Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[phi2008]

I've just migrated FTTC from TalkTalk to Plusnet, my Thinkbroadband monitoring has gone from this(TalkTalk) -



To this with Plusnet -



The graph between 10PM-12PM shows the typical packet loss I'm getting. My VDSL modem FEC errors have also gone from a couple of thousand over weeks to over 500,000 in 10 hours after the migration. Since it's Plusnet I've tried changing gateways and that made no difference. The Plusnet connection is currently syncing at 80Mb when under TalkTalk it synced at about 72Mb plus interleaving.


So it seems like migration was the start of packet loss, my line has been perfect up to this point in time(2 years). I'd like to know what physically happens during a migration like mine to try to get to the bottom of the newly acquired packet loss.

[ip75]

My guess (and it's just a guess) is that DLM was reset as part of the migration, and because you no longer have interleaving you are now exposed to all the errors that interleaving was previously protecting you from. If this is true, then in the next day or so DLM should re-apply the interleaving and get you back to where you were.

[Chrysalis]

I think ip75 is right.

[phi2008]

My guess (and it's just a guess) is that DLM was reset as part of the migration, and because you no longer have interleaving you are now exposed to all the errors that interleaving was previously protecting you from. If this is true, then in the next day or so DLM should re-apply the interleaving and get you back to where you were.

That's what I had assumed(probably - wanted to explore all avenues) as well ... having said that I've been running winmtr to gateways and constantly got about 3% loss - having swapped almost everything out I've now replaced an HG612(with latest unlocked firmware, used with TalkTalk) with an ECI - seems to be showing almost zero loss at the moment, though I'll have to wait for the graphs to update to get a better idea.

[phi2008]

Well the ECI modem seems to have solved the problem(swapped in about 4.30PM) -




Strange how the HG612 was fine on TalkTalk but didn't like Plusnet.

[Dray]

Due to the G.INP problem, the ECI modem causes your connection to have large amounts of interleaving applied.

[phi2008]

Wouldn't interleaving significantly increase the ping? That's not what the graph shows.

[Dray]

What's the ping, to BBC say?

[phi2008]

Pinging www.bbc.net.uk [212.58.244.66] with 32 bytes of data:
Reply from 212.58.244.66: bytes=32 time=10ms TTL=56
Reply from 212.58.244.66: bytes=32 time=9ms TTL=56
Reply from 212.58.244.66: bytes=32 time=9ms TTL=56
Reply from 212.58.244.66: bytes=32 time=9ms TTL=56

Ping statistics for 212.58.244.66:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 9ms, Maximum = 10ms, Average = 9ms

[Dray]

That looks like a fastpath ping to me, but I'm on BT so I can't really tell.

Are you connected to a Huawei cab or an ECI cab?

Did you have the latest firmware on the HG612?

[phi2008]

ECI cab AFAIK, latest unlocked G.INP firmware on HG612.

[Dray]

Maybe your line works better with the modem chipset matching the cab.

[burakkucat]

ECI cab AFAIK,

It might be best to confirm that one way or another.

So make a telnet connection to the HG612, log in with the admin/admin credentials, enter the busybox shell and issue a xdslcmd info --pbParams command.

Show the resulting output to the Eagle (Bald) who knows and he will disclose the cabinet's equipment manufacturer. 

[phi2008]

Have to wait a while before I swap the HG612 back in, here's my cabinet which I think is an ECI 256 -

[burakkucat]

A picture is worth a thousand words. 

Yes, unfortunately, that is a cabinet equipped with an ECI Hi-FOCuS Mini-Shelf M41 MSAN.