Sync drops when phone rings (subject was "Is this a record?)

[#1]

max and I had the same problem you did which improved when I went to 21CN but only after a complete pair change (exchange to cab) did the problem finally go away and I could have the upload uncapped

[#2]

It's not the same gap. Jim's old missing tones are on the Upstream 29-31 which fits in with the issues he was seeing initially after moving to 21CN.

[#3]

All I can say is WTF!
I've just tried swapping to a TG582n with 8.4.4.J firmware. To say the bit loading is odd is an understatement - think I'll try a resync!

[#4]

Looks corrupted (the first populated downstream tone appears to go beyond 16 bits for a start).
[Edit] Looking at the graph again it's not tone 31 - not very clearly labelled.

[#5]

I've just resynced and it looks the same. Connection is running OK:

[#6]

This is what the router is reporting:

{admin}=>xdsl debug bitloadinginfo
   Tone :  # Bits per tone
      0 :   0  0  0  0  0  0  0  0  2  3
     10 :   5  6  6  7  7  7  7  8  7  8
     20 :   7  7  7  7  7  6  6  5  4  4
     30 :   3  0 31 -27 32 80  0  0  0  0
     40 :  31 -27 63 -4  9 10 10 10 11 11
     50 :  11 11 12 12 12 12 12 12 12 12
     60 :  12 11 11 12  0 12 11 12 12 12
     70 :  12 11 11 11 11 11 11 11 11 11
     80 :  11 11 11 11 11 11 11 11 11 11
     90 :  11 11 11 11 11 11 11 11 11 11
    100 :  11 11 11 11 11 11 11 10 11 10
    110 :  10 10 10 10 10 10 10 10 10 10
    120 :  10 10 10 10 10 10 10 10 10 10
    130 :  10 10 10 10 10 10 10 10 10 10
    140 :  10 10 10 10 10 10  9 10 10 10
    150 :  10 10 10 10  2 10 10 10 10 10
    160 :  10 10 10 10  9 10 10 10 10  9
    170 :  10 10  9  9  9  8  8  8  8  9
    180 :   8  8  8  8  9  8  8  8  8  8
    190 :   8  8  8  8  8  8  8  8  8  8
    200 :   8  8  8  8  8  8  8  8  8  8
    210 :   8  8  8  8  8  8  8  8  8  8
    220 :   7  8  8  8  8  8  8  8  8  7
    230 :   7  7  7  7  7  7  7  7  7  7
    240 :   7  7  7  7  7  7  7  7  7  7
    250 :   7  7  7  7  7  0

[#7]

Do you get those -ve numbers everytime the command is run? If so, how odd - might possibly have expected it if Bit swapping was going on at the instant. Never mind the ones that are 0 when they shouldn't be

[#8]

This is after a full router restart:

{admin}=>xdsl debug bitloadinginfo
   Tone :  # Bits per tone
      0 :   0  0  0  0  0  0  0  0  3  3
     10 :   5  6  6  7  7  7  7  7  8  8
     20 :   7  7  7  7  6  6  6  5  4  4
     30 :   3  0 31 -27 32 80  0  0  0  0
     40 :  31 -27 63 -4  9 10 10 10 11 11
     50 :  11 11 12 12 12 12 12 12 12 12
     60 :  12 12 12 12  0 12 11 11 12 12
     70 :  11 11 11 11 11 11 11 11 11 11
     80 :  11 11 11 11 11 11 11 11 11 11
     90 :  11 11 11 11 11 11 11 11 11 11
    100 :  11 11 11 11 11 10 11 11 10 10
    110 :  10 10 10 10 10 10 10 10 10 10
    120 :  10 10 10 10 10 10 10 10 10 10
    130 :  10 10 10 10 10 10 10 10 10 10
    140 :  10 10 10 10  2 10  9 10 10 10
    150 :  10 10 10 10 10 10 10 10 10 10
    160 :  10 10 10 10  9 10 10 10 10 10
    170 :  10  9  8  9  9  8  8  8  9  9
    180 :   8  8  8  9  9  8  8  8  8  8
    190 :   8  8  8  8  8  8  8  8  8  8
    200 :   8  8  8  8  8  8  8  8  8  8
    210 :   8  8  8  8  8  8  8  8  8  8
    220 :   8  8  8  8  8  8  7  8  8  7
    230 :   7  7  7  7  7  7  7  7  7  7
    240 :   7  7  7  7  7  7  7  7  7  7
    250 :   7  7  7  7  7  0

Edit: I'm going to go back to the other TG582 with the latest firmware. Back in a bit.

[#9]

I had a similar Routerstats plot with 8.4.4.J which after a while righted itself, the tones were all shifted to the right with silly values plotted, as you have, but as you say the connection was OK at the time.
Apart from Firmware, are they the same, both without USB ports?

[#10]

Back to normality on the 10.2.2.B firmware:

{admin}=>xdsl debug bitloadinginfo
   Tone :  # Bits per tone
      0 :   0  0  0  0  0  0  0  0  2  3
     10 :   5  5  6  7  7  7  7  8  8  8
     20 :   8  7  7  7  6  6  6  5  4  4
     30 :   3  0  0  0  0  0  0  6  6  7
     40 :   7  8  8  9  9 10 10 10 11 11
     50 :  11 11 12 12 12 12 12 12 12 12
     60 :  12 12 11 12  0 12 11 12 12 12
     70 :  11 11 11 11 11 11 11 11 11 11
     80 :  11 11 11 11 11 11 11 11 11 11
     90 :  11 11 11 11 11 11 11 11 11 11
    100 :  11 10 11 10 11 10 11 11 11 10
    110 :  10 10 10 10 10 10 10 10 10 10
    120 :  10 10 10 10 10 10 10 10 10 10
    130 :  10 10 10 10 10 10 10 10 10 10
    140 :  10 10 10 10 10 10  9 10 10 10
    150 :  10 10 10 10 10 10 10 10 10 10
    160 :  10 10 10 10  9 10 10 10 10 10
    170 :   2  9  8  9  9  8  8  8  8  9
    180 :   9  8  8  8  9  8  9  8  8  8
    190 :   8  8  8  8  8  8  8  8  8  8
    200 :   8  8  8  8  8  8  8  8  8  8
    210 :   8  8  8  8  8  8  8  8  8  8
    220 :   8  8  8  8  8  8  8  8  8  7
    230 :   7  7  7  7  7  7  7  7  7  7
    240 :   7  7  7  7  7  7  7  7  7  7
    250 :   7  7  7  7  7  0

Yep, identical boxes, neither has USB

[#11]

I never got to the bottom of why it behaved like that with 8.4.4.J and as I said it righted itself, so whether it was some funny with RouterStats I'm not sure. I couldn't test for too long because it was 8.4.4.J on a TSTC line card and eventually the bit swaps on the Upstream got bad it started to drop connection so I gave up and went to 10.2.2.B
What was your previous exchange line card? If you feel like experimenting more it would be interesting to see if the 8.4.4.J eventually rights the RS plot but whether it will hold connection on the Marconi card. I suspect not - but you never know.
Edit: you haven't tried the Netgear yet have you?

[#12]

Out of interest, was there any significant difference in sync between the two modems?

[#13]

No, that was the odd thing. The sync is solid on 7616 so although the router is reporting channel mode fast I'm beginning to suspect otherwise.

[#14]

have you checked your ping

[#15]

D'oh - thanks for pointing out the obvious

C:\Users\John>ping ntp.plus.net
Pinging ntp.plus.net [212.159.13.50] with 32 bytes of data:
Reply from 212.159.13.50: bytes=32 time=15ms TTL=249
Reply from 212.159.13.50: bytes=32 time=15ms TTL=249
Reply from 212.159.13.50: bytes=32 time=15ms TTL=249
Reply from 212.159.13.50: bytes=32 time=14ms TTL=249
Ping statistics for 212.159.13.50:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 14ms, Maximum = 15ms, Average = 14ms

Definitely fastpath!