Trouble with mumble

[#1]

tracert to the 3 servers mentioned in this and other posts. 7:15pm
C:\>tracert Sweden.ventriloservers.biz
Tracing route to Sweden.ventriloservers.biz [188.126.79.15]
over a maximum of 30 hops:
  1    29 ms    98 ms    99 ms  dsldevice.lan [192.168.1.254]
  2    12 ms    15 ms    12 ms  lo0-central1.pcl-ag06.plus.net [195.166.130.46]
  3    11 ms    11 ms    12 ms  gi7-6-611.pcl-gw01.plus.net [84.93.254.129]
  4    12 ms    12 ms    12 ms  po4.pcl-gw02.plus.net [212.159.1.129]
  5    11 ms    11 ms    71 ms  ae3.pcl-cr02.plus.net [195.166.129.42]
  6    19 ms    11 ms    12 ms  ae1.ptw-cr02.plus.net [195.166.129.2]
  7    16 ms    11 ms    12 ms  po2.ptw-gw02.plus.net [195.166.129.39]
  8  662 ms  690 ms    *    teln.lon.portlane.net [195.66.225.159]
  9    *    1100 ms  824 ms  te-2-1.sto3.se.portlane.net [80.67.4.134]
10    68 ms    *      644 ms  sweden.clanwarz.com [188.126.79.15]
Trace complete.
C:\>tracert mumble.pandemic-legion.com
Tracing route to mumble.pandemic-legion.com [46.253.205.246]
over a maximum of 30 hops:
  1    5 ms    98 ms    98 ms  dsldevice.lan [192.168.1.254]
  2    13 ms    13 ms    13 ms  lo0-central1.pcl-ag06.plus.net [195.166.130.46]
  3    11 ms    11 ms    11 ms  gi7-6-611.pcl-gw01.plus.net [84.93.254.129]
  4    12 ms    11 ms    12 ms  ae3.pcl-cr01.plus.net [195.166.129.40]
  5    12 ms    12 ms    12 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
  6    11 ms    12 ms    12 ms  ae2.ptw-cr02.plus.net [195.166.129.5]
  7    12 ms    11 ms    11 ms  po2.ptw-gw02.plus.net [195.166.129.39]
  8  708 ms  692 ms  709 ms  teln.lon.portlane.net [195.66.225.159]
  9  731 ms  701 ms  756 ms  po-10.sto1.se.portlane.net [80.67.4.128]
10  690 ms  731 ms  635 ms  vl-3305-adminor-demarc.sto1.se.portlane.net [80.67.0.226]
11  774 ms  755 ms  764 ms  mumble.pandemic-legion.com [46.253.205.246]
Trace complete.
C:\>tracert thepiratebay.se
Tracing route to thepiratebay.se [194.71.107.15]
over a maximum of 30 hops:
  1    48 ms    99 ms    99 ms  dsldevice.lan [192.168.1.254]
  2    12 ms    12 ms    12 ms  lo0-central1.pcl-ag06.plus.net [195.166.130.46]
  3    12 ms    12 ms    11 ms  gi7-6-611.pcl-gw01.plus.net [84.93.254.129]
  4    12 ms    12 ms    11 ms  ae3.pcl-cr01.plus.net [195.166.129.40]
  5    12 ms    12 ms    24 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
  6    11 ms    11 ms    12 ms  ae2.ptw-cr02.plus.net [195.166.129.5]
  7    12 ms    11 ms    12 ms  po2.ptw-gw02.plus.net [195.166.129.39]
  8  700 ms  722 ms  670 ms  teln.lon.portlane.net [195.66.225.159]
  9  712 ms    *      757 ms  te-1-1.sto2.se.portlane.net [80.67.4.137]
10    *        *        *    Request timed out.
11    *        *        *    Request timed out.
12  709 ms  758 ms    *    thepiratebay.piratpartiet.se [194.14.56.2]
13    *    ^C
I don't use the first 2. It's the last one giving me probs

[#2]

Morning folks.
We have now shut-down peering with Portlane over LINX. This has has the effect of routing the data over Level3.

1 gi4-46.ptp-cr01.plus.net (84.93.216.81) 8.066 ms 5.828 ms 7.432 ms
2 gi1-22.pcl-gw01.plus.net (84.93.224.49) 14.480 ms 10.871 ms 9.618 ms
3 ae3.pcl-cr01.plus.net (195.166.129.40) 11.520 ms 10.742 ms 12.233 ms
4 xe-11-1-0.edge3.london2.level3.net (212.187.201.209) 12.732 ms 9.642 ms 11.036 ms
5 ae-32-54.ebr2.london2.level3.net (4.68.117.126) 16.024 ms 16.394 ms
ae-32-56.ebr2.london2.level3.net (4.68.117.190) 19.504 ms
6 ae-3-3.ebr1.london1.level3.net (4.69.141.189) 13.854 ms 15.642 ms 13.216 ms
7 ae-56-111.csw1.london1.level3.net (4.69.153.114) 12.638 ms
ae-57-112.csw1.london1.level3.net (4.69.153.118) 17.876 ms
ae-58-113.csw1.london1.level3.net (4.69.153.122) 16.690 ms
8 ae-1-51.edge3.london1.level3.net (4.69.139.73) 11.182 ms 14.829 ms 12.824 ms
9 gblx-level3-50g.london1.level3.net (4.68.110.158) 12.513 ms 22.585 ms 10.228 ms
10 ae6.scr4.lon3.gblx.net (67.17.106.150) 11.451 ms 10.688 ms 10.634 ms
11 ae5.csr2.arn3.gblx.net (67.16.146.113) 35.240 ms 35.702 ms 41.096 ms
12 te-4-4-gblx.sto1.se.portlane.net (209.130.172.178) 46.628 ms 46.460 ms 50.479 ms
13 po-10.sto3.se.portlane.net (80.67.4.129) 46.147 ms 46.032 ms 48.000 ms

So packet loss looks to be initially resolved, but we are clearly not in peak yet which will be the teller. Data is still going via Portlane, which is likely normal for the upstream provider of choice for the services you are using.
We can see higher latency in using Level3, but this is the trade-off unfortunately on making this work. We need to switch back to peering to resolve this, but Portlane appear to be ignoring us on the matter.
Can you folks let us know if you see anything during peak?

[#3]

yeh will do a tracert about 6:30 see how its looking.

[#4]

Microsoft Windows [Version 6.1.7601]
Copyright (c) 2009 Microsoft Corporation.  All rights reserved.
C:\Users\Paul>tracert sweden.ventriloservers.biz
Tracing route to sweden.ventriloservers.biz [188.126.79.15]
over a maximum of 30 hops:
  1    1 ms    <1 ms    <1 ms  local.gateway [10.0.0.2]
  2    45 ms    46 ms    42 ms  lo0-central5.pcl-ag02.plus.net [195.166.130.89]
  3    44 ms    42 ms    44 ms  gi7-21-252.pcl-gw02.plus.net [84.93.255.99]
  4  237 ms  241 ms  231 ms  po4.pcl-gw01.plus.net [212.159.1.128]
  5    44 ms    61 ms    42 ms  ae3.pcl-cr01.plus.net [195.166.129.40]
  6    42 ms    42 ms    44 ms  xe-11-1-0.edge3.London2.Level3.net [212.187.201.
209]
  7    46 ms    39 ms    70 ms  ae-32-56.ebr2.London2.Level3.net [4.68.117.190]
  8    39 ms    39 ms    39 ms  ae-3-3.ebr1.London1.Level3.net [4.69.141.189]
  9    42 ms    38 ms    40 ms  ae-59-114.csw1.London1.Level3.net [4.69.153.126]
10    45 ms    43 ms    41 ms  ae-1-51.edge3.London1.Level3.net [4.69.139.73]
11    42 ms    42 ms    47 ms  gblx-level3-50g.London1.Level3.net [4.68.110.158
]
12    41 ms    41 ms    42 ms  ae6.scr4.LON3.gblx.net [67.17.106.150]
13    76 ms    71 ms    71 ms  ae5.CSR2.ARN3.gblx.net [67.16.146.113]
14    74 ms    74 ms    76 ms  te-4-4-gblx.sto1.se.portlane.net [209.130.172.17
8]
15  123 ms  228 ms  207 ms  po-10.sto3.se.portlane.net [80.67.4.129]
16    71 ms    71 ms    71 ms  sweden.clanwarz.com [188.126.79.15]
Trace complete.
C:\Users\Paul>
done at 18:09

[#5]

Tracert @ 6:45pm
C:\>tracert Sweden.ventriloservers.biz
Tracing route to Sweden.ventriloservers.biz [188.126.79.15]
over a maximum of 30 hops:
  1    48 ms    98 ms    97 ms  dsldevice.lan [192.168.1.254]
  2    13 ms    13 ms    17 ms  lo0-central1.pcl-ag06.plus.net [195.166.130.46]
  3    12 ms    11 ms    11 ms  gi7-6-611.pcl-gw01.plus.net [84.93.254.129]
  4    12 ms    12 ms    10 ms  ae3.pcl-cr01.plus.net [195.166.129.40]
  5    12 ms    12 ms    13 ms  xe-11-1-0.edge3.London2.Level3.net [212.187.201.209]
  6    20 ms    17 ms    15 ms  ae-32-56.ebr2.London2.Level3.net [4.68.117.190]
  7    12 ms    12 ms    14 ms  ae-3-3.ebr1.London1.Level3.net [4.69.141.189]
  8    20 ms    22 ms    12 ms  ae-58-113.csw1.London1.Level3.net [4.69.153.122]
  9    11 ms    11 ms    13 ms  ae-1-51.edge3.London1.Level3.net [4.69.139.73]
10    12 ms    21 ms    12 ms  gblx-level3-50g.London1.Level3.net [4.68.110.158]
11    13 ms    12 ms    12 ms  ae6.scr4.LON3.gblx.net [67.17.106.150]
12    36 ms    36 ms    36 ms  ae5.CSR2.ARN3.gblx.net [67.16.146.113]
13    44 ms    43 ms    44 ms  te-4-4-gblx.sto1.se.portlane.net [209.130.172.178]
14    43 ms    44 ms    43 ms  po-10.sto3.se.portlane.net [80.67.4.129]
15    43 ms    43 ms    44 ms  sweden.clanwarz.com [188.126.79.15]
Trace complete.
C:\>tracert mumble.pandemic-legion.com
Tracing route to mumble.pandemic-legion.com [46.253.205.246]
over a maximum of 30 hops:
  1    76 ms    98 ms    98 ms  dsldevice.lan [192.168.1.254]
  2    14 ms    16 ms    12 ms  lo0-central1.pcl-ag06.plus.net [195.166.130.46]
  3    15 ms    13 ms    11 ms  gi7-6-611.pcl-gw01.plus.net [84.93.254.129]
  4    12 ms    12 ms    12 ms  ae3.pcl-cr01.plus.net [195.166.129.40]
  5    13 ms    12 ms    12 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
  6    13 ms    12 ms    12 ms  te9-4.ptn-gw01.plus.net [195.166.129.33]
  7    13 ms    14 ms    14 ms  195.99.126.144
  8  185 ms  200 ms  205 ms  195.66.224.243
  9    78 ms    61 ms    58 ms  po41-20g-r85.cr0-r86.hy-sto.se.p80.net [82.96.1.161]
10    46 ms    47 ms    46 ms  telelocation-hy-demarc0.sto.se.p80.net [82.96.53.70]
11    50 ms    49 ms    48 ms  91.142.184.10
12    47 ms    45 ms    45 ms  mumble.pandemic-legion.com [46.253.205.246]
Trace complete.
C:\>tracert thepiratebay.se
Tracing route to thepiratebay.se [194.71.107.15]
over a maximum of 30 hops:
  1    38 ms    98 ms  100 ms  dsldevice.lan [192.168.1.254]
  2    28 ms    32 ms    26 ms  lo0-central1.pcl-ag06.plus.net [195.166.130.46]
  3    12 ms    12 ms    11 ms  gi7-6-611.pcl-gw01.plus.net [84.93.254.129]
  4    15 ms    12 ms    12 ms  ae3.pcl-cr01.plus.net [195.166.129.40]
  5    13 ms    12 ms    11 ms  ae1.ptw-cr01.plus.net [195.166.129.0]
  6    13 ms    12 ms    12 ms  te9-4.ptn-gw01.plus.net [195.166.129.33]
  7    12 ms    13 ms    13 ms  t2a1-ge2-0-0.uk-lon1.eu.bt.net [166.49.211.45]
  8    12 ms    12 ms    13 ms  195.66.224.243
  9    45 ms    45 ms    45 ms  po41-20g-r85.cr0-r86.hy-sto.se.p80.net [82.96.1.161]
10    44 ms    44 ms    46 ms  robtex-kn1-demarc0.cr0-r84.kn1-sto.se.p80.net [93.158.74.238]
11    *        *        *    Request timed out.
12    45 ms    46 ms    45 ms  thepiratebay.piratpartiet.se [194.14.56.2]
13    *    ^C

[#6]

That's looking a bit better!

[#7]

is the extra bandwidth with Level3 in relation to this btw? (im assuming significant traffic was going over the portlane link before...?

Quote

What does the work involve? We're continuing work from recent weeks to increase the capacity of our transit links with Level3.

Or was this on the cards anyway..?
Regards

[#8]

The capacity on the LINX224 peering is the same as our Level3 transit per connection.  The issue lies with either portlanes capacity on the LINX or their hardware attached.

[#9]

looks good, I've been using mumble direct (no proxy) for the last couple of days and had very few issues.
I keep forgetting or working through peak time, but I will remember tomorrow. since its good friday it should be impressively congested and a good time to nab some stats.  

[#10]

Hop IP Address Host Name Response Time
=== ========== ========= =============
1 192.168.1.254 dsldevice.lan 87 ms
2 195.166.128.228 lo0-central3.ptw-ag01.plus.net 16 ms
3 84.92.6.112 gi1-5-131.ptw-gw01.plus.net 16 ms
4 195.99.126.146 20 ms
5 109.159.254.45 core2-te0-15-0-6.ealing.ukcore.bt.net 22 ms
6 109.159.254.108 peer1-xe9-1-0.telehouse.ukcore.bt.net 19 ms
7 195.66.224.243 17 ms
8 82.96.1.161 po41-20g-r85.cr0-r86.hy-sto.se.p80.net 94 ms
9 82.96.53.70 telelocation-hy-demarc0.sto.se.p80.net 53 ms
10 91.142.184.10 59 ms
11 46.253.205.246 mumble.pandemic-legion.com 59 ms
Trace complete

20:25
coms are p steady tonight, no worries. I'd say, result!
thanks a ton!

[#11]

Cracking.  Thanks for reporting the problem.

[#12]

a month down the line its been solid as a rock. no problems whatsoever.
thanks a lot everyone who chipped in to the thread and specially kelly/plusnet  for being awesome about the service they provide to scrubby end users and other unwashed proles like me.
ily.
o7o7o7

[#13]

No problem,
We have not yet switch back to Peering via Portlane on LINX yet, hence why this is all looking good still.
We are still awaiting for a reply from Portlane and chased them yesterday again.Given the lack of response, we are concious that we will continue to have problem if we were to switch back.
That is not to say we will not. A combination of cost, and the fact that peering should normally offer a better latency (so good for all) means we need to explore it, though we will put an equal case forward for not switching back.
I'll try to update you all on any progress we have here, given the impact this may or may not have.

[#14]

We have had a response on this now, which is looking like reasonable news to us and ultimately our customers longer term.
The latency has indeed been acknowledged, it's not as if it couldn't be. Work is in progress to resolve at this time and it is looking increasingly that work will be completed within the next 10 days.
We will look to get confirmation that this is complete before we make any tests and switch the relevant peering back on.

[#15]

Morning all,
We have received an update that means we are going to switch peering back on for Portlane.
I'm not sure specifically when yet, as this is not the highest priority (whilst it is working).
Do let us know if you do start to see issues and provide a trace if possible when you do.
Thanks.