24 replies to this topic

[General Solo]

Hello, lately I have been having a horrible time playing MWO due to an increase in latency of 50- 90ms.
I'm a Telstra Customer using a 4G celluar connection, which was very satisfactory untill recently (Fast 50 down/ 20 up but bloody expensive).

The reason I'm here is to find evidence and seek advice so that I can take action in order to resolve this issue.

It is already known from previous posts on the issue that some Australian data traffic is being routed to, Italy France and then Canada and that this the cause of the increased latenecy.

I had a gut feeling that PGI's host OVH might have something to do with the problem but was not sure. So I decided to investage further because I wish to avoid changing ISP's if possible.(Yes I contacted Telstra)

I ran a trace route to 3 different servers located in Montreal Canada

mwomercs.com hosted by OVH
www.ovh.com hosted by OVH
www.bluefur.com hosted by iWeb Technologies

I used the website http://www.ip-tracker.org to obtain the locations of the routers.




Trace Route 1 to mwomercs.com in Montreal Canada hosted by OVH

Spoiler

Tracing route to mwomercs.com [192.99.109.192]
over a maximum of 30 hops:
1 * * * Request timed out.
2 * * * Request timed out.
3 * * * Request timed out.
4 29 ms 28 ms 29 ms TenGigE0-5-0-15.chw-edge901.sydney.telstra.net [139.130.111.73]
5 39 ms 31 ms 37 ms bundle-ether13.chw-core10.sydney.telstra.net [203.50.11.98]
6 32 ms 27 ms 33 ms Bundle-ether17.oxf-gw2.sydney.telstra.net [203.50.13.70]
7 38 ms 29 ms 26 ms bundle-ether1.sydo-core01.sydney.reach.com [203.50.13.38] >> Australia
8 29 ms 28 ms 31 ms i-0-3-2-0.sydo-core02.bi.telstraglobal.net [202.84.220.189] >>Hong Kong
9 169 ms 226 ms 164 ms i-0-2-0-2.1wlt-core01.bx.telstraglobal.net [202.84.140.21 >>Hong Kong
10 200 ms 289 ms 319 ms i-0-4-0-0.eqla01.bi.telstraglobal.net [202.84.251.174] >>Hong Kong
11 256 ms 319 ms 319 ms tisparkle-peer.eqla01.pr.telstraglobal.net [134.159.61.26] >>Taiwan
12 181 ms 393 ms 319 ms xe-2-1-0.paloalto2.pao.seabone.net [89.221.35.145] >>Italy
13 * 236 ms 266 ms pal-5-6k.ca.us [178.32.135.206] >> France
14 521 ms * 443 ms 198.27.73.188 >> Montreal, Canada
15 359 ms 319 ms 319 ms mtl-1-6k.qc.ca [198.27.73.183] >> Montreal, Canada
16 308 ms 586 ms 319 ms bhs-g2-6k.qc.ca [198.27.73.3] >> Montreal, Canada
17 304 ms 298 ms 300 ms 198.27.73.233 >> Montreal, Canada
18 303 ms 540 ms 319 ms mwomercs.com [192.99.109.192] >> Montreal, Canada
Trace complete.

The data goes from Australia to Hong Kong, Taiwan, Italy, France, then Canada
Also the router in Taiwan which makes the descision to sent the data to Europe maybe a Telstra router going by the hostname
Note - Hop 11 - tisparkle-peer.eqla01.pr.telstraglobal.net .



Trace Route 2 to www.bluefur.com in Montreal Canada hosted by iWeb Technologies

Spoiler

Tracing route to bluefur.com [209.172.44.79]
over a maximum of 30 hops:
1 * * * Request timed out.
2 * * * Request timed out.
3 * * * Request timed out.
4 31 ms 26 ms 39 ms TenGigE0-5-0-15.chw-edge901.sydney.telstra.net [139.130.111.73]
5 26 ms 29 ms 30 ms bundle-ether13.chw-core10.sydney.telstra.net [203.50.11.98]
6 27 ms 31 ms 27 ms Bundle-ether17.oxf-gw2.sydney.telstra.net [203.50.13.70]
7 35 ms 33 ms 36 ms bundle-ether1.sydo-core01.sydney.reach.com [203.50.13.38] >> Australia
8 29 ms 35 ms 27 ms i-0-2-0-6.sydo-core02.bi.telstraglobal.net [202.84.223.46] >>Hong Kong
9 174 ms 449 ms 170 ms i-0-2-0-3.1wlt-core01.bx.telstraglobal.net [202.84.140.222]>>Hong Kong
10 259 ms 204 ms 424 ms i-0-4-0-3.eqla01.bi.telstraglobal.net [202.40.149.246]>>Hong Kong
11 201 ms 212 ms 210 ms cogent-peer.eqla01.pr.telstraglobal.net [134.159.63.198] >>Taiwan
12 515 ms 215 ms 423 ms be2181.mpd21.lax01.atlas.cogentco.com [154.54.41.113] >>USA
13 244 ms 425 ms 319 ms be2067.ccr21.iah01.atlas.cogentco.com [154.54.7.161] >>USA
14 250 ms 249 ms 384 ms be2172.ccr41.atl01.atlas.cogentco.com [154.54.29.17] >>USA
15 309 ms 269 ms 306 ms be2168.ccr21.dca01.atlas.cogentco.com [154.54.31.94] >>USA
16 269 ms 270 ms 276 ms be2148.ccr21.jfk02.atlas.cogentco.com [154.54.31.118] >>USA
17 251 ms 408 ms 319 ms be2106.ccr21.alb02.atlas.cogentco.com [154.54.3.50] >>USA
18 560 ms 319 ms 253 ms be2088.ccr21.ymq02.atlas.cogentco.com [154.54.43.17] >>USA
19 352 ms 541 ms 399 ms te8-4.cl-core06.cogent.mtl.iweb.com [38.104.154.202]>> Montreal,Canada
20 560 ms 639 ms 639 ms te8-4.dr1.mtl.iweb.com [184.107.1.114]>> Montreal,Canada
21 520 ms 368 ms 595 ms corp.bluefur.com [209.172.44.79]>> Montreal,Canada
Trace complete.

The data goes from Australia to Hong Kong, Taiwan, USA, Montreal Canada, Direct as it should be.




Trace Route 3 to www.ovh.com in Montreal Canada hosted by OVH

Spoiler

Tracing route to www.ovh.com [198.27.92.1]
over a maximum of 30 hops:

1 * * * Request timed out.
2 * * * Request timed out.
3 * * * Request timed out.
4 27 ms 36 ms 29 ms TenGigE0-5-0-15.chw-edge901.sydney.telstra.net [139.130.111.73]
5 28 ms 37 ms 31 ms bundle-ether13.chw-core10.sydney.telstra.net [203.50.11.98]
6 31 ms 35 ms 31 ms Bundle-ether17.oxf-gw2.sydney.telstra.net [203.50.13.70]
7 28 ms 39 ms 28 ms bundle-ether1.sydo-core01.sydney.reach.com [203.50.13.38] >> Australia
8 29 ms 39 ms 26 ms i-0-3-2-0.sydo-core02.bi.telstraglobal.net [202.84.220.189] >>Hong Kong
9 185 ms 177 ms 221 ms i-0-2-0-3.1wlt-core01.bx.telstraglobal.net [202.84.140.222] >>Hong Kong
10 505 ms 211 ms 427 ms i-0-4-0-3.eqla01.bi.telstraglobal.net [202.40.149.246] >>Hong Kong
11 385 ms 220 ms 418 ms tisparkle-peer.eqla01.pr.telstraglobal.net [134.159.61.26] >>Taiwan
12 179 ms 485 ms 195 ms xe-11-1-0.paloalto2.pao.seabone.net [89.221.35.129] >>Rome, Italy
13 235 ms 300 ms 319 ms pal-5-6k.ca.us [178.32.135.206] >>Paris, France
14 245 ms 239 ms 719 ms 198.27.73.156 >> Montreal,Canada
15 237 ms 417 ms 248 ms snj-2-6k.ca.us [178.32.135.168] >>Paris, France
16 520 ms 319 ms 639 ms snj-1-6k.ca.us [178.32.135.148] >>Paris, France
17 241 ms 358 ms 242 ms www.ovh.com [198.27.92.1]>> Montreal,Canada
Trace complete.

The data goes from Australia to Hong Kong, Taiwan, Italy, France, Canada, France and back to Canada (Thats some crazy routing there.)



In summary I found that servers located in Montreal Canada if hosted by OVH are routed via Europe whilst servers hosted via another hosting provider such as iWeb Technologies are routed direct to North America.

But at the same time it is telstra's router in Taiwan sending data for Montreal OVH Customers via Europe and iWeb Technologies customer direct to the North America.

So were do I look next?

Telstra because their router in taiwan (tisparkle-peer.eqla01.pr.telstraglobal.net) is routing inapproriately.

Or do do we look at OVH via PGI, as it affects OVH Montreal Customers but not servers hosted by iWeb Technologies customers also located in Montreal.

Just looking for advice how to proceed next.

Thanks you

PS:
A fellow Aussie SirLankyIII, had the same problem , and posted it on Telstra Crowd Support.

http://crowdsupport....-p/380209#M3356

Was the issue resolved?
Does this problem still persist or is it just me?

Thanks again

Edited by OZHomerOZ, 06 October 2014 - 11:46 PM.

[Kiiyor]

Have you contacted Telstra about it? I have a seething hatred for them, but they may be able to help.

Maybe if someone with unaffected ping posted their results, it could help. Is anyone else with Telstra similarly affected (or unaffected)?

[Petard]

LOL, I too, have a seething hatred for Telstra...

[Kiiyor]

Petard, on 07 October 2014 - 01:34 AM, said:

LOL, I too, have a seething hatred for Telstra...

OH YEAH. Did they do the same thing to you that they did for me? They disconnected my phone contract, bacause I hadn't paid my bill. Only I had. After being forced to pay the bill (again) they then discovered that it was indeed their fault. I then had to pay a reconnection fee, after which they informed me that they could not reconnect my phone for 30 days, because "that's just the way the system worked". They then refused to refund any of the free money I gave them, because it was "lodged as a dispute". 12 days later, still no progress.

Called the ombudsman, provided some details, he told me he'd call me back once it was sorted. 1 hour and 22 minutes later, all was well. Turns out the ombudsman is a badass.

The moral of the story is THIS - homer, if all else fails, call the ombudsman.

[N0MAD]

Strange that you can trace any addy that isnt OVH in Montreal and you get your normal route, its just OVH that has the strange EU routing,, whats that telling you.
BTW im a telstra customer, my trace goes thru California to Canada..

[General Solo]

Thanks for the replies

Agreed, it maybe a OVH issue, and this was why I posted, to get more input on this possibility.
Or perhaps Telstra treats prepaid traffic like a second class citizen. Yes I did contact Telstra, It got escalated to level 3, and they said they couldn't help, but I'll try again tomorrow. After that, the ombudsman.

Edited by OZHomerOZ, 07 October 2014 - 02:55 AM.

[Zanathan]

I have the luxury of being able to use two connections, my folks are on Telstra and I'm on TPG. When the datacentre move happened I also noticed the same behaviour on the Telstra connection, the calls to MWO where routing via Europe. This bumped the ping rates from an average of 250 to 350+. However on the TPG connection it was fine.

Long story short, the routing issue has stopped on the Telstra connection now and its routing properly. Although pings are a little higher then before, just shy of 300ms but one thing I regularly get with the Telstra connection is lag spikes which can get as bad as 2000ms+ and I never experience this on the TPG connection. Some preliminary testing has shown that the lag spikes don't occur for any other services/games I've been testing with so it looks like it's only occuring with MWO.

I haven't been patient enough to delve further but I would contact the big T and find out if they can do anything given other Telstra customers have had the routing issue resolved.

[Impyrium]

I had this exact problem with the new servers came online, which persisted for a few weeks... then it sort of fixed itself. I dunno, it stopped routing through Europe and went straight to America. I assumed Telstra or whoever must have picked up on something or changed it on their own. Not entirely sure why, but it works now, so I'm not complaining.

[xengk]

might be related to your problem.
currently affecting southeast asia's connection to north america
I have been seeing an extra 100 ping on my connection the past 2 week.

http://www.lowyat.ne...southeast-asia/

[General Solo]

The problem seems to be fixed, guess I better thank Telstra or OVH via PGI or someone.

Oh Yeah thank you guys.

[Troutmonkey]

I blame "Mr Broadband"

[riTk0MA]

Hey, Aussies this is a little off-topic but you're best suited to help me:
Since the server move I've noticed that when my latency increases above 250ms, lasers deal reduced damage or stop dealing damage at all. Are you guys experiencing the same thing?

[SirLankyIII]

OZHomerOZ, on 06 October 2014 - 11:40 PM, said:

PS:
A fellow Aussie SirLankyIII, had the same problem , and posted it on Telstra Crowd Support.

http://crowdsupport....-p/380209#M3356

Was the issue resolved?
Does this problem still persist or is it just me?

Thanks again

They claim it's fixed, but my traceroute still looks similar to the one you posted (and my ping is still 290-320ms on average campared to 250-280ms prior), but Jon mentioned in a PM that my trace route is going via hong kong to the palo alto servers in LA. So either Telstra is still screwy (I wish they'd stop going via Hong Kong and just go straight to the US), or there's a problem somewhere else down the line. It's worth mentioning that there is a huge spike once it enters canada which is either down to some issue with one of the hops in canada, or it's still crossing back over the Atlantic and Telstra is full of ****.

Edit: the funny thing is, when i Ping OVH I get a ping of 240ms (but my traceroute times out, which isn't entirely unexpected) and bluefur is high like mwo

Edited by SirLankyIII, 08 October 2014 - 04:05 AM.

[King Arthur IV]

"you're on an island" huehuehuehue

by you i really mean we...

Edited by King Arthur IV, 08 October 2014 - 07:33 AM.

[Petard]

Kiiyor, on 07 October 2014 - 01:52 AM, said:

OH YEAH. Did they do the same thing to you that they did for me? They disconnected my phone contract, bacause I hadn't paid my bill. Only I had. After being forced to pay the bill (again) they then discovered that it was indeed their fault. I then had to pay a reconnection fee, after which they informed me that they could not reconnect my phone for 30 days, because "that's just the way the system worked". They then refused to refund any of the free money I gave them, because it was "lodged as a dispute". 12 days later, still no progress.

Called the ombudsman, provided some details, he told me he'd call me back once it was sorted. 1 hour and 22 minutes later, all was well. Turns out the ombudsman is a badass.

The moral of the story is THIS - homer, if all else fails, call the ombudsman.

Not quite, my story is slightly different...Some years ago, after moving to a new house, I organised to have the phone hooked up via Telstra, they informed me that I needed to be home between the hours of 12 noon and 5.30 pm to enable this, no problem, I organised to leave work early that day to get it done....

Due to unforeseen circumstances, i.e. being right in the middle of work that had to be finished, I ended up leaving work a bit later than I wanted to, and drove home at slightly illegal speeds in order to make it on time, saw the flash of a speed camera on Port Wakefield road as I went past it...DOH!!!, and duly made it home at 5 minutes to 12...And sat there, and sat there, and sat there...5.30 came and went, and still no Telstra guy...Went out to check the mail box, and found a card from the technician saying that he was there at 11 am, and left because NOBODY WAS HOME...

Needless to say, I rang them on my mobile straight away, and told them the story, and they offered me a 12 dollar rebate on my first bill to compensate me for the inconvenience...A WHOLE 12 dollars, after them costing me half a day's pay, and a 300 dollar speeding fine, for ab-so-f*****g-lutely nothing...

So I told em to get stuffed, and stick their 12 dollars where the sun don't shine, organised my phone through Optus, and resolved to NEVER, EVER, EVER give that pack of scabby diseased goat rectums another cent of my hard earned cash ever again....So yeah, it's fair to say that Telstra have been off of my Xmas list for quite some time.

Edited by Petard, 08 October 2014 - 05:01 AM.

[Reitrix]

Your first 3 hops are timing out, you should look into that first.

[Kiiyor]

riTk0MA, on 08 October 2014 - 03:46 AM, said:

Hey, Aussies this is a little off-topic but you're best suited to help me:
Since the server move I've noticed that when my latency increases above 250ms, lasers deal reduced damage or stop dealing damage at all. Are you guys experiencing the same thing?

YES.

The trouble is, my evidence is all anecdotal. I love laser boats. Sometimes I think I've had an amazing game, and have rubbish damage. Sometimes it's the opposite. It seems to vary depending on the time of day.

It's odd, and all I've been able to narrow it down to is HSR. When it works, it's GREAT. I can aim at lights, and actually HIT what I aim at. Other times, I have no chance, and have to resort to leading. Takes me back to the pre-HSR days of leading between 1 and 4 mech lengths to score hits on ravens.

[Troutmonkey]

Petard, on 08 October 2014 - 04:54 AM, said:

So I told em to get stuffed, and stick their 12 dollars where the sun don't shine, organised my phone through Optus, and resolved to NEVER, EVER, EVER give that pack of scabby diseased goat rectums another cent of my hard earned cash ever again....So yeah, it's fair to say that Telstra have been off of my Xmas list for quite some time.

Internode is your friend. Saying that net still sucks, because it's all owned by Telstra anyway. At least I don't have to deal with them directly.

[XxXAbsolutZeroXxX]

OZHomerOZ, on 06 October 2014 - 11:40 PM, said:

It is already known from previous posts on the issue that some Australian data traffic is being routed to, Italy France and then Canada and that this the cause of the increased latenecy.

It isn't routed through europe.

There's an undersea telecom cable connecting australia to europe that provides australia with internet access.

Unfortunately, there is no telecom cable from australia directly to canada. Connecting that way isn't an option for geographic reasons.

Satellite internet might offer better connectivity if it can route directly from australia, to a satellite in orbit, directly to canada.

Using a hardline instead of a wifi router is probably the best upgrade for most.

[Petard]

Troutmonkey, on 08 October 2014 - 06:17 AM, said:

Internode is your friend. Saying that net still sucks, because it's all owned by Telstra anyway. At least I don't have to deal with them directly.

My Internet provider is Adam Internet at the moment, so far, after 2 years, I have been very pleased with the service and support they have provided...I have heard very good reports about Internode too, but, seeing as how i get download speeds of between 1 mbps and 1.2 mbps, and 250 ping to MWO, plus 200 gig limit per month for 60 bucks, I dont feel any need to change right now....

Unless, of course, you can tell me that you get a far better service than that, my friend...

Edited by Petard, 08 October 2014 - 09:49 PM.