Packet loss and increased latency

Text

1 23 45 67 89 1011 1213 1415 1617 1819 20

Computing statistics for 200 seconds...             Source to Here   This Node/LinkHop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address   0                                           {removed) [192.168.1.10]                                0/ 100 =  0%   |   1    0ms     0/ 100 =  0%     0/ 100 =  0%  192.168.1.1                                0/ 100 =  0%   |   2    9ms     0/ 100 =  0%     0/ 100 =  0%  losubs.subs.dsl2.wh-man.zen.net.uk [62.3.87.147]                                0/ 100 =  0%   |   3   23ms     4/ 100 =  4%     4/ 100 =  4%  ae1-114.cr1.wh-man.zen.net.uk [62.3.87.169]                                0/ 100 =  0%   |   4   29ms     0/ 100 =  0%     0/ 100 =  0%  ge-3-0-0-0.cr2.th-lon.zen.net.uk [62.3.80.45]                                0/ 100 =  0%   |   5   28ms     0/ 100 =  0%     0/ 100 =  0%  te-0-2-0-0.cr01.the-lon.pulsant.net [195.66.224.84]                                2/ 100 =  2%   |   6   29ms     2/ 100 =  2%     0/ 100 =  0%  te-0-0-0-0.cr01.dc1-mhd.pulsant.net [193.29.223.70]                                1/ 100 =  1%   |   7   29ms     5/ 100 =  5%     2/ 100 =  2%  ge-5-3.ds02.dc1-mhd.pulsant.net [46.249.195.26]                                0/ 100 =  0%   |   8   25ms     3/ 100 =  3%     0/ 100 =  0%  (removed)

Looks like congestion on the core Zen network.
Any comment Zen?

One to zeninternet.co.uk

Text

1 23 45 67 89 1011 1213 1415 1617 1819 2021 22

Computing statistics for 200 seconds...             Source to Here   This Node/LinkHop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address   0                                          (removed)  [192.168.1.10]                                0/ 100 =  0%   |   1    0ms     0/ 100 =  0%     0/ 100 =  0%  192.168.1.1                                0/ 100 =  0%   |   2   10ms     0/ 100 =  0%     0/ 100 =  0%  losubs.subs.dsl2.wh-man.zen.net.uk [62.3.87.147]                                2/ 100 =  2%   |   3   26ms     2/ 100 =  2%     0/ 100 =  0%  ae1-114.cr1.wh-man.zen.net.uk [62.3.87.169]                                0/ 100 =  0%   |   4   23ms     6/ 100 =  6%     4/ 100 =  4%  ae0-0.dr1.sp-roch.zen.net.uk [62.3.80.90]                                0/ 100 =  0%   |   5   25ms     3/ 100 =  3%     1/ 100 =  1%  ae1-0.dr2.sp-roch.zen.net.uk [62.3.80.94]                                0/ 100 =  0%   |   6   23ms     8/ 100 =  8%     6/ 100 =  6%  ge-1-1-104.access.ar1.sp-roch.zen.net.uk [62.3.81.146]                                0/ 100 =  0%   |   7   23ms     4/ 100 =  4%     2/ 100 =  2%  ge-0-0-15-100.fwhosting1a.sp-roch.corp.zen.net.uk [88.98.128.202]                                0/ 100 =  0%   |   8   25ms     2/ 100 =  2%     0/ 100 =  0%  core-212-23-0-100.zen.net.uk [212.23.0.100]  Trace complete.

mmm, what a surprise, I already started having issues like this a few months back however I was far too busy with other issues relating to moving home to argue with them...

Since left!

In reply to a post by mlmclaren:

mmm, what a surprise, I already started having issues like this a few months back however I was far too busy with other issues relating to moving home to argue with them...

Since left!

I have just moved to them, today. My ping monitoring graph looks atrocious from about lunchtime today, and it is almost clearly contention (or an overloaded router), because the graph has got even worse over the evening in line with peak time, but I'll give them the benefit of the doubt and assume that they are actively working on it...

Exactly the same as myself then.... it would correct itself for 12-24 hours when Zen rebooted a router at there end but then it returned and back at full issue after 72-96 hours

In reply to a post by mlmclaren:

Exactly the same as myself then.... it would correct itself for 12-24 hours when Zen rebooted a router at there end but then it returned and back at full issue after 72-96 hours

Better not be doing that with me else I'll be moving pretty sharply The "quality" tonight is unacceptable, but, I appreciate these things happen so I'll see what they say first...

How is BT infinity?

Edited by Pipexer (Wed 27-Jan-16 22:57:26)

BT's fine, not seen any congestion like issues yet, as you can see from link in my sig BQM is very good compared to my Zen connection.... I'll have a quick look to see if I archived any Zen shots

EDIT: You can find some examples of my BQM for the last 2 months, shame I can't find anything else like a full graph though.

Edited by mlmclaren (Thu 28-Jan-16 01:08:14)

care to share the live graph in your sig?

umm I see no BQM links in your sig.

Click the "Quality Monitor" link in my sig.

I did, it points to mouselike.org not tbb bqm.

Post deleted by Bryer

Mmm your BT connection does look pretty good indeed.

Still seeing congesion tonight on Zen but not as bad as was yesterday - but peak time seems to be around 9pm so I'll report back later on the evening. Either way it will need addressing.

Support said they have put a "ping monitor" on my connection - wonder where they are pinging from.

Will update in due course!

In reply to a post by Chrysalis:

I did, it points to mouselike.org not tbb bqm.

Erm, Yes that is a BQM/CQM.... I don't use ThinkBroadband's as it doesn't ping hostnames...

In reply to a post by mlmclaren:

In reply to a post by Chrysalis:

I did, it points to mouselike.org not tbb bqm.

Erm, Yes that is a BQM/CQM.... I don't use ThinkBroadband's as it doesn't ping hostnames...

And also pings from two places

In reply to a post by mlmclaren:

In reply to a post by Chrysalis:

I did, it points to mouselike.org not tbb bqm.

Erm, Yes that is a BQM/CQM.... I don't use ThinkBroadband's as it doesn't ping hostnames...

iic it does ping host names or it is supposed to do

ThinkBroadband's CQM only pings IP addresses

ok, I dont use failure because they auto delete if you dont login every so often.

So you had issues on zen and now they gone on BT? but you did move address as well.

I checked if it was a localised issue when I was at the old address and confirmed it was not

In reply to a post by Pipexer:

And also pings from two places

Which is a problem for Fritz!Box users, as it doesn't reply if there's too many echo requests.

Too bad for FritzBox users

I also have my own personal smokeping monitoring my line, mouselike I would use if I didnt need to login every 30 days to keep it going.

27th (unfortunately firebrick monitoring had not activated at this stage)

http://i63.tinypic.com/2quorc3.png

28th, 29th, 30th
http://i67.tinypic.com/20p4ps2.png

AAISP by comparison:
http://i64.tinypic.com/ileuyf.png


Clear contention and also visible packet loss at peak times occuring on the core zen network, it would seem.

http://i68.tinypic.com/16gl93o.jpg

All in all I am a little bit dissapointed to see this. I went with Zen (and paid the premium) so that this sort of thing would not be happening. Should have just gone with BT or TalkTalk.

Would welcome other people's thoughts...

Edited by Pipexer (Sun 31-Jan-16 22:10:34)

So your Avg ping times are around 20ms and its maxing out around 75ms and you are loosing less then 0.1% of ping's. Looks pretty good to me, its not something I study much though.

Ping isn't a true reflection of how normal traffic will flow, and occasional loss of packets is expected (yes expected) this is why we have TCP/IP and apps that use UDP are also designed to cope with lost packets.

Ok now i have stopped ranting ill try to be more helpful.

You try to work out where the delay is coming from... start with a traceroute (at a time you are seeing lots of delay) and then try pining every hop:




-bash-4.1$ traceroute aaisp.co.uk
traceroute to aaisp.co.uk (81.187.30.81), 30 hops max, 40 byte packets
1 gw (82.70.x.x) 0.644 ms 0.528 ms 0.567 ms
2 losubs.subs.dsl4.th-lon.zen.net.uk (62.3.84.19) 25.632 ms 25.784 ms 25.291 ms
3 ae1-119.cr2.th-lon.zen.net.uk (62.3.84.209) 25.800 ms 25.579 ms 25.825 ms
4 xe-0-0-2-0.mario.as34270.net (195.66.225.36) 25.535 ms 26.212 ms 25.918 ms
5 no-dns-yet.inetc.co.uk (85.91.238.219) 25.541 ms 26.193 ms 26.023 ms
6 * a.weightless.thn.aa.net.uk (90.155.53.61) 26.188 ms 26.476 ms
7 a.needless.thn.aa.net.uk (81.2.80.193) 26.804 ms 26.947 ms 26.288 ms
8 a.homeless.tch.aa.net.uk (217.169.11.93) 28.073 ms 28.278 ms 27.954 ms
9 limitless.aa.net.uk (81.187.30.81) 28.799 ms 28.808 ms 28.289 ms

-bash-4.1$ ping -s 62.3.84.19
PING 62.3.84.19: 56 data bytes
64 bytes from losubs.subs.dsl4.th-lon.zen.net.uk (62.3.84.19): icmp_seq=0. time=25.160 ms
64 bytes from losubs.subs.dsl4.th-lon.zen.net.uk (62.3.84.19): icmp_seq=1. time=24.979 ms
64 bytes from losubs.subs.dsl4.th-lon.zen.net.uk (62.3.84.19): icmp_seq=2. time=26.082 ms

-bash-4.1$ ping -s 62.3.84.209
PING 62.3.84.209: 56 data bytes
64 bytes from ae1-119.cr2.th-lon.zen.net.uk (62.3.84.209): icmp_seq=0. time=25.500 ms
64 bytes from ae1-119.cr2.th-lon.zen.net.uk (62.3.84.209): icmp_seq=1. time=25.889 ms
64 bytes from ae1-119.cr2.th-lon.zen.net.uk (62.3.84.209): icmp_seq=2. time=26.107 ms


-bash-4.1$ ping -s 195.66.225.36
PING 195.66.225.36: 56 data bytes
64 bytes from xe-0-0-2-0.mario.as34270.net (195.66.225.36): icmp_seq=0. time=74.655 ms
64 bytes from xe-0-0-2-0.mario.as34270.net (195.66.225.36): icmp_seq=1. time=57.076 ms
64 bytes from xe-0-0-2-0.mario.as34270.net (195.66.225.36): icmp_seq=2. time=25.931 ms


-bash-4.1$ ping -s 85.91.238.219
PING 85.91.238.219: 56 data bytes
64 bytes from no-dns-yet.inetc.co.uk (85.91.238.219): icmp_seq=0. time=25.560 ms
64 bytes from no-dns-yet.inetc.co.uk (85.91.238.219): icmp_seq=1. time=25.334 ms
64 bytes from no-dns-yet.inetc.co.uk (85.91.238.219): icmp_seq=2. time=26.051 ms


-bash-4.1$ ping -s 90.155.53.61
PING 90.155.53.61: 56 data bytes
64 bytes from a.weightless.thn.aa.net.uk (90.155.53.61): icmp_seq=0. time=26.182 ms
64 bytes from a.weightless.thn.aa.net.uk (90.155.53.61): icmp_seq=1. time=25.942 ms
64 bytes from a.weightless.thn.aa.net.uk (90.155.53.61): icmp_seq=2. time=26.410 ms


-bash-4.1$ ping -s 81.2.80.193
PING 81.2.80.193: 56 data bytes
64 bytes from a.needless.thn.aa.net.uk (81.2.80.193): icmp_seq=0. time=26.258 ms
64 bytes from a.needless.thn.aa.net.uk (81.2.80.193): icmp_seq=1. time=26.220 ms
64 bytes from a.needless.thn.aa.net.uk (81.2.80.193): icmp_seq=2. time=26.601 ms

I did that in my OP, perhaps go back and read it.

Is your connection using Zen LLU or is it BT WBMC ?

I'm not currently seeing peak time issues ,well nothing that is evident of ping graphs at least, But i have seen similar on my graphs several months back, It was down to BT and it's over contended Svlans ect, as there's no Zen LLU presence in my connected exchange, I have had this sorted twice so far by zen and that is within 12mths

Those hops in your trace route that are showing a small drop in packets isn't really a sign of an issue, just shows those switches are busy and icmp isn't programmed as a high priority so they start to not respond to every ping, if all successive hops from a particular hop to the target where showing similar packet loss then that could be an indication of a issue,

Edited by tommy45 (Mon 01-Feb-16 09:30:38)

Follow up. I lost PPP during a morning this week and since reconnecting things have been much better - looks like I have landed on a different gateway which is not hitting its limits.

http://oi68.tinypic.com/212s6tg.jpg

^ Much happier with the above!