to Newark, NJ, USA, and from there to Canada. I'll try again tonight and post the results.

EDIT:

Now it's 4:30pm here and the results are different:


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

1 * * * Request timed out.
2 19 ms 72 ms 29 ms 218.101.61.29
3 20 ms 19 ms 20 ms ie2-g-0-0-0.telstraclear.net [203.98.50.2]
4 18 ms 19 ms 36 ms ge-0-2-0-1.xcore1.acld.telstraclear.net [203.98.50.251]
5 22 ms 19 ms 21 ms unknown.telstraglobal.net [134.159.174.37]
6 145 ms 148 ms 153 ms i-0-5-1-1.1wlt-core01.bx.telstraglobal.net [202.84.142.114]
7 205 ms 206 ms 206 ms 202.84.136.90
8 276 ms 279 ms 276 ms i-0-1-2-0.ulhc-core01.bx.telstraglobal.net [202.84.249.26]
9 282 ms 292 ms 281 ms i-0-0-0-5.ulco01.bi.telstraglobal.net [202.84.178.10]
10 * * * Request timed out.
11 298 ms 298 ms 299 ms sbg-g1-a9.fr.eu [37.187.232.86]
12 301 ms 298 ms 301 ms sbg-g5-a9.fr.eu [37.187.36.153]
13 * * * Request timed out.
14 375 ms 374 ms 375 ms mwomercs.com [192.99.109.192]

The routing is now through Europe, instead of straight through to the US. That's why it's taking so long; I'm being routed the long way around. It's probably due to traffic congestion on the Southern Cross link to the US (because of outages elsewhere) that's causing the re-route.

And this morning around 6:40am:

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

1 * * * Request timed out.
2 19 ms 66 ms 18 ms 218.101.61.29
3 22 ms 17 ms 18 ms ie2-g-0-0-0.telstraclear.net [203.98.50.2]
4 20 ms 19 ms 26 ms ge-0-2-0-1.xcore1.acld.telstraclear.net [203.98.
50.251]
5 20 ms 24 ms 19 ms unknown.telstraglobal.net [134.159.174.37]
6 144 ms 154 ms 146 ms i-0-5-1-1.1wlt-core01.bx.telstraglobal.net [202.
84.142.114]
7 204 ms 206 ms 205 ms 202.84.136.90
8 277 ms 278 ms 279 ms i-0-1-2-0.ulhc-core01.bx.telstraglobal.net [202.
84.249.26]
9 292 ms 282 ms 280 ms i-0-0-0-5.ulco01.bi.telstraglobal.net [202.84.17
8.10]
10 * * * Request timed out.
11 310 ms 309 ms 309 ms sbg-g1-a9.fr.eu [37.187.232.86]
12 306 ms 309 ms 308 ms sbg-g5-a9.fr.eu [37.187.36.153]
13 * * * Request timed out.
14 376 ms 379 ms 379 ms mwomercs.com [192.99.109.192]

Trace complete.

Still via the EU.

Edited by EasyPickings, 18 February 2016 - 09:40 AM.

[Batch1972]

OK so this is the routing for Telstra from Sydney as at 7:15pm on the 18th Feb:

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

1 97 ms 97 ms 98 ms BigPond.BigPond [10.0.0.138]
2 * * * Request timed out.
3 18 ms 18 ms 18 ms 144.130.209.210
4 20 ms 19 ms 19 ms bundle-ether14.chw-core10.sydney.telstra.net [203.50.11.100]
5 19 ms 19 ms 20 ms bundle-ether1.oxf-gw11.sydney.telstra.net [203.50.6.93]
6 20 ms 18 ms 19 ms bundle-ether1.sydo-core03.sydney.reach.com [203.50.13.98]
7 199 ms 201 ms 201 ms i-0-1-0-6.tlot-core01.bx.telstraglobal.net [202.84.141.178]
8 260 ms 262 ms 258 ms i-0-4-0-6.tnrt-core01.bi.telstraglobal.net [202.40.149.145]
9 336 ms 337 ms 335 ms i-0-3-0-0.ulhc-core01.bx.telstraglobal.net [202.84.143.54]
10 365 ms 365 ms 365 ms i-0-0-0-4.ulco01.bi.telstraglobal.net [202.84.178.6]
11 * * * Request timed out.
12 518 ms 517 ms 517 ms sbg-g2-a9.fr.eu [37.187.232.94]
13 519 ms 518 ms 518 ms sbg-g5-a9.fr.eu [37.187.36.155]
14 * * * Request timed out.
15 519 ms 518 ms 518 ms mwomercs.com [192.99.109.192]

Trace complete.

I presume we need to raise this as an issue with the isp

[Karamarka]

Yeah. It's really a shame that the OCE server has so high ping for many... 360 ping here lol...

OCE is good for a whole of 10 people, since everyone else at min gets like 160-200. Might as well just make a Los angeles server. Cheaper, better for everyone except those living in singapore (10 people)

[jjm1]

Karamarka, on 18 February 2016 - 01:07 AM, said:

Yeah. It's really a shame that the OCE server has so high ping for many... 360 ping here lol...

OCE is good for a whole of 10 people, since everyone else at min gets like 160-200. Might as well just make a Los angeles server. Cheaper, better for everyone except those living in singapore (10 people)

I'm sure there is a lot more than 10 players in the South East Asia region. Loads more than Oceania if its anything like other games.

It simply should be:

South East Asia (SEA) = Singapore
Oceania = Sydney

I'm usually happy playing on Singapore but it seems to be the rubber bandiest of all the servers despite having the lowest ping.

[Grinster]

Anecdata point:

Oceania server OK for me and no disconnects.
Ping times from my Internode (CBR) to the Oceania server is consistently ~150-180ms.

[Karamarka]

Grinster, on 19 February 2016 - 11:41 PM, said:

Anecdata point:

Oceania server OK for me and no disconnects.
Ping times from my Internode (CBR) to the Oceania server is consistently ~150-180ms.

Im optus and 360 to OCE now lol. used to be 165. lmao cable cutting or routing.

[EasyPickings]

I've still been getting infrequent disconnects. The next thing to check is if the disconnects still happen if I've just got MWO up and running without background programs like Team Speak competing for CPU time. With the sporadic nature of disconnects, this one will be hard to verify.

[Luca M Pryde]

Hey guys,

I have noticed in the last 2 weeks sustained poor ping results in the oceanic server.

I only just looked into now with tracert and it seems my packets arrive to Singapore on time but then suddenly are delayed in Singapore.

λ tracert 101.100.201.86

Tracing route to 86.201.100.101.in-addr.arpa [101.100.201.86]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 10.1.1.1
2 19 ms 19 ms 20 ms left out on purpose
3 * * * Request timed out.
4 20 ms 20 ms 20 ms 198.142.250.229
5 20 ms 20 ms 20 ms 198.142.139.116
6 19 ms 20 ms 20 ms 198.142.139.132
7 170 ms 240 ms 169 ms 203.208.174.189 ------ Hello Singapore!
8 211 ms 170 ms 174 ms 203.208.172.222
9 349 ms 355 ms 354 ms 203.208.173.73 ------ SING A POOOOOOOOOOOOOOOORRRRRR!
10 334 ms 360 ms 333 ms 203.208.153.253
11 360 ms 359 ms 359 ms 203.208.158.2
12 352 ms 340 ms 340 ms 203.208.190.38
13 354 ms 354 ms 389 ms 29.168.22.103.in-addr.arpa [103.22.168.29]
14 348 ms 409 ms 346 ms tl1.tl1.sg.vodien.com [103.22.168.154]
15 * * * Request timed out.
16 349 ms 369 ms 371 ms r1.nc1.sg.vodien.com [103.245.92.18]
17 341 ms 345 ms 333 ms 86.201.100.101.in-addr.arpa [101.100.201.86]

Trace complete.

[EasyPickings]

Luca M Pryde, on 23 February 2016 - 05:52 PM, said:

Hey guys,

I have noticed in the last 2 weeks sustained poor ping results in the oceanic server.

I only just looked into now with tracert and it seems my packets arrive to Singapore on time but then suddenly are delayed in Singapore.

λ tracert 101.100.201.86

Tracing route to 86.201.100.101.in-addr.arpa [101.100.201.86]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms 10.1.1.1
2 19 ms 19 ms 20 ms left out on purpose
3 * * * Request timed out.
4 20 ms 20 ms 20 ms 198.142.250.229
5 20 ms 20 ms 20 ms 198.142.139.116
6 19 ms 20 ms 20 ms 198.142.139.132
7 170 ms 240 ms 169 ms 203.208.174.189 ------ Hello Singapore!
8 211 ms 170 ms 174 ms 203.208.172.222
9 349 ms 355 ms 354 ms 203.208.173.73 ------ SING A POOOOOOOOOOOOOOOORRRRRR!
10 334 ms 360 ms 333 ms 203.208.153.253
11 360 ms 359 ms 359 ms 203.208.158.2
12 352 ms 340 ms 340 ms 203.208.190.38
13 354 ms 354 ms 389 ms 29.168.22.103.in-addr.arpa [103.22.168.29]
14 348 ms 409 ms 346 ms tl1.tl1.sg.vodien.com [103.22.168.154]
15 * * * Request timed out.
16 349 ms 369 ms 371 ms r1.nc1.sg.vodien.com [103.245.92.18]
17 341 ms 345 ms 333 ms 86.201.100.101.in-addr.arpa [101.100.201.86]

Trace complete.

These numbers can be misleading. The response at each hop is sent back to you on a lower priority basis than the actual traffic. In other words, when you're sending regular traffic (MWO connects) to the OC server it is done at regular priority. Most routers set their priorities to pass traffic first and foremost, but assign lower priorities to "service" requests like ping and traceroute. If you look at the overall ping time it reflects the whole route, but you'll occasionally see individual nodes send their response much later than the round trip takes.

In your case, you've got 3 responses for the entire route timed at 341 ms 345 ms 333 ms respectively, but node number 14 (and others) takes longer to respond to your request than the entire route: 348 ms 409 ms 346 ms.

Each node is giving its response time to your request, but isn't necessarily assigning the same priority to it as regular traffic.

Keep this in mind when trying to discover where the bottleneck lies.

[ShinVector]

EasyPickings, on 24 February 2016 - 03:45 PM, said:

These numbers can be misleading. The response at each hop is sent back to you on a lower priority basis than the actual traffic. In other words, when you're sending regular traffic (MWO connects) to the OC server it is done at regular priority. Most routers set their priorities to pass traffic first and foremost, but assign lower priorities to "service" requests like ping and traceroute.

Not exactly true. Results from ping and traceroute are usually quite reliable.

If you results is sCrap...

Typical causes of high ping are usually 'bad' route, congestion, congestion caused by ddos attackers, physical interface issues, network equipment high cpu, etc....

[EasyPickings]

ShinVector, on 24 February 2016 - 05:08 PM, said:

Not exactly true. Results from ping and traceroute are usually quite reliable.

If you results is sCrap...

Typical causes of high ping are usually 'bad' route, congestion, congestion caused by ddos attackers, physical interface issues, network equipment high cpu, etc....

You may have misunderstood what I was trying to say. The results that Luca posted may mislead him into believing he has a problem with a specific router in Singapore, whereas the returns from ping and traceroute are sent back with a lower priority and, therefore, could be delayed.

When a packet destination is the router itself, this packet has to be process-switched. The processor has to handle the information from this packet, and send an answer back. This is not the main goal of a router. By definition, a router is built to route packets. Answering a ping is offered as a best-effort service. -Understanding the Ping and Traceroute Commands

There's a good explanation of why traceroute could be misleading here.

The real question is why is there such a delay between when his packets arrive at Optimus.net AU (198.142.139.132) and get to the next hop in Singapore? Could it be that the problem is with his provider?

Edited by EasyPickings, 25 February 2016 - 04:40 PM.

[ShinVector]

Just stating my general opinion that icmp ping and traceroute are usually very accurate for determining the quality of the connection.

EasyPickings, on 25 February 2016 - 04:31 PM, said:

The real question is why is there such a delay between when his packets arrive at Optimus.net AU (198.142.139.132) and get to the next hop in Singapore? Could it be that the problem is with his provider?

Usually depends of the path that the packet took go out of Aust to Singapore and Singapore back to Aust.

Often enough it is not the same.

If outgoing path is good and return path is BAD... It is still going to suck...
Best is the lower latency a path and symmetrical routes.
Asymmetrical routes tends to have more network jitters.. Which HSR hates.

But Australia ISP of route traffic to Singapore or South East Asia from East instead of the more direct West path which sucks it as it the more round about way, I guess because it is cheaper ?

Those taking the west path probably get pings between 80ms to 150ms.

Refer to :
http://www.submarinecablemap.com/

Edited by ShinVector, 26 February 2016 - 01:11 PM.

[slide]

FWIW my ping which has been up around 400 for the last few weeks has dropped back to around 260 which is better than it ever was at around 280. I guess they fixed that cable or Telstra found a better route.

Edited by slide, 26 February 2016 - 04:32 PM.