Author Topic: Package loss 60% (Read 1970 times)

Apar · « **Reply #30 on:** May 05, 2002, 10:59:45 AM »

Cable here, 50% PL,

Wilbus · « **Reply #31 on:** May 05, 2002, 11:02:49 AM »

Sweden, ADSL, 400-800 ping to HTC server. The problem is at HTC, I have 150ms to the server before HTC then it goes up to nearly 800 when it hits HTC.

Alino · « **Reply #32 on:** May 05, 2002, 11:22:22 AM »

Problem seems to be the same in Germany, Netherlands, Spain and Sweden. It must be a HTCproblem but no react from them.

Vector · « **Reply #33 on:** May 05, 2002, 11:46:47 AM »

..and Finland as well. Never had this bad conx. 450-750 ms (usually solid 180ms) and discos that costs 70 perks / each.

Skuzzy · « **Reply #34 on:** May 05, 2002, 12:00:22 PM »

Quote

Originally posted by Wilbus
Sweden, ADSL, 400-800 ping to HTC server. The problem is at HTC, I have 150ms to the server before HTC then it goes up to nearly 800 when it hits HTC.

Wilbus, again, it is an asynchrounous route. This is not something HTC can control. It very well could be the return path, but do not take the last hop jump as a HTC problem. They cannot control how data is routed over the Internet.

Alino · « **Reply #35 on:** May 05, 2002, 12:12:33 PM »

Quote

Originally posted by Skuzzy

Wilbus, again, it is an asynchrounous route. This is not something HTC can control. It very well could be the return path, but do not take the last hop jump as a HTC problem. They cannot control how data is routed over the Internet.

Hi Skuzzy, the Problem is in 5 Countries in europe. It`s Germany,Netherlands,Sweden,Spain and Finnland. Who the hell if not HTC can find out the Problem. ME NOT.

Wilbus · « **Reply #36 on:** May 05, 2002, 01:21:04 PM »

Clumpsy said Skuzzy, what I ment is that it's at HTC or nearby, it's not our connections that is for sure, unless we all happen to have a pretty bad day down here in europe. Connection is sollid untill it gets to the HTC server.

Skuzzy · « **Reply #37 on:** May 05, 2002, 01:26:50 PM »

It is not about who can find out the problem Alino. Depending on where the problem lies, there may be nothing HTC can do but wait it out.
I do not know exactly where the problem is. Looking at all the trace data I have seen, it mostly appears to be a return data path issue. I do not have access to the network where the MA is hosted, so it is difficult for me to say.
However, I have been through issues like this one, and it is not usually something that is resolved quickly.

Also, while many are complaining, very few are posting any helpful data. I see a ton of complaints in the Internet Forum, but only 2 traceroutes. Without data, there is absolutely nothing anyone can do. So any complaints without any trace data is pretty much worthless and would/should be ignored.
You are one of the few that has posted any trace data Alino, and it is appreciated. This is what is needed to find the problem.

tofri · « **Reply #38 on:** May 05, 2002, 02:04:15 PM »

Quote

Originally posted by Skuzzy

Also, while many are complaining, very few are posting any helpful data. I see a ton of complaints in the Internet Forum, but only 2 traceroutes. Without data, there is absolutely nothing anyone can do. So any complaints without any trace data is pretty much worthless and would/should be ignored.

Why should I bother stuffing my post with trace routes, when every other trace route that is already posted, shows the same.
Up to theplanet-2 erverything is fine.
I do not know and I cant test it, if there are any switches and routers between theplanet-2 and tpl.HTC .
But if there is any rerouting on routers or more likely balancers, then it happens between the last two hops.
The ISP had problem with the bandwith??
It seems, that they have solved the problem in a very unorthodox way.
They mask out a certain group of IP-adresses and let them run into a bottleneck.

Skuzzy, please take the Main-server back as soon as possible. Even with a T1-Connection it was never that bad. We had lags, but we could play at least.

Here is my actual trace route

NeoTrace Version 2.12a - Shareware (2-1-2000)
Destination: tpl.hitechcreations.com

-#--------------Node Name---------------IP Address------Location-----------RT*--High---Low---Avg-Tot---D-Who
1 tom 192.168.1.1 50.000N, 6.000E 0 0 0 0 1 0
2 217.5.98.141 Unknown 54 54 54 54 1 0
3 217.237.156.94 Unknown 53 53 53 53 1 0
4 vienna-gw12.usa.net.dtag.de 62.156.131.182 Unknown 150 150 150 150 1 0
5 dt-gw.wswdc.ip.att.net 194.25.6.234 38.908N, 77.017W 147 147 147 147 1 0
6 gbr4-p50.wswdc.ip.att.net 12.123.9.54 38.908N, 77.017W 146 146 146 146 1 0
7 gbr4-p30.attga.ip.att.net 12.122.2.225 33.750N, 84.383W 158 158 158 158 1 0
8 gbr3-p20.dlstx.ip.att.net 12.122.2.89 32.783N, 96.800W 171 171 171 171 1 0
9 gbr5-p60.dlstx.ip.att.net 12.122.5.73 32.783N, 96.800W 172 172 172 172 1 0
10 gar2-p360.dlstx.ip.att.net 12.123.17.25 32.783N, 96.800W 172 172 172 172 1 0
11 gar1-p3100.dlrtx.ip.att.net 12.123.196.101 Unknown 172 172 172 172 1 0
12 12.119.136.22 Unknown 172 172 172 172 1 0
13 border3.ge4-1-bbnet2.ext1.dal.pnap.net 216.52.191.83 Unknown 171 171 171 171 1 0
14 theplanet-2.border3.003.dal.pnap.net 216.52.189.222 Unknown 173 173 173 173 1 0
15 tpl.hitechcreations.com 216.234.247.240 32.934N, 97.095W 360 360 360 360 1 0
--------------------------------------------------------------------------------------------------------
*All times in milliseconds (ms), D=Dropped packets
------------------------------------------------------------------------------------------------------------
May 5, 2002 21:2:23
NeoTrace Copyright ©1997-1999 NeoWorx inc
http://www.neoworx.com

Duedel · « **Reply #39 on:** May 05, 2002, 02:18:57 PM »

Skuzzy appreciate ur work to explain the whole problem. Thx and

Target Name: tpl.hitechcreations.com
IP: 216.234.247.240
Date/Time: 05.05.2002 21:23:40

1 55 ms 69 ms 110 ms 55 ms [217.5.98.61]
2 55 ms 68 ms 82 ms 69 ms [217.237.153.194]
3 193 ms 151 ms 165 ms 164 ms Vienna-gw12.USA.net.DTAG.DE [62.156.131.182]
4 165 ms 151 ms 165 ms 151 ms dt-gw.wswdc.ip.att.net [194.25.6.234]
5 165 ms 164 ms 151 ms 151 ms gbr4-p50.wswdc.ip.att.net [12.123.9.54]
6 179 ms 165 ms 192 ms 164 ms gbr4-p30.attga.ip.att.net [12.122.2.225]
7 192 ms 178 ms 192 ms 192 ms gbr3-p10.dlstx.ip.att.net [12.122.3.37]
8 193 ms 192 ms 193 ms 192 ms gbr6-p60.dlstx.ip.att.net [12.122.5.81]
9 192 ms 179 ms 178 ms 178 ms gar2-p370.dlstx.ip.att.net [12.123.17.29]
10 178 ms 178 ms 179 ms 179 ms gar1-p3100.dlrtx.ip.att.net [12.123.196.101]
11 179 ms 179 ms 219 ms 178 ms [12.119.136.22]
12 192 ms 398 ms 261 ms 192 ms border3.ge3-1-bbnet1.ext1.dal.pnap.net [216.52.191.22]
13 193 ms 179 ms 248 ms 192 ms theplanet-2.border3.003.dal.pnap.net [216.52.189.222]
14 * * * * [-]
15 371 ms * 385 ms 357 ms [216.234.247.240]

Alino · « **Reply #40 on:** May 05, 2002, 02:26:05 PM »

Ok Skuzzy here is one actually

1 69 ms 55 ms 1442 ms 217.5.98.143
2 55 ms 55 ms 55 ms 217.237.156.110
3 151 ms 137 ms 137 ms Vienna-gw12.USA.net.DTAG.DE [62.156.131.182]
4 1071 ms 151 ms 151 ms dt-gw.wswdc.ip.att.net [194.25.6.234]
5 151 ms 151 ms 151 ms gbr4-p50.wswdc.ip.att.net [12.123.9.54]
6 151 ms 165 ms 165 ms gbr4-p30.attga.ip.att.net [12.122.2.225]
7 179 ms 178 ms 179 ms gbr3-p20.dlstx.ip.att.net [12.122.2.89]
8 178 ms 179 ms 178 ms gbr6-p60.dlstx.ip.att.net [12.122.5.81]
9 178 ms 179 ms 165 ms gar2-p370.dlstx.ip.att.net [12.123.17.29]
10 178 ms 179 ms 178 ms gar1-p3100.dlrtx.ip.att.net [12.123.196.101]
11 178 ms 165 ms 164 ms 12.119.136.22
12 178 ms 179 ms 178 ms border3.ge3-1-bbnet1.ext1.dal.pnap.net [216.52
91.22]
13 178 ms 179 ms 178 ms theplanet-2.border3.003.dal.pnap.net [216.52.1
.222]
14 357 ms 370 ms * 216.234.247.240
15 370 ms 357 ms * 216.234.247.240
16 357 ms 385 ms * 216.234.247.240
17 * * * Zeitüberschreitung der Anforderung.
18 357 ms 370 ms 357 ms 216.234.247.240

Like ToFri said take it back soon please.

Skuzzy · « **Reply #41 on:** May 05, 2002, 03:32:02 PM »

Ah hah! Anyone notice that all the traces start from the same Class A address of 217.x.x.x?

Looks like we got a route problem going on here. I did a trace back to one of the posters in the Internet forum and the data going back to him was getting hung up in a router on his local ISP (telia).
Looks like someone has propogated a bad route table for the 217.x.x.x IP address block.

The owner of that Class A block is Deutsche Telekom AG. They are apparently propagating bad route tables for this block.

This is why we need traceroutes. It is not neccessarily the times in the routes, it is also the addresses. When the routes are asynchronous, there is no way for you to know where the problem really is.

Now, getting it fixed could be a problem. I have not had much luck in getting overseas issues corrected. Anyone familiar with the above company?

Skuzzy · « **Reply #42 on:** May 05, 2002, 03:34:26 PM »

tofri, Duedel and Alino, tracing back to you, the data stops at the following router - 217.237.156.109, which is a dtag.de router.

Alino · « **Reply #43 on:** May 05, 2002, 04:01:43 PM »

Skuzzy that`s good to know. I will phone the Telekom and confront them with that.

Thank`s a lot for your work man.

Alino · « **Reply #44 on:** May 05, 2002, 04:14:21 PM »

Skuzzy here i am again.
I phoned to the Telekom. And told them of that 217...router
They asked from who i got that information and asked for your mailadress is ok to give it to them ??

They will check it out tomorrow. So i can`t do anything more then wait now.