[SATLUG] Grande Comunications down last night
firestorm.v1 at gmail.com
Sun Jan 11 14:59:20 CST 2009
As someone that works in the cable industry, there is something I need
to say on this issue. I don't think that a lot of it applies, but a
well-informed customer is a smart customer. Since I'm in Houston and
went through Ike, I got to experience firsthand how uninformed people
are about how the cable network actually works.
First off, a little bit of terminology:
Headend - the point where the cabl signals are originated. These are
the large windowless buildings where the server farms, muxes,
switches, signal aggregators, etc are that generate the cable TV, and
high speed data signal. The headend has a backup generator that runs
on diesel and is staffed with technicans in the event of a major
disaster. A typical headend has several large satellite dishes and
Hubsite - This is a small windowless building, usually about 400sqft
that has additional equipment for the more localized area such as VOD
server farms and vaults, fiber receivers and transmitters, and the
digital upstream and downstream for the settop boxes and high speed
data services (both digital phone and high speed data)
Node - This is the "neighborhood endpoint" where the RF signals are
translated from the fiber into coaxial RF and is sent down the coax to
the taps and eventually ends up at the customer site. The node gets
its power from the coaxial cable from power inserters.
Power Inserter - does what the name implies. These devices are small
cabinets that contain a voltage regulator and several batteries (24V
marine deep cycle batteries if memory serves). The cabinets are
usually about 3ft tall, grey and have a utility meter on them or close
by. They are connected to utility power and are where the coax plant
get their power for operating the nodes, line extender amplifiers and
system amplifiers. There are many power inserters for a single cable
plant, depending on how large the area is that is being served by the
node. Think of the coax plant as one big buss system not too unlike
10base2, where the power inserter is not necessarily next to the node,
but it is somewhere close.
Tap - This is the point where the customer's drops are connected to
the cable RF plant. These devices are usually in the backyard or up
on a pole. The underground covers are about a foot in diameter
cylinder that is in the backyard every two to three houses depending
The layout is something like this:
During Ike, where commercial power service was out, we had deployed
many generators to the power inserters to keep the cable plant
operational and many of our techs pulled insane hours repairing and
verifying that the fiber plant and the coax plant were operational.
We had full plant (all nodes responding) within I believe four days of
Ike's passing, and the Ike-related outages were resolved a week later.
Our biggest hurdles were people stealing our generators, destroying
the power inserters for the batteries, flooding and the lack of
utility power. We had almost 1,200 techs in the field for the two
weeks after Ike to get as many trouble tickets resolved.
It's not that the ISP's treat the Internet any less than the other
services that offer (at least not in Houston), far from it. Our
network is constantly being monitored for trending and a tech is
dispatched if 10% of a node's cablemodems drop offline. If we get more
than 15 people calling in to open video issues on a single node, then
we roll a tech to address the customer's issues and also roll a plant
tech to address the plant to see if there's something else that could
have caused it.
Now, I can't speak for any company, (not even the one I work for) but
I can tell you that we do everything in our power to ensure that our
services are available as much as possible but sometimes the reason
for the outage is something we have no control over.
just my .02c, well actually a bit more, lol, guess I got wordy.
On Sun, Jan 11, 2009 at 12:55 PM, Henry Pugsley <henry.pugsley at gmail.com> wrote:
> On Sun, Jan 11, 2009 at 12:13 PM, Bruce Dubbs <bruce.dubbs at gmail.com> wrote:
>> Jennie Haywood wrote:
>>> On Sun, Jan 11, 2009 at 9:07 AM, Todd W. Bucy
>>> <toddwbucy at grandecom.net>wrote:
>>>> at about 5:00 pm I started noticing that we were having problems with
>>>> our Internet service by 6:00pm we noticed that our cable tv service was
>>>> down. upon calling Grande they said that they were currently having
>>>> technical difficulties in the San Antonio Area. Service returned to
>>>> normal by about 8:30 pm. Anyone on the list have any idea what happened?
>>> I don't know what happened, but the service was sporadic for those hours.
>>> Everything was out - internet, phone and cable tv for about an hour.
>> This is exactly the problem I have with internet based phone systems like
>> Vontage or Time Warner's digital phone. I feel the most reliable phone is
>> still copper from AT&T.
>> -- Bruce
>> SATLUG mailing list
>> SATLUG at satlug.org
>> http://alamo.satlug.org/mailman/listinfo/satlug to unsubscribe
>> Powered by Rackspace (www.rackspace.com)
> Copper is going the way of the dinosaur. With AT&T pushing U-Verse,
> the only copper involved is the possibly 100 yard run from the FiOS
> box in the neighborhood to your house. Up here in Dallas, AT&T only
> provides 8 hours of battery backup in the event of power loss. People
> without U-Verse may not have copper either .. since the late 1990s the
> telcos have been running fiber from the CO to neighborhoods, then
> plugging into multiplexers to provide copper to the houses (one of the
> reasons DSL provisioning sucks).
> Personally I've never had an outage that has affected my VoIP for the
> year or so I've been in this location, but I also know that there is
> no SLA from Time Warner to keep my Internet access up for my phone.
> In San Antonio there were frequent cases where the power went out in
> the neighborhood which took down the cable head-end .. I had battery
> backup on my stuff, but there was nothing alive on the other side.
> After Time Warner started selling digital phones, they added backup
> power to the head-end and a battery pack in the actual cable modem
> (not sure of how long it lasts).
> Unfortunately Internet access is not simply a convenience or a luxury
> now, but something that people may depend on for their lives. ISPs
> need to take this into consideration when they build their
> infrastructure and setup SLAs.
> SATLUG mailing list
> SATLUG at satlug.org
> http://alamo.satlug.org/mailman/listinfo/satlug to unsubscribe
> Powered by Rackspace (www.rackspace.com)
More information about the SATLUG