See bottom of message

Levi Smith wrote:
> OK, Something is REALLY messed up here....
> I rebooted the router, modem.  No change...
> 
> I went up front and tried another computer.  No problem... What???
> So I said, well let's just try reboots on these computers.  The problem is
> gone...
> 
> What the heck is going on??
> This was two completely different computers.  One with Linux, one with
> Windows, in both IE and Firefox.
> They both stopped working at the same time...
> Only via http to a single site...
> 
> Man, computers suck.  (:
> 
> Levi (:
> 
> On 5/17/06, Gus Wirth <[EMAIL PROTECTED]> wrote:
>>
>> Levi Smith wrote:
>> > Not that I know of.  I am the IT department for about 12 computers that
>> > make
>> > up our network.  I haven't touched our Linksys router or the DSL modem
>> > (seemingly the only things that would affect all the computers at
>> once),
>> it
>> > was working this morning, and then it wasn't.
>> >
>> > Oh, and one other tidbit.  It does on a RARE occasion actually come
>> > through.  (like 1 out of 20 or more times), but it's dead again by the
>> time
>> > I can try another page...
>> >
>> > I may try rebooting the router and modem just to see if it makes a
>> > difference...
>> >
>> >
>> > Thanks!
>> > Levi
>> >
>> > On 5/17/06, Michael J McCafferty <[EMAIL PROTECTED]> wrote:
>> >>
>> >>
>> >> Proxy server ?
>> >> Firewall ?
>>
>> Your ISP may be running a caching proxy without you knowing it. There
>> are some ISPs that do this so they can claim "faster" Internet access.
>> If it has a problem, you have a problem. Call them and ask them directly.
>>
>> Also, I have occasionally had problem with my Linksys router (WRT54G)
>> where if I have something like bittorrent running on the inside, even if
>> it's throttled, will slow down tremendously. I think it has problems
>> with the routing table when there are a lot of connections. A power
>> cycle will usually clear it up.
>>
>> Gus
>>
>>
>>
>> -- 
>> [email protected]
>> http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
>>
> 

<aside>
I think there were 2 separate top-posters in this thread which makes it
hard to follow the history, *and* perhaps a good example of why
bottom-posting is generally recommended on technical mailing lists.
</aside>

..anyway, might this be something like a stale (incorrect) internal
routing cache on both linux & windows boxes. How could that have
happened -- well, maybe there was a change to internal routing or
nameserver setup during the experience? I donno - does anyone know:
could an external routing/dns switchover show up as this symptom?

(I presume the actual destination itself didn't change IP address, or
you would have probably mentioned that.)

It may have been /interesting/ to do some command line pings to IPs and
DNS-names when the symptom existed to see what kind of error you got --
"unknown host", "destination unreachable", "no route to host", ..?

On a linux box route -nC or route -C may have been informative.

The command
  ip route flush cache
might be worth a try if you see this again.  Seems like there should be
 a way to do that in windows, too?

If that were to fix the problem, it would be nice to compare the data
shown by "route -C" from before and after.

..jim


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to