[TriLUG] RR Network Reliablity

[jonc]

Time Warner is a "bulk" internet provider. Their infrastructure can
handle QoS and CIR (Committed Information Rate) but their business model
can't. 

We've been hammering on them for years to offer higher priced
connections that give users of streaming services (like VoIP) better
service. Their solution is always to throw more bandwidth at the user -
which never, never solves the problem (latency spikes caused by
over-subscription of their distribution switches and lack of simple
reservation queues on those distribution switches).

Time Warner also lacks accountability for providing quality service -
especially on accounts that need consistent low latency. We have had
clients that ran fine on RoadRunner for years then suddenly it all goes
to heck: dropped packets, long latency spikes, etc. Time Warner's only
response in that situation is to send a tech out who does a bulk
bandwidth test and then declares that there is no problem...

I guess that client just ran into a fluke that lasted for a couple of
years - kind of a Roadrunner "el Nino" of calm internet access.

So when it comes to Time Warner, *know* what you are buying. It's bulk
internet access for a nice cheap price. Not something you generally want
your  network to be dependent on.

Jon Carnes

On Tue, 2005-11-08 at 08:36, Chad Thomsen wrote:
> Interesting. Is anybody running RR business class at work? Reason I ask is
> my Frame Relay contract for seven sights is up for renewal in about 4
> months. Two years ago I looked into RR business class but they did not have
> the lines run to my facilities and they wanted me to incure the costs of
> running the lines so I said no and stuck with frame. I also could not
> swallow the pill of having no CIR or QOS since we run VOIP. I doubt things
> (performance) has changed that much in two years. I also spoke with a guy at
> a conference who switched to Broadband business class only to swith back to
> frame due to lack of CIR and QOS.
> 
> Any suggestions would be appreciated.
> 
> CHad
> 
> On 11/7/05, Jeff Groves <jgroves at krenim.org> wrote:
> >
> > I wouldn't tolerate any ping losses to any of the RoadRunner servers
> > such as their news, mail, or smtp servers.
> >
> > Jeff G.
> >
> > Ron Joffe wrote:
> >
> > >What would be expected as far as reliability?
> > >
> > >Ron
> > >
> > >On Monday 07 November 2005 12:06, Jeffrey A. Groves wrote:
> > >
> > >
> > >>This is a valid test.
> > >>
> > >>Call RoadRunner tech support and ask for a second level technician as
> > soon
> > >>as they answer the phone.
> > >>
> > >>Work with them until they see the packet losses too. It may take a
> > while,
> > >>but take the time.
> > >>
> > >>They followed this procedure when this was happeneing to me:
> > >>
> > >>1) Dispatched a technician to my house to check the drop from the pole
> > >> Result: technician replace the coax from the telephone pole to the box
> > >> on the side of my house, but did not fix the problem.
> > >> Technician opened a ticket for a problem on the main line.
> > >> This ticket was pencil-whipped a day or two later -- problem
> > >> still was occuring
> > >>
> > >>A week later...
> > >>
> > >>2) Dispatched a technician to house to check the local drop again --
> > >> Result: found everything in good order on the drop and opened
> > >> a ticket on the main line. This time they actually
> > >> worked the ticket. Two days later problem was fixed.
> > >>
> > >>This resulted in getting the cable modem problem fixed and a noticeable
> > >>improvement in my Cable TV signal in the house :-D
> > >>
> > >>Jeff G.
> > >>
> > >>Ron Joffe said:
> > >>
> > >>
> > >>>Hey folks,
> > >>>
> > >>>Looking for a suggestion. I have noticed that my VPN tunnels seem to
> > drop
> > >>>on a random order.
> > >>>
> > >>>So I have set up a simple ping cron job, I have it sending out 60 pings
> > >>>every minute to three servers (one on the RR network {DNS}, one on the
> > >>>east coast, one on the west coast).
> > >>>
> > >>>The job is run every minute, so in fact this is a continuous ping that
> > >>>sends results every minute.
> > >>>
> > >>>I see that of my 60 packets, I loose a few almost once every 3 minutes.
> > >>>This seems to affect all three servers.
> > >>>
> > >>>So my questions are:
> > >>>
> > >>>1. Is this providing me with a valid network test?
> > >>>2. What kind of results should I expect?
> > >>>3. What type of network reliability test would you recommend?
> > >>>4. What type of network reliability should I expect?
> > >>>
> > >>>Thanks,
> > >>>
> > >>>Ron
> > >>>
> > >>>
> > >>>
> > >>>--
> > >>>TriLUG mailing list :
> > >>>http://www.trilug.org/mailman/listinfo/trilug TriLUG Organizational FAQ
> > >>>: http://trilug.org/faq/
> > >>>TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
> > >>>
> > >>>
> > >>--
> > >>
> > >>Jeffrey A. Groves jgroves at krenim.org
> > >> Web Site: http://www.krenim.org/
> > >>
> > >>
> >
> > --
> > Jeff Groves
> > email: jgroves at krenim.org Web Site: http://www.krenim.org/
> >
> >
> > --
> > TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
> > TriLUG Organizational FAQ : http://trilug.org/faq/
> > TriLUG Member Services FAQ : http://members.trilug.org/services_faq/
> >