>I have noticed the same thing. Several customers have reported that their
>connect speed has dropped. One modem is a Motorola Modem Surfer 56k V.90
>connection. It did connect consistantly at 49,333 now (after the upgrade) is
>connects at 46,666 others have reported that connects are now in the 30'ks.
>Anyone else?
Since we made the change to 3.8.2c2 from 3.8.2, we havent seen much overall
change at all...
Before
v34 54.62% 1756/3215
v90 34.15% 1098/3215
v90medium 22.24% 715/3215
v90hi 6.10% 196/3215
21600 6.00% 193/3215
14400 5.23% 168/3215
v90low 4.57% 147/3215
after
v34 55.91% 1301/2327
v90 35.75% 832/2327
v90medium 21.40% 498/2327
v90hi 7.22% 168/2327
v90low 5.46% 127/2327
14400 4.51% 105/2327
21600 3.82% 89/2327
Again, from another terminal server at a different POP... before
v34 54.26% 1158/2134
v90 35.05% 748/2134
v90medium 22.40% 478/2134
14400 7.97% 170/2134
v90hi 6.00% 128/2134
v90low 5.53% 118/2134
21600 2.72% 58/2134
After...
v34 56.02% 1178/2103
v90 34.57% 727/2103
v90medium 18.97% 399/2103
14400 7.56% 159/2103
v90hi 7.23% 152/2103
v90low 6.99% 147/2103
21600 1.85% 39/2103
The changes are hardly statistically signicant, as there is as much
variation from day to day... However, I suspect the changes did break
someone's modem somewhere. There are invariably always some modem revision
that gets hosed in the process.
The stats for disconnects are about the same as well.
It would be *REALLY* nice Lucent, if total retrains and renegotiations were
recorded in RADIUS somewhere...e.g. instead of just
Connect-Info = "44000 LAPM/V42BIS", have something like
Connect-Info = "44000 LAPM/V42BIS 3,5" to capture retrains and
renegotiations.. Or put it in as a VSA... But just record it somewhere.
---Mike
Mike Tancsa (mdtancsa@sentex.net)
Sentex Communications Corp,
Waterloo, Ontario, Canada
-
To unsubscribe, email 'majordomo@livingston.com' with
'unsubscribe portmaster-users' in the body of the message.
Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>