The problem exhibits itself mostly during rainy periods - it starts to rain and I lose sync
I had Be* check things their end and line monitoring for 24 hours showed no issues
BT voice engineer came out and checked everything and replaced the master socket as it was slightly damaged. As the problems are the same plugged into the test socket, I knew however that this wouldn't make any difference
Around August this year, things started getting worse. When first connected to Be*, my downstream attenuation was 38db. Last year, it started hanging around 37.5db, then around August dropped down to 37db and is now sitting at 36.5db. Downstream output power also used to be around 19db, but has gone to 17.5db.
I've been using a Netgear DG834GT for about 6 months to see if there was any difference and it also doesn't spontaneously restart like the Speedtouch 780WL after a large amount of FEC errors are generated. I got a slightly more stable connection at first, but things progressively worse. I have on average about 10 LOS errors a day. Connected at the test socket was just as bad, if not worse.
I normally use an extension socket. Also, in October I fitted an iPlate mostly for the sake of curiousity.
Since about August time also, my modulation keeps going to ADSL2 instead of ADSL2+. I've never had problems with the modulation dropping down to ADSL2 before this. I also only averaged 5 - 10 LOS a week even during the old, wet winter months (apart from during periods of heavy rain and storms)
Last night, I connected the BeBox (780WL) back up for troubleshooting and to be ready for Be* support. I also took the downstream SNR up to 9db. Reliabity matters more at the moment than speed for increased level of online gaming over the Xmas period.
Stats up until Iost sync earlier todayat 18:42 and lost Internet connection. After a restart, got Internet back, but was on ADSL2. Took 4 restarts of the Bebox to get back onto ADSL2+
Uptime: 0 days, 0:18:31
Modulation: G.992.5 Annex A
Bandwidth (Up/Down) [kbps/kbps]: 1,325 / 7,709
Data Transferred (Sent/Received) [KB/KB]: 18.00 / 69.00
Output Power (Up/Down) [dBm]: 12.0 / 18.0
Line Attenuation (Up/Down) [dB]: 18.5 / 36.5
SN Margin (Up/Down) [dB]: 7.0 / 8.5
Vendor ID (Local/Remote): TMMB / �
Loss of Framing (Local/Remote): 20 / 0
Loss of Signal (Local/Remote): 5 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): 0
Error Seconds (Local/Remote): 141 / 0
FEC Errors (Up/Down): 0 / 19,140
CRC Errors (Up/Down): 124,331 / 647
HEC Errors (Up/Down): 1,332,180 / 591
As this has been going on for at least 3 months, I think I can rule out Xmas lights being the problem.
The router has been connected to the same extension socket for over a year, I'm using XF-1E filters and I've tried different filters. As mentioned above everything is just as bad at the test socket.
I'm on Highams Park exchange, E17 and they have been digging the roads up around here since June for Thames Water upgrade work, so it's possible this has had affect. They finished the work last week (for the time being at least) and things have gotten markedly worse
This is a a DMT graph after monitoring has been running for 8 hours
http://rapidshare.com/files/174667975/dmt20081218_1556.png.html
This is a DMT graph taken just now
http://rapidshare.com/files/174667978/dmt20081218_2119.png.html
Current stats
Uptime: 0 days, 2:26:27
Modulation: G.992.5 Annex A
Bandwidth (Up/Down) [kbps/kbps]: 1,321 / 7,427
Data Transferred (Sent/Received) [MB/MB]: 7.11 / 68.68
Output Power (Up/Down) [dBm]: 12.0 / 17.5
Line Attenuation (Up/Down) [dB]: 18.5 / 36.5
SN Margin (Up/Down) [dB]: 7.0 / 9.0
Vendor ID (Local/Remote): TMMB / �
Loss of Framing (Local/Remote): 0 / 0
Loss of Signal (Local/Remote): 0 / 0
Loss of Power (Local/Remote): 0 / 0
Loss of Link (Remote): 0
Error Seconds (Local/Remote): 23 / 0
FEC Errors (Up/Down): 0 / 26,747
CRC Errors (Up/Down): 124,430 / 86
HEC Errors (Up/Down): 1,333,534 / 50
I'm away over the weekend, so am going to leave things until Sunday and if still bad will log with Be* support. At the moment I cannot get to the master socket to test from the test socket, but as it's just the same from there it really doesn't make a difference.
Just wondering in the meantime if anyone here has any thoughts
Thanks
Nicola
Be* Unlimited



Pages in this thread:
Print Thread
nredwood