purleigh : exchange congestion on EAPUR ?

[#1]

While not as bad as others in this thread, last night I was seeing some packet loss and unusually high (blue) average latency on pcl-ag02.
I very rarely see ANY red packet loss showing.
The blue average latency is rarely above 60 to 80ms when streaming 720p, but last night it was nearer 140ms !.

[#2]

I was also wondering if there is congestion somewhere, because I'm not doing anything different from usual and only using a small percentage of connection available bandwidth (<30% down / <15% up) and router CPU resources (~4%), yet for a second night have really bad average latency but thankfully no packet loss tonight.  Still on gateway pcl-ag02.
My modem looks fine, and has recorded zero errors in more than 35 days, and is showing the same great stats as it always has. 
Unfortunately I can't tell whether my exchange is congested as the Plusnet Exchange Checker hasn't been updated since 31 Jan 2014  

[#3]

@'Kelly' - Can you do something about updating the Exchange Checker ?
because as I said, it says "Record last updated: 31 Jan 14" and that shows the last status for "03 Jan 14"
What use is out of date status information, other than looking for historical trends ?
See attached screenshot -

[#4]

We don't formally support tools on the usertools/usergroup box as they were owned by the PUG members.  We've been trying to see what we need to go to get that updated though.

[#5]

As the Plusnet Exchange Status Checker isn't showing current congestion status,  does anyone know of an equivalent exchange congestion checker that works ?,  as I would like to see whether I can eliminate exchange congestion as the most likely cause of my worsening evening latency and last nights reducing bandwidth (despite my modem connection stats remaining near perfect) -

[#6]

I hadn't appreciated how bad the latency got last night by just looking at the TBB ping chart above !  
I just checked my SamKnows whitebox report which says latency hit 581ms average and 876ms maximum around 22:00 !  
See attachment below for detailed figures.  
Anyone got any ideas where to view up to date exchange congestion status ?  

[#7]

PLEASE can someone from DCT check my exchange PEAK TIME congestion status ("Purleigh" / EAPUR) and upstream broadband endpoints as last night my connection was worse than ever.
I had RouterStats running all night and my modem connection was PERFECT, with full 8128Mbps sync, flat noise margin, zero errors, nice bit/tone shape, etc, and my router was running <5% CPU usage.
However my SamKnows latency hit 1083ms and the connection became near unusable, and when attempting to max out my connection I could only pull 2.1Mbps when normally I get 6.7Mbps download doing the same thing.
Of course, running any tests during the day is giving me my full profile speed  
It is only this week that peak time latency and loss of bandwidth have appeared. 
Has Plusnet got enough capacity on 20CN connections ? 

[#8]

Posts and replies split off to its own topic at the request of the OP

[#9]

Ah, wondered why I was getting errors trying to response to the previous thread.
@Purleigh,
I'll ask Chris Pettitt to take a look shortly.

[#10]

Thanks Chris,  I asked to have the above replies split off, as my latency issue was getting lost when mixed up with other problems in the original thread.

[#11]

Just as a side note, the exchange checker was something we updated but the sheet we used to update it hasn't been published for a while. We've raised this issue with BTW to get an updated version of the sheet.

[#12]

Hi Purleigh,
Your exchange does have congestions issues, our suppliers are aware of the issue and have plans for this to be resolved by 30/06/2014 - I appreciate that this isn't exactly soon, however, this is the earliest they are able to allocate the work for this.
With regards to the Exchange Checker, as Chris has mentioned, BTW have changed some of the formatting from the data we pulled, we're trying to get this reverted so we can update the checker - it's a little out of our hands at the moment 😕

[#13]

Fantastic !
At least I now know that I don't have some bizarre fault with my modem connection,
and won't waste any more time blindly fault finding from my end.

It is a pity that after waiting until 30/06/2014 for BTW to increase the capacity, my exchange will still be 20CN  

Thanks to both Chris's and OldJim for helping me get to an answer,  and I hope you all have a great weekend.  

[#14]

If your service is below the acceptable levels (i.e. is red when you run the Further Diagnostics on the BTw tester), it's classified as a fault. So technically speaking, you shouldn't be paying the monthly charge until it's fixed.

[#15]

When I did that test earlier it said that my acceptable range of speeds is 1.2 to 7.15Mbps 
The slowest I've seen it this week (multiple downloads and reading my Linux firewall for throughput) was around 2Mbps.

I am mostly relieved that I don't have a (modem) line fault,  and thankful that I use a squid caching http proxy - so general web browsing is still very responsive,  it is just video streaming that is slowing down during the peak time - so I will just use lower resolution for the next six weeks.
It is just frustrating that all the exchanges around me are to be upgraded thanks to BDUK, annoying that those exchanges already get good speed are to be upgraded to fibre,  yet my local up to 8Mbps 20CN exchange will not be upgraded and at the same time has run out of bandwidth.