ccnp

All posts tagged ccnp

I’m thinking of doing some video blogging and kicking it off with a series with my thoughts on technical certifications.  Are they valuable or just a vendor racket?  Should you bother to invest time in them?  Why do the questions sometimes seem plain wrong?

Meanwhile, a little Netstalgia about the first technical certification I (almost) attempted:  The Apple Certified Server Engineer.

Back in the 1990’s, I worked for a small company doing desktop and network support.  When I say small, I mean small.  We had 60 employees and 30 of them had computers.  Still, it was where I first got into the computer industry, and I learned a surprising amount as networking was just starting to take off.

I administered a single AppleShare file server for the company, and I even set up my very first router, a Dayna Pathfinder.  I was looking for more, however, and I didn’t have much of a resume.  A year and a half of desktop support for 30 users was not impressing recruiters.  I felt I needed some sort of credential to prove my worth.

At the time Microsoft certifications, in particular the MCSE, were a hot commodity.  Apple decided to introduce its own program, the ACSE.  Bear in mind, this was back before Steve Jobs returned to Apple.  In the “beige-box” era of Apple, their products were not particularly popular, especially with corporations.  Nonetheless, I saw the ACSE as my ticket out of my pathetic little job.  I set to work on preparing for it.  If memory serves (and I can find little in the Wayback machine), the certification consisted of four exams covering AppleTalk networking, AppleShare file servers, and Backup.

Apple outsourced the certification development to a company called Network Frontiers, and its colorful leader, Dorian Cougias.  I had seen Dorian present at Macworld Expo once, and he clearly was very knowledgeable.  (He asked the room “what’s the difference between a switch and a bridge?” and then answered his own question.  “Marketing.”  Good answer.)  Dorian wrote all of the textbooks required for the program.  He may have known his stuff, but I found his writing style insufferable.  The books were written in an overly conversational tone, and laced with constant bad jokes.  (“To remove the jacketing of the cable you need a special tool…  I’d call it a ‘stripper’ but my mother is reading this.”  Ugh…)  A little levity in technical documentation is nice, but this got annoying fast.

This was in the era before Google, and despite my annoyance I did scour the books for scarce information on how networking actually worked.  I didn’t really study them, however, which you need to do if you want to pass a test.  I downloaded the practice exam on my Powerbook 140 laptop and fired it up.  I assumed with my day-to-day work and having read the book, I’d pass the sample exam and be ready for the real deal.

Instead, I scored 40%.  I used to be a bit dramatic back in my twenties, and went into a severe depression.  “40%???  I know this stuff!  I do it every day!  I read the book!  I’ll never get out of this stupid job!!!”  I had my first ocular migraine the next day.

In reality, it doesn’t matter how good or bad, easy or hard an exam is.  You’re not going to pass it on the first go without even studying.  And this was a practice exam.  I should have taken it four or five times, like I learned to do eventually studying Boson exams for my CCNP.

Instead, I gave up.  And, very shortly later, Apple cancelled the program due to a lack of interest.  Good thing I didn’t waste a lot of time on it.  Of course, I managed to get another job, and pass a few tests along the way.

I learned a few things about technical certifications from that.  In the first place, they can often involve learning a lot of knowledge that may not be practical.  Next, you can’t pass them without studying for them.  Also, that the value and long-term viability of the exams are largely up to the whims of the vendors.  And finally, don’t trust a certification when the author of the study materials thinks he’s Jerry Seinfeld.

 

As a part of my job at Cisco I’ve been looking into Zscaler and their offerings.  It started me thinking back to the early days of remote access, and I figured it would make a good topic for Netstalgia.

I wrote in the past about how bulletin board systems (BBSs) work, and in another article I resurrected my old BBS in an Apple II emulator.  In a nutshell, a computer with a BBS set up had a modem on it and users dialed in using their own modem over dial-up phone lines.  I’m not sure how many readers are young and don’t remember modems, and how many are dinosaurs like me, but as a reminder, modems connect computers to phone lines.  One modem is set to answer any call that comes in, and waits.  Then another user with a modem inputs the phone number of the other end into his software.  His modem dials out, the phone rings, and the other modem answers with a carrier tone.  Then the dialer responds and after some negotiation on the line, a connection is established and data is sent.

Now in my first job, at a small company in Marin California in the mid-1990’s, we had one computer set up as a dedicated remote access server.  It had a single modem with a single phone line, and ran Apple Remote Access server, since we were a Mac shop.  We only had one user with a laptop, the CEO, so when he traveled he would dial-in and be able to access basic functions like email and our file server.  There was no Internet access back then.

When I moved on to a consulting company, I did a few more industrial set ups.  Usually these involved remote access servers that were comprised of a bunch of modems and a LAN port.  The remote access server would accept a bunch of phone lines and then provide TCP/IP or AppleTalk connectivity to the network.  By this time users had Internet connectivity.  The Shiva LanRover is one example of this sort of device.

Shiva LANRover

When I worked at the San Francisco Chronicle, we had an Ascend Max which served this purpose.  The Max had two DS3 lines plugged into it.  It was the first time I had seen a DS3, and I remember being excited to learn the phone company could deliver a circuit over coax.  (It actually entered the building on fiber and went over coax from the MPOE.)  The DS3 was an ISDN PRI, with 24 dial-in phone lines multiplexed over a single digital circuit.  It took me months to find someone who had the password to the Max, and when I finally got in I found out that the second DS3 was unconfigured.  Users had been complaining about busy signals and all I had to do was change a few menu settings.

Ascend Max

Remote access dial-up was heavily used at the Chronicle.  Reporters filed their stories via modem.  VPN was just coming out, and I decided to replace the dial-up with VPN + dial-up.  A company called Fiberlink provided a dialer with a vast database of local Internet dial-up lines from a variety of carriers they contracted with.  Our users would pick a local phone line and then dial into it.  They then launched our Nortel VPN client to establish connectivity.  This saved us a fortune on 800-number charges, but our users hated it.  As a good senior guy, I did the initial design and left implementation to a junior guy.  I’m amazed he still talks to me.  (And he’s not junior anymore!)

Despite being a long-time Cisco guy, I never touched the Cisco remote access stuff.  I did use 2500-series routers with serial ports as terminal servers in the lab, but I never connected modems to them.  Still, when I passed my CCNP, one exam covered remote access and I needed to know a lot about modems.

Nowadays I rarely log into VPN.  Most systems I need to access can authenticate through our Zero Trust/SSO system without the need for a connection to Cisco’s network.  We’ve come a long way since the days of dial-up.  And while I said I missed wiring in another post, I sure don’t miss modem tones!