100M : Shocking performance over last month - CV10

[Wiggz]

Hi all,

The performance of my VM line has been shocking for the last month or so. Previously I used to get a constant 100Mbit service...using multiple Internet services was a dream! I would have recommended the VM service to anyone wanting something stable following my 100Mbit upgrade.

However in the last 4-6 weeks, it's become intolerable.

Packet loss is constant, latency is also at shocking proportions to many nodes (VM cores and internet sites) making general use an absolute no no.

Thinkbroadband meter is averaging 10% packetloss for crying out loud.

Just wondering if anyone has any advice to look at as the first 2 questions from technical support before I got ummed and arred at were reboot superhub, and reboot router.

CV10 area..

This is what I'm dealing with currently...

[qasdfdsaq]

Although possibly a faulty Superhub, more likely you have a line fault. Nothing you can do about it, VM need to send a tech.

Signal/power/noise levels most likely the culprit.

Posting your power levels would help confirm, but are rarely confirmative of anything.

[Wiggz]

Originally Posted by qasdfdsaq

Although possibly a faulty Superhub, more likely you have a line fault. Nothing you can do about it, VM need to send a tech.

Signal/power/noise levels most likely the culprit.

Posting your power levels would help confirm, but are rarely confirmative of anything.

[horseman]

Originally Posted by Wiggz

Assuming your VM alias is CPB on the H&S forum then ForumTeam there will respond usually within 2-4days (+ bank holiday) to check FEC/SNR levels + tech visit as appropriate.

[Wiggz]

Turns out VM know there is an over-subscription/congestion in my area and are doing something about it....in September!

Not happy.

[qasdfdsaq]

Highly unlikely to be oversubscription. Oversubscription causes graphs to look completely different.

[Wiggz]

That was what I thought also - upstream seems absolutely fine too which would be the first to be hit in high subscription areas.

[Wiggz]

Turns out that others are experiencing the same...latest BQM

[qasdfdsaq]

There's definitely some congestion in that graph Wiggz, but it still doesn't look like the main problem. I have heard in the past, IIRC, that VM won't allow agents to raise new or additional faults if there is an area-wide fault (such as congestion) already active, even if it is totally unrelated. Which is stupid IMO.

[Wiggz]

Luckily the VM community was helpful and I've got a Tech coming out tomorrow to check things out. Doubtful anything can be done as the PL has dropped WAY down now also, but nice to get somebody to look at it.