[nSLUG] Re: Troubleshooting LCP & PAP on a Telus dialup NAS

Mike Spencer mspencer at tallships.ca
Tue Dec 9 01:56:06 AST 2014


Well, thanks, Dop, for your thoughtful reply.

At the risk of boring the list, here's the (lack of) progress:

I finally got both "escalation specialists" on the phone at once. [1]
They said that they (or someone) had installed the dialup software
"out of the box" -- exactly what I had suspected -- and that it
"worked for everyone else".  But they were unable to tell me exactly
what the "OOtBox" configuration was.

That inability notwithstanding, they were firm that I had to
authenticate using the prompted ASCII interface before starting
LCP/PPP.  That doesn't work, either manually in Minicom or in a script
called by pppd.  The now-defunct Telus NAS was originally configured
like that, was changed to the "ignore prompt, go directly to LCP/PPP"
mode in 2012 without notice. The "escalation specialists" are sure, on
the word of their nameless engineer, that the ASCII/prompt mode is now
in effect.  It doesn't work with any of the candidate passwords (see
below).

They both opined that I knew more about this stuff than they did.
There were adamant that I couldn't talk to the "engineer" with
relevant responsibility.

It gets worse. Originally, we had:

     Login password:  foo
     Mail-1 password: foo
     Mail-2 password  bar

At some point, years ago, we felt that the login password had been
compromised so we used the ca.inter.net HTML interface to change
it. The result was:

    Login password:  baz
    Mail-1 password: foo
    Mail-2 password: bar

which has been the state of affairs for years. The two mail accounts
can still be accessed on the mail servers (when connected via a
different ISP) with the same foo and bar passwords.

But now the "escalation specialists" tell me that the login password
is once again (or, according to them and erroneously, has always has
been) the original foo.  Moreover, they say that my wife has been
typing a '=' in foo where a '-' is correct.  Only she never types
anything, just uses a Win-XP "thing" that hasn't changed in years.

It's interesting (to me, anyway) to note that, in resolving the 2012
mode switch, a support person (not these guys) was sure that my login
password was foo when it had not been and proved (after accommodating
the login mode change) not to be so then.  Already, in 2012, something
was screwed up -- what support could see was not the effective facts.

So as far as I can see, 

   + The "escalation specialists" don't actually know how *anything*
     actually *works*.  They can query their RADIUS server and other
     databases or logs but have no clue what any of it means.

   + Their relevant engineer is so busy rolling out cool new user
     experiences, digital telephony, colo, cloud, big data, SaaS
     etc. etc.  that he doesn't have time to troubleshoot old tech for
     a customer of 20 years standing.

   + Their authentication data is probably fragmented and has never
     been correctly ported from Uniserve (previous owners of
     ca.inter.net) servers to Fibernetics servers.

I think it may be time for a registered letter to their CEO asking to
be employed to work from home at $300/hr as their dialup technology
specialist, requiring I be sent all manuals for all relevant software
and hardware.  I don't really want such a job but I could do it better
than it's being done now and a few weeks at $6000/wk would be nice.

This has totally distracted me from more Linux-relevant stuff such as
compiling a new kernel.  If I weren't listening to Bach's cello suite,
this would probably be written in ALL CAPS.  Sigh...  Sorry.

Thanks, again, for your thoughtful reply.  If you have any suggestions
for social engineering, short of applying for a job, suitable for
provoking meaningful support from Fibernetics, I'm all ears.


- Mike


[1] http://www.fibernetics.ca/team/

    Scroll down to Neil Clarke and Dave Majury.  I don't know which,
    if any, of the folks pictured there is the relevant "engineer".


More information about the nSLUG mailing list