Are you referring to the TAP3 Test?

Up until Tuesday, I wasn't affected by this at all. Then Tuesday night I got bumped to a different gateway. I wasn't in last night but my TBB shows a sorry tale:

Tonight my throughput is radically down and my latency radically up. Changed gateway and back to normal. Results from the last 3 days (I'm on 21CN):
26/02/2013  8.35pm G/way pcl-ag07 Down: 71.87Mbit/s Up: 17.94Mbit/s Ping: 7ms
28/02/2013  8.02pm G/way ptw-ag04 Down: 15.42Mbit/s Up: 10.53Mbit/s Ping: 29ms
28/02/2013  8.35pm G/way ptn-ag02 Down: 71.67Mbit/s Up: 17.92Mbit/s Ping: 6ms
NOTE: I was getting problems on gateway ptw-ag04 tonight and moved FROM this gateway. AndyH moved TO the same gateway and now gets normal performance.
So the problem doesn't seem to be the gateway per se. Perhaps an effect that builds up over time on a gateway? Or it's not gateway related at all and all this gateway hopping is just misleading!?

On the TAP3 test (when it works), as AndyH says, surely that's only going to be useful if a customer is already logged into the test account when experiencing the performance problem. Disconnecting the router and logging in again to the test account is similar to changing gateways, except this time its from a Plusnet gateway to a BT gateway. So chances are it won't show up as a problem.

You are sort of right there Bright, I think, but at least it eliminates your connection to the BT network IIRC, but I don't know if that's to the Core nodes.
I was intrigued by some of Kelly's comments a few days back, where, when they see this start to happen they do something (on a particular Gateway?) and it goes away - for some
@Chris Pettitt
Yes, AndyH is talking about the Tap3 test. That IP address is the BTw Tester and putting it into my browser the tester ran ok. I haven't tried the Tap3 test again recently as I'm reluctant to lose my current Gateway, but as I posted earlier, the last time I tried it didn't work. At least waiting for 217.32.105.42 is a step in the right direction for AndyH.
Anyway 20CN, ptn-ag04 no problems so far as of 10 minutes ago. All normal.

@ Chris
Yep. Still not working for me.
@ Bright
At least it would be proof (definitive?) where the problem lies?

Okay, if you're still having issues with that, can we start a new thread as we'd like to keep this one for Peak Time Performance issues

Quote from: Bright

Up until Tuesday, I wasn't affected by this at all. Then Tuesday night I got bumped to a different gateway. I wasn't in last night but my TBB shows a sorry tale:

That big green spike was last night yeah?  That is actually the problem which started this thread which is specifically a hardware problem.  I explained it somewhat here: http://community.plus.net/forum/index.php/topic,111170.msg959759.html#msg959759
I think the other issue you guys are seeing which involves general slowdowns and poorer performance is a different problem.  If *could* be related, but definitely isn't the hardware I describe above, at least it's not manifesting the same

Quote from: Kelly

That big green spike was last night yeah?   That is actually the problem which started this thread which is specifically a hardware problem.   I explained it somewhat here: http://community.plus.net/forum/index.php/topic,111170.msg959759.html#msg959759 I think the other issue you guys are seeing which involves general slowdowns and poorer performance is a different problem.  If *could* be related, but definitely isn't the hardware I describe above, at least it's not manifesting the same

So are you saying that you changed something in the network yesterday which caused the hardware problem to be triggered again?
Or you haven't got enough additional kit in the network (while your supplier works on a fix) to spread the load enough to prevent the hardware problem affecting some users every evening?

Bright, unless I misread it, I thought Plusnet had said there was no capacity problem on their Network. Hardware faults are a slightly different kettle of fish are they not?
20CN ptn-ag04. Pings more variable now, but speed tests still fine -
Reply from 212.159.6.10: bytes=32 time=44ms TTL=250
Reply from 212.159.6.10: bytes=32 time=38ms TTL=250
Reply from 212.159.6.10: bytes=32 time=38ms TTL=250
Reply from 212.159.6.10: bytes=32 time=46ms TTL=250
Reply from 212.159.6.10: bytes=32 time=42ms TTL=250
Reply from 212.159.6.10: bytes=32 time=65ms TTL=250
Reply from 212.159.6.10: bytes=32 time=44ms TTL=250
Reply from 212.159.6.10: bytes=32 time=40ms TTL=250
Reply from 212.159.6.10: bytes=32 time=39ms TTL=250
Reply from 212.159.6.10: bytes=32 time=59ms TTL=250
Reply from 212.159.6.10: bytes=32 time=3914ms TTL=250
Reply from 212.159.6.10: bytes=32 time=43ms TTL=250
Reply from 212.159.6.10: bytes=32 time=41ms TTL=250
Reply from 212.159.6.10: bytes=32 time=45ms TTL=250
Reply from 212.159.6.10: bytes=32 time=36ms TTL=250
Reply from 212.159.6.10: bytes=32 time=36ms TTL=250
Reply from 212.159.6.10: bytes=32 time=38ms TTL=250
Reply from 212.159.6.10: bytes=32 time=44ms TTL=250
Reply from 212.159.6.10: bytes=32 time=38ms TTL=250
Reply from 212.159.6.10: bytes=32 time=47ms TTL=250
Reply from 212.159.6.10: bytes=32 time=47ms TTL=250
Reply from 212.159.6.10: bytes=32 time=47ms TTL=250
In view of Kelly's comment wouldn't it be better to have a new thread for Peak Time Performance issues and I'd suggest on Broadband & Routers where it would be more obvious and save a lot of separate threads cropping up as at present. I'm sure the Mods would be willing to make it a sticky for the duration - and keep this thread for the specific hardware issue as outlined by Kelly - What do you think Kelly, Chris?

We are in the process of adding 2 more bits of kit (one is going live tomorrow) which should put enough kit in the network to prevent a reoccurrence of the issue.  Because it is traffic level related, I suspect we had  a customer being DDOS on PCL-AG08 which pushed the device into failure mode last night.  It's the first time this has happened in quite a while due to the early scaling work we've been doing.

Eek, my speed has plummeted to about 2Mbps (still synced at 66362) - this has never happened to me before. I'll try a disconnect.

What gateway you on?
I was just about to see that we seemed to be seeing less problems tonight as well 😕
If you guys are reading and seeing ok performance/speeds tonight, please post and let us know what gateway you are on too.

I was on pcl-ag02, just hopped to pcl-ag07 and it's fixed it, back up to 60 meg+.

@Anotherone
It's not strictly a hardware fault/failure (other than that the hardware doesn't perform to spec), more a network issue with a particular piece of hardware. Not sure how Plusnet prefers to handle the forum issue, but I guess this thread has now become the place where people report their stats when experiencing peak-time performance problems. Very hard for customers to differentiate their symptoms between the two possibly different causes.
@Kelly
Thanks. Understood. Although the symptoms seem remarkably similar (reduced throughput/flow control/dropped packets for all traffic types), just not such a hard cut-over between normal/fault condition, and presumably not locatable to a particular bit of kit...?