Dropouts & latency spikes

[#1]

I've been having dropouts of the service several times a day. Plusnet monitored the service and say it's fine - so I paid for a fixed IP and set up ping monitoring. This shows bad (but intermittent) packet loss spikes (see attached pdf).
If the back-end is fine, could this be the copper line from the FTTC cabinet?  It's intermittent but I've had these drop-outs from day 1 of the service (a major reason for replacing the supplied router with my own as the Technicolour took around 7 minutes each time to re-boot to a usable connection and had to be rebooted manually as it wouldn't re-establish a connection on its own - hopeless!).
How might I get this service to be more reliable?  (Incidentally the quiet-line test never shows any issues. Went through all that many times).
Andrew

[#2]

Hi there,
That graph looks horrible, sometimes that type of graph can be caused by large amounts of upload happening, but this is not the case in this situation. I've looked over the test results from our faults testing and they look fine, no issues we can detect. I wonder if we need to hit the test at the time you're having the problems for something to be picked up.
I know you've tested the phone line, but I'm going to run a quick test on it anyway for peace of mind and to eliminate that now.

[#3]

As expected, that's showing as fine with no issues seen.

Your line is showing a few more disconnections than I'd like really. I'd just been through the same test our faults team performed on your line and again it's showing no issues. It seems odd that there is so much packet loss on the line but there are very low error rates. We'll need to do some more investigation on the faults ticket to see what's going on.

[#4]

Thanks Chris - all help appreciated!  Since 5am there have been c. 10 latency spikes but brief (1 to a few minutes) so difficult to 'hit'.
Is traceroute data any use?  I tried it to your primary DNS IP and got this:
traceroute to 212.159.13.49 (212.159.13.49), 64 hops max, 72 byte packets
1  192.168.1.1 (192.168.1.1)  78.882 ms  1.097 ms  0.998 ms
2  lo0.11.central11.pcl-bng04.plus.net (195.166.130.178)  28.878 ms  115.155 ms  41.343 ms
3  irb.11.pcl-cr01.plus.net (84.93.249.97)  299.119 ms  498.611 ms  85.052 ms
4  po2.pcl-gw01.plus.net (195.166.129.41)  16.435 ms  102.362 ms  105.155 ms
5  cdns01.plus.net (212.159.13.49)  204.641 ms  65.724 ms  340.215 ms
Some of those times look high so I tried ping:
PING 212.159.13.49 (212.159.13.49): 56 data bytes
Request timeout for icmp_seq 0
64 bytes from 212.159.13.49: icmp_seq=1 ttl=60 time=8.527 ms
Request timeout for icmp_seq 2
64 bytes from 212.159.13.49: icmp_seq=2 ttl=60 time=1068.713 ms
64 bytes from 212.159.13.49: icmp_seq=3 ttl=60 time=111.379 ms
64 bytes from 212.159.13.49: icmp_seq=4 ttl=60 time=10.798 ms
64 bytes from 212.159.13.49: icmp_seq=5 ttl=60 time=10.340 ms
64 bytes from 212.159.13.49: icmp_seq=6 ttl=60 time=10.602 ms
64 bytes from 212.159.13.49: icmp_seq=7 ttl=60 time=11.657 ms
64 bytes from 212.159.13.49: icmp_seq=8 ttl=60 time=10.493 ms
64 bytes from 212.159.13.49: icmp_seq=9 ttl=60 time=10.670 ms
64 bytes from 212.159.13.49: icmp_seq=11 ttl=60 time=10.522 ms
64 bytes from 212.159.13.49: icmp_seq=12 ttl=60 time=10.951 ms
64 bytes from 212.159.13.49: icmp_seq=13 ttl=60 time=41.731 ms
Request timeout for icmp_seq 14
64 bytes from 212.159.13.49: icmp_seq=14 ttl=60 time=1401.694 ms
64 bytes from 212.159.13.49: icmp_seq=16 ttl=60 time=12.032 ms
Request timeout for icmp_seq 17
64 bytes from 212.159.13.49: icmp_seq=18 ttl=60 time=41.357 ms
64 bytes from 212.159.13.49: icmp_seq=19 ttl=60 time=10.867 ms
64 bytes from 212.159.13.49: icmp_seq=20 ttl=60 time=10.387 ms
64 bytes from 212.159.13.49: icmp_seq=21 ttl=60 time=25.178 ms
64 bytes from 212.159.13.49: icmp_seq=22 ttl=60 time=10.366 ms

[#5]

Are you connected wired or wirelessly?
On the traceroute, it looks like the issue is possibly within your network (I am guessing you're connected wirelessly).

[#6]

I was indeed, although the graphs are external monitoring.  I connected to the router with a wire and got this:
traceroute to 212.159.13.49 (212.159.13.49), 64 hops max, 72 byte packets
1  192.168.1.1 (192.168.1.1)  1218.214 ms *  1761.092 ms
2  * lo0.11.central11.pcl-bng04.plus.net (195.166.130.178)  3865.334 ms *
3  irb.11.pcl-cr01.plus.net (84.93.249.97)  3333.164 ms  688.949 ms  123.789 ms
4  *
Then couldn't access the internet at all so tried the router admin page. Couldn't display that even after rebooting (never happened before).
Now using the Openreach box and a different router (Airport Express).
traceroute is now good (wifi connected):
traceroute to 212.159.13.49 (212.159.13.49), 64 hops max, 72 byte packets
1  10.0.1.1 (10.0.1.1)  1.102 ms  0.780 ms  3.341 ms
2  lo0-central10.ptn-ag04.plus.net (195.166.128.193)  9.697 ms  9.875 ms  34.238 ms
3  link-a-central10.ptn-gw01.plus.net (212.159.2.140)  9.996 ms  9.859 ms  9.484 ms
4  xe-8-2-0.ptw-cr01.plus.net (212.159.0.116)  9.439 ms  9.887 ms  9.582 ms
5  ae2.ptw-cr02.plus.net (195.166.129.5)  9.831 ms  10.190 ms  60.659 ms
6  po2.ptw-gw02.plus.net (195.166.129.39)  9.749 ms  9.961 ms  9.889 ms
7  * * *
8  cdns01.plus.net (212.159.13.49)  12.069 ms  9.651 ms  9.625 ms
Obviously the router looks suspicious but I had these line drops with the original Technicolour router so will run with the Airport for a while...
Update after 24 hours (at 18:20 Sunday): Connection has been almost perfect so far with the different router.  Will update tomorrow night. Was using a Vigor 2760n...

[#7]

Looks like it was the router (Draytek Vigor 2760n) causing the problems (14 months old).  The connection has been perfect since using the original BT VDSL box and Apple Airport Express router.
Time for a new WiFi router...
Thanks to all who responded to previous posts.

[#8]

Glad you got to the bottom of this.