|
|
Hi
As some will know Coms seem pretty terrible and have been suffering with packetloss on an evening. Got to the point im looking to move on, but before i do i just wanted to prove it wasnt my issue, and was a coms issue. So i decided to run a pingpath test so see where the packetloss was. Can someone confirm if what i think im seeing
Tracing route to bbc.co.uk [212.58.246.104]
over a maximum of 30 hops:
0 User-PC [192.168.0.4]
1 192.168.0.1
2 * 109.224.191.52
3 149.6.9.170
4 gi8-10.mag01.lhr01.atlas.cogentco.com [149.6.9.169]
5 te0-5-0-5.rcr21.lhr01.atlas.cogentco.com [130.117.0.90]
6 be2036.ccr42.lon13.atlas.cogentco.com [130.117.48.153]
7 be2494.ccr22.lon01.atlas.cogentco.com [154.54.39.129]
8 level3.lon01.atlas.cogentco.com [130.117.15.246]
9 * * *
Computing statistics for 200 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 User-PC [192.168.0.4]
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.0.1
25/ 100 = 25% |
2 36ms 25/ 100 = 25% 0/ 100 = 0% 109.224.191.52
2/ 100 = 2% |
3 37ms 27/ 100 = 27% 0/ 100 = 0% 149.6.9.170
0/ 100 = 0% |
4 49ms 28/ 100 = 28% 1/ 100 = 1% gi8-10.mag01.lhr01.atlas.cogentco.com [149.6.9.169]
0/ 100 = 0% |
5 35ms 28/ 100 = 28% 1/ 100 = 1% te0-5-0-5.rcr21.lhr01.atlas.cogentco.com [130.117.0.90]
0/ 100 = 0% |
6 36ms 32/ 100 = 32% 5/ 100 = 5% be2036.ccr42.lon13.atlas.cogentco.com [130.117.48.153]
0/ 100 = 0% |
7 37ms 27/ 100 = 27% 0/ 100 = 0% be2494.ccr22.lon01.atlas.cogentco.com [154.54.39.129]
2/ 100 = 2% |
8 48ms 29/ 100 = 29% 0/ 100 = 0% level3.lon01.atlas.cogentco.com [130.117.15.246]
Trace complete.
109.224.191.52 is the coms gateway and hop 3 is Cogentco. So am i right in thinking that the packetloss is happening at the gateway as seen in hop 2 (25% loss) and the other packetloss after that is probably a result of the drop packets at the gateway?
So in others words its a coms issue?
|
|
|
|
|
|
|
Microsoft Windows [Version 6.1.7601]
1 <1 ms <1 ms <1 ms 192.168.0.1
2 36 ms 41 ms 35 ms 109.224.191.52
3 36 ms 35 ms 44 ms 149.6.9.170
4 35 ms 45 ms 36 ms gi8-10.mag01.lhr01.atlas.cogentco.com [149.6.9.169]
5 36 ms 45 ms 35 ms te0-5-0-5.rcr21.lhr01.atlas.cogentco.com [130.117.0.90]
6 37 ms 54 ms 37 ms be2036.ccr42.lon13.atlas.cogentco.com [130.117.48.153]
7 37 ms 45 ms 38 ms be2494.ccr22.lon01.atlas.cogentco.com [154.54.39.129]
8 * 44 ms 45 ms level3.lon01.atlas.cogentco.com [130.117.15.246]
9 * * * Request timed out.
10 44 ms 44 ms 62 ms ae-120-3506.edge4.London1.Level3.net [4.69.166.5]
11 36 ms 37 ms 38 ms BBC-TECHNOL.edge4.London1.Level3.net [212.113.14.222]
12 * * * Request timed out.
13 37 ms 37 ms 43 ms ae0.er01.telhc.bbc.co.uk [132.185.254.109]
14 38 ms 38 ms 45 ms 132.185.255.149
15 43 ms 41 ms 37 ms fmt-vip72.telhc.bbc.co.uk [212.58.244.18]
Trace complete.
C:\Users\User>
Packetloss only occurs around 5-6pm and goes away around 11-12pm. Outside of those times everything is perfect. Weirdly thats the time when coms throttle p2p
Edited by bobble_bob (Mon 20-Oct-14 19:16:33)
|
|
Register (or login) on our website and you will not see this ad.
|
|
|
Nothing dramatic to be seen in the trace route !
|
|
|
Yea packetloss comes and goes. When i ran that test it wasnt there going on my bqm. It is there though as seen in the pathping, just wondering where.
http://www.thinkbroadband.com/ping/share/0cb991750c5...
Thats my live graph. Was using p2p around 4 hence the high pings, but the packetloss is visable later
Edited by bobble_bob (Mon 20-Oct-14 19:35:31)
|
|
|
Might be hop 2 might not. If it was hop 2 all ongoing hops should be as bad too
Need to ping each hip for longer to show its in coms network and not just a bad set if peering links
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
not just a bad set if peering links
Sorry not quite sure what that means. Is it possible this issue is at my end? Whether its coms, or there supplies im not too fussed aslong as i know ive done all i can and im not the cause.
The packetloss between my router and the gateway is worrying, does this suggest congestion at the exchange?
Edited by bobble_bob (Mon 20-Oct-14 19:55:03)
|
|
|
May mean loss over the dsl, at the exchange, wholesalers network or coms network
Traffic is often handled in a way that you do not see exchange and other links
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
|
The fact the packet loss is during peak hours around early/late evening suggests its congestion somewhere on the network, where it is like you say difficult to pinpont.
Although others have complained of packetloss and slow browsing while being with coms, they fact they throttle p2p/usenet to within an inch of its life i dont know why there would be congestion. Then again i dont know how they have their network set up
|
|
|
|
What effects are you seeing of packetloss? As far as I know an ICMP ping isn't going to tell you anything definitive as this will be the first traffic a router drops if there are any performance issues. Pings aren't a great test. Are you seeing real world performance issues with downloads and the like?
Also, have you done speedtests on TBB at the time so that we can compare the x1 and x6 to see if there is likely to be congestion?
|
|
|
|
Im seeing slower browsing although streaming seems to be uneffected. Ive had tests that show 90% packet loss and streaming has been fine, but then webpages have been slower (this could be a result of coms treating streaming with a higher priority)
Ive run the speedtest yea and the x1 is slower than the x6 when it happens, but download a raw file i get full speeds
|
|
|
|
Well me and someone else on coms here ran bqm's tonight, both on different exchanges in the country and both started getting packetloss at exactly the same time. So cant think what else it can be other than a coms issue
|
Pages in this thread: 
Print Thread
bobble_bob
