Connection problems? [Tracert needed]

[kemor]

Connection problems? [Tracert needed] Updated

Please check this thread in the Technical forum.


UPDATE
Now, i need traceroutes from players who do NOT have problems while their friends are dropping like flies all around the realms.
Same procedure, with just a different subject : [CLEAN TRACERT]

Thanks.

 

[Jupitus]

Guys - GOA are trying to get our help tracking the lag issues we have all been bitching about recently. Please read the thread and send them the information they are looking for, so hopefully they can use it to slap Wanadoo network technicians around a bit

Jup.

(Bump btw)

 

[Jupitus]

I do hope they manage to find the problems with the router ...

(shameless, aren't I?)

Bump.

Jup.

 

[SFXman]

[cough]*slurp slurp*[/cough]
j/k mate, hope the problems get solved.

 

[old.LandShark]

SFXman woz 'ere



no need to bump this as it's been sticky'd

 

[c0ngo]

Maybe GOA should just go here http://www.traceroute.org/ and try a few.

Tracerts work fine for me here though.

 

[SilverHood]

at least they are ackowledging there's a problem....

 

[Jupitus]

This does actually remind me of the old DOS attacks this time - we're still connected but with crap lag - other nights have been all l/d, comes back up, and off we go again...

I hope the techies aren't sitting in the pub.....

 

[old.moriath]

Well i sent mine and it shows the same as all the otther

Wanadoo sucks big time



But they french so what u expect

Sort it pls

 

[Frieza]

I can't play what so ever have sent my mail but this is ridiculous.

Since patch i've had so many LD's and now tonight i'm just lagged out

 

[Jupitus]

I'm sure that Kemor is asking for these to gather evidence to shove up Wanadoo's nose - keep em coming and we'll give them a right good nosebleed

 

[old.drac]

nose?...try another orifice...and make em go in sideways.

 

[Jupitus]

*coughs*

I was kinda being nice about it, but I do agree!

 

[old.smurflord]

OMG!!!!

I get an average of 60 - 80ms per hop on my tracert from my PC and out across europe. That is until I get to wanadoo's router where it jumps to 1500ms!!!!
I think we know where the fault lies.

:touch: Wanadoo - wake up or there will be much ing.

 

[kemor]

Alright folks, will do a bigger news tomorrow once we have more news but, after shoving..err..showing the problem and the tracerts to the opentransit folks, we got some different news :

ps : bear with me, i'm not a network monkey

"for reasons still unknown (but they are actively looking for them), it seems that there is a problem on one of the Opentransit router. However, this problem is a bit weird and could be related to a more important problem affecting the european internet network.

As you may know, many companies dealing with internet connection went/are going bankrupt lately and a lot of people are re-routed on the same routers causing what some call a "bottleneck".

The problem we have lately with connection seems to be "upstream" (like at the top of a river, some rocks blocking the flow of the water, slow the whole current). Many ISPs are trying to rout their clients to opentransit network, and they are slowing down everyone.

Some players can play without ANY problem, with a perfect ping to our servers, while some are stopped on Opentransit router.

Really cannot give more information right now, but hopefully the night will bring us more details....

Stay tuned, more details tomorrow...

 

[SFXman]

Mwahha... I am on with a perfect ping. After 426 attemps I had an average below 70 (66 or 67) and a PL of 0%. Lovely.

 

[Orin Askhammare]

Thanks Kemor. I can confirm the upstream thing as my lag seemed to be exclusively on commands I was giving, not on other things happening around me in game. I could see health bars come down fine, but had to wait many seconds for my character to start casting after pressing the button.

 

[old.Tohtori]

Ok..shiiit...waht the hell is that page and how do i get ping pings up the wazoo or the annihilation of anything?

 

[Asonn]

Thanks for the info Kemor

[bump]

 

[Blood]

is it too much to ask that goa tries to get a system in place that does these traceroutes when someone goes LD?

if GOA did a traceroute from their end to client on connect, and then another on disconnect they could very quickly begin to troubleshoot these things themselves? (feed it to a database with relation to the account database, and they would have a powerfull tool to help us customers pinpointing where the LD's occur)

i gave up counting after the relic raid where i went LD 5 times, but in total i must have gone LD 15 times today... if i had been xping that would = a huge lost of gold to conpoints, and negative xp gain..

i'd love to help by pingplotting myself, but alas my entire connection is reset when the daoc connection ld's (i have a special routing table myself, and allthou i've tried disable "EnableDeadGWDetect" my win2k still insist's on killing my netcard for 45 secs) so i cant do your job for you.

<-- paying customer (got another 3 month invoice to show for it)

 

[Zebarmy]

I had a feeling it'd be related to the recent KPNQwest/Ebone issues

The problem with doing a tracert from goa when someone goes LD as far as I can see Blood, is that a lot of users in the UK (and probably elsewhere, but I'm hardly an expert on such demographics ), use dialup PSTN and ISDN connections with a set time cut off. You'd be gathering a tonne of useless information, just because someone forgot what time it is whilst they're exping in a good group or RvRing (Dare I even mention crafting? ). Also with mass LD's I doubt you'd be able to collect decent information either, for one because goa would suddenly have to run 500 (or more/or less ) seperate tracert processes, and secondly with the number of people all trying to reconnect again, the information gathered would not be conclusive. Pings/Tracerts are one of the lowest priority packets on the internet, and would be dropped in favour of sending all the client game information back to the (re)connecting players.

Therefore in my eyes, it makes sense that the players theirselves do the tracert's, because they will know more about their connection than Goa could possibly ever do (i.e. Crap I'm LD - Oh my 2 hour cut off... )

Just my two copper

Zebby

 

[SFXman]

If KPNQwest had problems I would be in major shit tbh... so I really really doubt that. They are somehow directly related to my ISPs cable service (remember that from somewhere)...

 

[Zebarmy]

Hate to break it to you mate, but KPNQwest have gone tits up. They're continuing running their network with a skeleton crew, but there really isn't much hope for em. I believe they've gone into receivership.

More and more major parties are moving their traffic away from KPNQwest, which is why we're seeing congestion If your ISP is still using them, then in theory, you should have a nice non laggy connection > Less contention, but how long this will last is anyones guess

Zebby

 

[nerve]

Skeleton crew no way

And hate to break it to you, but that skeleton crew left the premises a few days ago. If you don't believe me, it was on Belgian national news. The Ebone network from KPNQwest has been totally disabled. That's 30% of European internet capacity down the drain -> I guess I don't have to draw a picture where all our problems are coming from (yup rerouting over routers that have now become bottlenecks).

Have a lagless day,

Nerve

4x Mercenary/Excalibur

 

[Zebarmy]

Heh, oh dear

(makes mental note to keep up with topical subjects that he should know about )

Zebby

 

[Sarnat]

I guess all those idiots on IRC who start screaming how GOA's servers suck and relic raids cause LDs should read this thread.

If a raid crashes a server/zone, it won't crash the other servers/zones. It will not crash the web server or patch server either. There just seems to be a relic raid going on every day from 18:00 to midnight on prydwen atm So when lag starts somewhere during the evening, everyone starts to yell 'relic raid lag' etc even though last raid with same amounts of people worked just great.

 

[Brannor McThife]

Hmm, who ISN'T having problems when everyone else goes LD? Speak up now.

I think it may help GOA as well if they know who's works, and who's doesn't.

<prepares to make a dash for the door>
Maybe, if you DO NOT go LD when everyone else does, could you perhaps still exit the game, and do that TRACERT stuff, and submit it to Kemor with a [WORKING TRACERT] subject header?

I think they'll appreciate it. And I know I will.

<slams door and runs off into the street.>



-G

 

[kemor]

As updated in the threads...

UPDATE
Now, i need traceroutes from players who do NOT have problems while their friends are dropping like flies all around the realms.
Same procedure, with just a different subject : [CLEAN TRACERT]

Thanks.

 

[Brannor McThife]

<opens the door>
I think WORKING TRACERT sounds nicer.
<slams door again>

-G

 

[DeSlisser]

Tracert

Got this, no idea what the stars mean:

1 core-7206-fe-3.netsys.com (199.201.232.2) 0.536 ms 0.476 ms 0.429 ms
2 core-savvis.netsys.com (209.176.21.1) 18.650 ms 9.564 ms 6.042 ms
3 at-1-2-0707.usnycm-j20c.savvis.net (64.240.136.57) 6.374 ms 6.545 ms 6.408 ms
4 500.POS1-1.GW8.NYC1.ALTER.NET (65.194.72.53) 6.369 ms 14.117 ms 23.575 ms
5 197.at-5-0-0.XR2.NYC1.ALTER.NET (152.63.20.2) 7.352 ms 15.389 ms 6.818 ms
6 208.at-1-0-0.XR4.NYC4.ALTER.NET (152.63.29.70) 7.185 ms 7.193 ms 7.476 ms
7 188.ATM7-0.GW7.NYC4.ALTER.NET (152.63.25.109) 7.511 ms 36.705 ms 53.022 ms
8 opentransit-gw.customer.alter.net (157.130.60.218) 7.675 ms 11.354 ms 8.462 ms
9 P10-0.NYKCR2.New-york.opentransit.net (193.251.241.237) 7.185 ms 6.977 ms 7.517 ms
10 P4-0.PASCR1.Pastourelle.opentransit.net (193.251.241.133) 83.210 ms 89.595 ms 104.368 ms
11 P4-0.BAGCR3.Bagnolet.opentransit.net (193.251.241.118) 102.027 ms 84.181 ms 107.306 ms
12 P1-0.BAGBB1.Bagnolet.opentransit.net (193.251.241.145) 89.755 ms 82.913 ms 83.821 ms
13 P2-0-0.BAGAR1.Bagnolet.opentransit.net (193.251.128.102) 83.538 ms 89.099 ms 84.192 ms
14 WanadooPortails.GW.opentransit.net (193.251.251.214) 83.865 ms 84.352 ms 84.361 ms
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * *

Did this on the following site: http://www.netsys.com/tools.html