Peak time latency spikes

[#1]

Kelly as per my query on Usenet speed dropping I'm now on ptw-ag03, usenet is still crawling at 1mbps but speedtest give me over 56Mbps

[#2]

I'm happy for people to report specific large scale peak time slow downs in here.   If we think someone has a more generic fault, we can report he post to be split out.
It could be that the two problems are connected, however, with the first problem, we see one of the devices stuck at 100% CPU.  We don't see that associated with the more generic slowdowns.
,

[#3]

@Kelly
See the second half of my post. In particular the note about the gateway that worked perfectly for AndyH but at exactly the same time was giving me severe performance problems.

[#4]

Yep!  Seen that problem before.  Truly weird.  I'm sure once we work out what's going on here it will make perfect sense....

[#5]

Currently on ptw-ag04 and so far the speed seems to be ok, although the TBB graph is showing higher peaks than earlier and speed is a little erratic:


Jitter is climbing and the line rating is decreasing.
TBB graph:

I'll have another check on this later - 10ish last night was bad, so will try again then.

[#6]

Speed is still good for me tonight im on gateway ptw-ag04 which i have remained on since yesterday latency has been pretty good touch would it stays like this .

[#7]

I am suffering from slow speeds tonight again, been happening for the past few nights. Usually a gateway hop solves it but this is getting annoying having to do this every day. Currently on ptn-ag06 and my speeds are around 25-30Mb slower than usual.
Pages are slowing and basic websites such as Facebook that usually loads instantly on my 78Mb fibre connection is taking 10-20 seconds to load completely so I am not sure if DNS is maybe involved in this as well.

[#8]

I've been seeing really slow speeds in the evenings for a while, but wasn't sure if it was an isolated problem or not. Anyway, decided to run a speedtest this morning at about 7am before work and got this:

And this is just now:

And this was last night:

All of these were on: pcl-ag02
Not great. 
Edit: although my monitor looks suprisingly healthy:

[#9]

Quote from: bobboulby

Really bad this evening on pcl-ag08 - iPlayer is unusable. $ ping bbc.co.uk PING bbc.co.uk (212.58.241.131) 56(84) bytes of data. 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=1 ttl=245 time=196 ms 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=2 ttl=245 time=196 ms 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=4 ttl=245 time=197 ms 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=5 ttl=245 time=184 ms 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=6 ttl=245 time=181 ms 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=7 ttl=245 time=183 ms 64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=8 ttl=245 time=184 ms ^C --- bbc.co.uk ping statistics --- 9 packets transmitted, 7 received, 22% packet loss, time 8007ms rtt min/avg/max/mdev = 181.459/189.066/197.015/6.644 ms Bob

Much better this evening, still on pcl-ag08

PING bbc.co.uk (212.58.241.131) 56(84) bytes of data.
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=1 ttl=245 time=14.5 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=2 ttl=245 time=14.2 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=3 ttl=245 time=14.7 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=4 ttl=245 time=26.7 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=5 ttl=245 time=14.6 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=6 ttl=245 time=14.5 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=7 ttl=245 time=15.1 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=8 ttl=245 time=15.8 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=9 ttl=245 time=14.9 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=10 ttl=245 time=26.8 ms
64 bytes from virtual-vip.thdo.bbc.co.uk (212.58.241.131): icmp_req=11 ttl=245 time=14.5 ms
^C
--- bbc.co.uk ping statistics ---
11 packets transmitted, 11 received, 0% packet loss, time 10013ms
rtt min/avg/max/mdev = 14.238/16.993/26.895/4.663 ms
Bob

[#10]

Bob: Your problems yesterday were the bit spike issue on PCL-AG07/08, rather than the more general slowdowns problem.

[#11]

Things are slightly strange here at 2202 - 20CN ptn-ag04 -
BTw tester 0.4-0.5Mbps slower than normal (usually 6.2Mbps)
BBC Diagnostic attached - figures are normally all over 6000kbps,
but pings not as bad as earlier -
Reply from 212.159.6.10: bytes=32 time=41ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250
Reply from 212.159.6.10: bytes=32 time=42ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250
Reply from 212.159.6.10: bytes=32 time=42ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250
Reply from 212.159.6.10: bytes=32 time=37ms TTL=250
Reply from 212.159.6.10: bytes=32 time=38ms TTL=250
Reply from 212.159.6.10: bytes=32 time=42ms TTL=250
Reply from 212.159.6.10: bytes=32 time=37ms TTL=250
Reply from 212.159.6.10: bytes=32 time=37ms TTL=250
Reply from 212.159.6.10: bytes=32 time=46ms TTL=250
Reply from 212.159.6.10: bytes=32 time=37ms TTL=250
Reply from 212.159.6.10: bytes=32 time=44ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250
Reply from 212.159.6.10: bytes=32 time=40ms TTL=250
Reply from 212.159.6.10: bytes=32 time=37ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250
Reply from 212.159.6.10: bytes=32 time=37ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250

[#12]

still no issues for me on ptw-ag02 so far tonight, swapped to this one last night at 7:30

[#13]

Things are starting to slide, speed is down 5Mb:

Oddly the pingtest results are better:

But the TBB graph has just started producing big spikes again - only just visible but this is similar to last night 

I'll do another set before I sign off for the night - I'm feeling slightly less grumpy tonight too, perhaps it shows 

[#14]

Traceroute still looking good no issue as yet .
Microsoft Windows [Version 6.0.6002]
Copyright (c) 2006 Microsoft Corporation.  All rights reserved.
C:\Users\Private>tracert www.bbc.co.uk
Tracing route to www.bbc.net.uk [212.58.246.94]
over a maximum of 30 hops:
  1    <1 ms    <1 ms    <1 ms  192.168.1.1
  2    16 ms    22 ms    15 ms  lo0-central10.ptw-ag04.plus.net [195.166.128.194
]
  3    12 ms    11 ms    11 ms  link11-central10.ptw-gw01.plus.net [84.93.248.11
6]
  4    11 ms    13 ms    11 ms  xe-10-0-0.ptw-cr01.plus.net [212.159.1.40]
  5    11 ms    11 ms    11 ms  kingston-gw.thdo.bbc.co.uk [212.58.239.6]
  6    *        *        *    Request timed out.
  7    13 ms    13 ms    12 ms  ae0.er01.cwwtf.bbc.co.uk [132.185.254.93]
  8    16 ms    13 ms    13 ms  132.185.255.165
  9    12 ms    12 ms    12 ms  bbc-vip015.cwwtf.bbc.co.uk [212.58.246.94]
Trace complete.
C:\Users\Private>

[#15]

jamesanstee:  We put you on the elevated product yesterday.  This could mean that the problems you've been seeing are BT network congestion related....