![]() |
Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
Rich Braun wrote: > Starting about 10 days ago, my cable-modem connection started acting up. I > was seeing high packet loss and occasional loss of sync (the "cable" link > light would go out for up to a minute of time). > > Had service techs come out from Comcast last Friday; the first one yanked out > the short piece of cable connecting my old Best Data CMX110 to the wall jack > and said "aha, here's your problem right here" pointing to a loose connector. > Well that wasn't the problem so hours later I had a more-competent tech come > out. Both techs reported strong signal quality; the latter did a > more-comprehensive DOCSIS test at the packet level. Scratching our heads, I > had the tech take a look at my own Linux-based test: we concluded the only > viable test at that point was to swap out the CMX110 cable-modem unit. I > rented a Motorola SB5120 for the month to see if that works better. > > Well it sure does! I now get over 10 megabits/sec in download speed, and > don't see the dropouts and packet loss. > > So I'll get rid of the old cable modem and get a new one. It's not obvious > what failed but my hypothesis is that Comcast quietly did an upgrade of their > network that rendered my old cable modem unusable. It has a 10BaseT > connector; I'm theorizing that the speed upgrade simply choked the poor thing, > in a way that made it appear that something--almost anything--else besides the > cable modem was the culprit. > > Anyway I'm posting this here to share my experience and see if perhaps my > hypothesis proves correct over time. If so it'll save a lot of aggravation: > the service techs really don't have much expertise so you have to figure these > things out pretty much on your own. I'm more than happy with the Comcast > network's performance and reliability but it'd be nice if this mystery were > solved. > > -rich >
![]() |
|
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |