[colug-432] DNS delays for IPv6

[Angelo McComis]

Mine is a Motorola SurfBoard DOCSIS. At 192.168.100.1, I see Signal strength
is not a field of information on mine, but Signal-to-noise ratio is, and
mine is 37.9dB...

I hear less than 36dB can be an indicator of problems.  (My power level is
-1.8 dBmV)

For what its worth, broadbandreports.com has FAQs with loads of useful info
on t-shooting cable modem connectivity issues.

Bonus FWIW: Apparently columbus.rr.com (Columbus franchise of RoadRunner)
does allow one to BYO cable modem. Not sure if it saves any money or not,
but on the FAQ, it says Columbus allows it.


Rick -- I would point out that beyond the initial NS lookup on the host
name, pages should not be slow. These days with pages calling in ads and
plug ins and what-not from facebook, doubleclick, etc., a given page might
do 10 or more lookups, but once they're done, they should be cached locally
according to their TTL rules (or your own rules for caching that might
override what's in TTL).  Coding those addresses locally into /etc/hosts
explains immediate relief on first lookup, but doesn't explain why
subsequent lookups (before adding to /etc/hosts) were slow.

-----Original Message-----
From: colug-432-bounces at colug.net [mailto:colug-432-bounces at colug.net] On
Behalf Of Judd Montgomery
Sent: Tuesday, June 19, 2012 11:37 AM
To: colug-432 at colug.net
Subject: Re: [colug-432] DNS delays for IPv6

It could be the signal strength.  To check open a web browser to the 
cable modem, usually http://192.168.100.1  The default passwords can be 
found on the web.  Try admin/motorola.  Then look at the power levels, 
probably under the connection tab.  The power levels should be between 0 
dBmV and -10 dBmV.  If you get much below -10 there will be connection 
problems and a tech can come out and fix that.

Judd

On 06/18/2012 09:43 PM, Rick Troth wrote:
> Can someone lend me a clue stick?  (Or perhaps just slap me with it.)
>
> I got a new cable modem.  Since that time, I've had degrading internet
access.
> To be specific, service went from intermittent extremes (all there, or
> nothing there) to slower and slower.
> IMPORTANT COINCIDENCE:  The new hardware was installed (by a TWC tech)
> on June 7, one day after "World IPv6 Launch".  So I cannot tell for
> sure if the problem is with the new hardware or is because of a change
> in the network.
>
> The most obvious symptom is slow loading web pages.  Most other
> services (XMPP, IMAP, SSH, PING, so on) seem to be working fine.
> Google has it worst, but other sites too.  Direct 'wget' of some files
> has been immediate, once the hostname resolved.
>
> Two experiments suggest that it is #1 related to IPv6 and #2 related
> to DNS.  Of note, I disabled the IPv6 address on my daughter's machine
> and it immediately got better.  (No details, just her report.)  Since
> Google was/is a big hurt, I collected the IPv4 addresses for a number
> of its hosts and hard-coded them in /etc/hosts and here too got
> immediate relief.
>
> I cannot find news or articles or reports of similar experience.
> (Certainly possible that it's just me.)  My home network has an IPv6
> /48 routed over a SixXS tunnel which has worked well for more than a
> year.
>
> I run my own 'named', forwarding to the servers provided by the ISP
> (TWC/RR).  I don't see delays with 'nslookup'.  But when a given web
> page is "slow", the behavior is as if it were resolving fetchable
> objects in the page.
>
> -- R;<><
> '::1, not-so-sweet ::1'
>
> _______________________________________________
> colug-432 mailing list
> colug-432 at colug.net
> http://lists.colug.net/mailman/listinfo/colug-432
>
>
> !DSPAM:4,4fdfddd176531423178057!

_______________________________________________
colug-432 mailing list
colug-432 at colug.net
http://lists.colug.net/mailman/listinfo/colug-432