[TriLUG] Griping about Time-Warner outage

[matt at noway2.thruhere.net]

It has been my experience too that TWC outages are caused primarily by a
failure of DNS.  It is easy enough to verify with a numeric ping.

Since switching to my own DNS (BIND) I haven't had this problem.  I have
had to power reset the cable modem a few times since I switched to
business class, whereas my residential Toshiba went a couple of years
without a reset.

With regards to their poor customer service, I have also found that the
business class support is FAR better than the residential.  The
residential internet support is heavily oriented around the PEBKAC model
and fails when it is not.  For example, with my recent issue regarding
their email relays getting put on RBL lists, when I contacted BC support,
they asked me to show them a copy of the problem with a pastebin like
site, which they then acknowledged that there was a problem, verified it
with MXToolBox, and brought their security department in to the mix, and
created an internal work order to have the culprit identified and shut
down.  The residential help desk kept saying, "we don't see a problem with
the account sir.  If your having problems with receiving spam you need to
get a filter program."

> Not everyone will want to go to the lengths that I have to deal with flaky
> ISP DNS.
>
> I have had trouble with TWC DNS as well as DNS provided by a couple
> previous
> ISPs I have used. I use an internal caching name server with Google
> (8.8.8.8 and
> 8.8.4.4) as the forwarders. I have not experienced a DNS outage since
> starting
> that. I have tried OpenDNS as my forwarders, but they do too much
> filtering of
> "undesirable" IP addresses so I gave up on them. My "undesirable" is not
> the
> same as theirs. ;-)
>
> I have recently started using my DNS server for internal name resolution
> instead
> of /etc/hosts files, but that is intended as a learning experience for me.
> And I
> have learned a lot about DNS that way.
>
>