I have encountered the egregious breach of that mantra just today.  Read on and have caution!

About a month ago a purchased a new laptop, as my Gateway was LONG in the tooth and was lumbering under the bloat of the latest development tools and frameworks.  It is an Acer Travelmate 8572.  With an Intel i7 64 bit processor (Quad Core!) I expected some great gains in speed and productivity.  Alas, I am not completely happy with it, even at 64 bits it won’t support more than 4GB RAM and the graphics resolution isn’t as high as I want  (but that is all another story).

I had installed Windows 7 Ultimate and have had very good luck with it.  Some incompatibilities and hassles, but overall, a good experience.  However, in the last few days my bluetooth headset has been acting flaky so I decided to install some updated bluetooth drivers.  No joy, I still had intermittent loss of connection.  So I decided to remove the device drivers, reboot and then reinstall.

Removing the drivers, no problem.  I rebooted and then attempted to reinstall the software.  It couldn’t find the hardware. DOH!  Function-F3 activated the hardware and the software started to install,  but first it began by uninstalling the software.  Progress bar got to about 75% and then just sat, and sat, and sat.  I saw plenty of hard drive activity but no progress.  So I decided to check email while I was waiting.  The shortcut said it couldn’t find the thunderbird.exe file that was referenced.  That’s weird!  I followed the link to the folder for Mozilla Thunderbird.  Not there! NO FOLDER!!!! Then my icons started to disappear off of my desktop.  So, I immediately shut down the system.  I restarted and now the entire desktop was gone and I was down to just some basic folders and programs on the start menu.  Basically nothing left in the Program Files folder.  The system was hammered.  After pondering what I should do and thinking I had picked up some nasty malware I started a System Restore.  It ran for about 45 minutes.  While I was running the restore, I decided to Google WIDCOMM and uninstall and delete and I found this:

http://social.answers.microsoft.com/Forums/en-US/w7performance/thread/ffe5f236-da04-4c3c-8df9-696080ec9d0c

Basically the WIDCOMM uninstall will go rogue and delete the data on the C drive.  This is confirmed by Microsoft (signed driver!)  and so MS is an accessory to hard drive murder IMHO.  The negligence of WIDCOMM in producing such a flawed POS (piece of software) is unconscionable.  These folks apparently are clueless software developers and, no doubt, the software was probably developed off shore by low cost programmers in order to maximize their profits.

The System Restore did very little to get things functioning again.  So, now I will have to reformat and reinstall EVERYTHING!!  Thanks Widcomm!

Anyone game for a Class Action Suit against Widcomm?  I’ll lose at least a day of productivity.  That is worth a few bucks to me….

As an FYI to others that go this route, just treat the http.conf file like any other Apache configuration file and all will be well.  Very pleased that the move was so trouble free (well, once I figured it out).

Next challenge is the upgrade WordPress itself.  I have usually ended up running WP updates manually because of of authority issues with the way the Automatic Update routine runs, but you can pretty much guess that the next post will be on WP 3.0.  Now that the management of the i instance is so much easier, I plan to move the rest of the PHP and Apache applications to the i.  If I could get Glassfish to run on i, I’d move my Glassfish apps as well.  Apache, PHP, Ruby, JRuby and Tomcat all run nicely on i.

However, RTC relies on WAS and my past experiences with WAS have been anything but pleasant.  I have run Tomcat and Glassfish as servlet containers and they install smoothly and run great. WAS, on the other hand, has always been a challenge to install and run.  But, if it is required it is required, so I started down the simple path of getting WAS 7 Express installed.

WAS on i installs multiple ways.  I first took the approach of running the installer from Windows, which *seemed* the easiest.  The installer ran until about the 40% mark on the configuration step, and, 8 hours later, still hadn’t progressed but had not errored out either.  So I shot the installer and restarted.  I quickly learned that it wasn’t quite so simple to end and restart the install.  The installer complained about a .lck file in the ‘tmp’ folder that had to be removed.  However, the installer wasn’t intelligent enough to give me the full path as to where to find this file (as an aside, this ALWAYS aggravates me about installers.  They rarely tell you where they are stuffing things even though they may ask you where to install the software).  After a full scan on my laptop hard drive looking for the mysterious file, it occurred to me that since the installer was installing to the i, perhaps I should look there.  Voila! So I cleaned up the installer files in the IFS and restarted the installer.  Same result. So I cleaned up after install #2 and decided to do some more reading and perhaps download the CD images from the IBM web site to make sure I had the latest and greatest.

And yes, I DID review the prerequisites and made sure I was current on PTF’s and groups.  I had downloaded the WAS Group as well, ready to install it once I had the product installed.  I tried the GUI install one more time with the new CD’s I had downloaded, same result (#3). The next approach I tried, this time using the new CD images, was to attempt an install from PASE.  Although not as sexy as the GUI, I was hoping I would get better feedback on what was going on.  This install (#4) again reached the 40% mark on the configuration step and, after about an hour, dumped an error message to the console (at least I got an error!).  Install fails on configuration action 80Fos400.

Googling turned up very little.  I again revisited the requirements and double checked them.  After one more try (#5) I then submitted a PMR and began  to work with IBM trying to solve the problem.  I will spare you the details, but the next 5 attempts also failed (so now we are up to 10 install attempts).  Finally, an IBM’er jumped on the box which led to a couple of revelations:

1) Even though all the licensed programs indicated that they were happily installed, a command called CHKPGMOPT indicated otherwise.  That lead to two PTF applies that wouldn’t have been uncovered any other way.

2) Although I had downloaded the WAS group PTF, my attempt to apply it resulted in a error about the product not being installed (DUH – I hadn’t been able too).  However, it turns out I was taking the wrong install approach on the group.  I need to take option 8 on the PTF  menu.

3) My path environment variables weren’t correct.  I sorted them out and I doubt it contributed to the fix but, what the hey…

With those actions taken, the install just zippity-do-dah’ed along.  WAS is now installed and happily sucking up memory and CPU cycles.  I guess that the lesson here is that even if you RTFM, the devil is in the details of carrying out the steps.  Counter intuitive as it was, installing the WAS group BEFORE installing the product was the way to go.

The VIOS folks tossed it over the wall to the Linux folks.

The Linux folks determined that I did NOT have a support agreement for Linux on the blade.  They closed the call.

I reopened the call, arguing that the problem manifested itself when using the terminal window to install Linux but the terminal window, being a part of VIOS (IVM) would be part of a supported product.

Silence.

I then, through trial and error, determined it to be a Firefox issue.  I *could* install with IE8 but not with FF.  I reported this back to IBM expecting that they would proceed with determining the cause of the incompatibility and get it fixed.  Au Contraire!  Browser incompatibilities are not part of the mission of IBM support so they closed the call.

I did pick up this tidbit though, and I document it here so I can find it again: You can start the terminal for a given partition in VIOS by issuing the mkvt -ID X command  from the command line where X is the number of the partition.  To exit the partition terminal use the key combination ENTER, ~, . (period).   Whether the VIOS terminal is running in the “Remote Control” window from the AMM or from Telnet or SSH, the commands work.  So now I have a workaround.

I was able to finally get the SLES 11 Linux (PPC64)  partition running and accessible from VNC using a technique found at the Novell site where I posted. Read it all the way through.  The solution is posted at the end of the link in the post: http://forums.novell.com/novell-product-support-forums/suse-linux-enterprise-server-sles/sles-configure-administer/373530-sles-11-vnc-issue-no-desktop-just-gray.html

So that gets Linux (SLES 11 PPC64)  running again.  Now if I can get WAS 7 going on my i, the software application installation challenges for the week are done.

Having experienced a quick and easy install with a “difficult” operating system, I decided to try something easy.  A well supported, well known, easy to install Linux distro.  Since Suse Linux Enterprise Server is supported in the Power PC 64 bit world, I downloaded SLES 11 and pulled up the install instructions and have now spent two days trying to get the distro to install.  I have posted to a few forums but apparently Linux on Power, particularly on a JS12 blade running VIOS and the IVM to install the OS is a bit of a mystery.

The maddening part is that I have done this somewhat successfully before but I am not making much progress at the moment. The issue is getting a terminal window to display so that I can begin the install process. The JS12 implementation guide says to start a Remote control session using the AMM (easy to do) and then activate the defined partition in the IVM (also easy to do), and use the IVM option to “Open a Terminal Window” (another easy step). So far so good. The part that doesn’t seem to be working is the “terminal window” popup in the IVM that is supposed to appear to authenticate only appears for a second or two and then it disappears without an opportunity to log in. The AMM Remote Control terminal window changes from displaying the VIOS console to a blank screen. Seems to me that in my last successful attempt at installing Linux using the IVM I saw all the installation prompts in the remote console of the AMM but I can’t be sure.

The missing terminal window is a show stopper since there is no way to install the OS headless.  In fact, the original install of SLES11 was deleted because I couldn’t get the GUI to ever run, even after installing the OS in text mode (which is no easy feat).

Ah! Would it were that Linux was as easy to install as i5/OS!  Whenever I DO get it installed, I will post the “trick” here (whatever it is) so at least I can find it again…

I have been a VERY satisfied QWEST customer for years.  I don’t know when Qwest originally deployed DSL in my area but I was an early adopter (could it have been 10 years ago?).  In any case, that move from 56k to 1.2mbps was fantastic.  And, in those many years I have never experienced an unanticipated interruption in service.  They have been as close to perfect and unchanging as is possible.  But, there’s is the rub.  No changes. DSL in the 21st century, especially as we begin a new decade, is the 56k of 90′s.  It is just too slow to sustain a growing small business.

I have to also blame Qwest’s wonderful market machine that has dropped monthly “bingo” cards into my mailbox extolling the virtues of “high speed” broadband.  YES.  I *want* high speed Internet up to 7mbps!!!  But, having developed the desire, they couldn’t produce.  After patiently waiting two years, it was time to move, so I pulled the plug on Qwest and switched to Comcast.

Now I gotta admit, this was more stressful than waiting for the birth of my first child.  I imagined all of the things that could possibly go wrong and each of them gave me buyers remorse on deciding to switch.  The installation day came and went with a smooth install of the modem and over less than an hours time I switched over the modem, SBS 2003 and then my DNS entries to the new IP addresses and….nothing broke!  Everything kept humming right along.  The only suspicious wrinkle is that my VOIP traffic seems not to be working on the Comcast network so I need to do more sleuthing to determine the cause (admittedly it took a while to fine tune my Asterisk server even on Qwest).  With a planned Asterisk upgrade in my future, I may wait until I upgrade and then do the troubleshooting.

But wait, there’s more!  Last week I cut the phones over and….nothing broke again!  The phones continued to ring, this time with the included Caller ID (which was extra and therefore not present with my Qwest account).  I know this is all too good to be true and a few of my Facebook friends have warned me of the coming Armageddon but I just can’t get over how easy this was.

To do’s: I had XO handling my 800 lines and due to some arcane rules for Comcast, I need to have a bill with activity on all 3 800 lines before I can swap them over.  But, sometime in July I should receive my very first consolidated Internet/Phone/Long Distance bill and it should be lower than the 3 bills I had before.  That is actually wrinkle #2.  I have gotten my first Comcast bill and I am having trouble reconciling the current bill against the promised costs.  It is about $30 more than I think I was promised.  Oh, and a $99 Comcast special arrived in the mail today.  I have a little bit more than a $99 bill in front of me….

So, it may all just be too good to be true.  We’ll see.  I have faster Internet and free outbound long distance.  Oh, and I also have cable TV, for the first time in my life, so things are looking up (maybe).

With Qwest being sold to CenturyTel (whoever they are) I may have gotten out just in time.  In fact, time will tell…

That is not to say everything was perfect. The austerity measures taken by the board last year did impact the evening activities, particularly the “Ask the experts” night and the “fun and games” night. Without substantial food and at least subsidized alcohol (the staple of the IBM i user) the events were poorly attended. I *can* tell you that the Common staff did their very best to provide as much as the budget allowed but when some budgets were cut completely (like the CaN budget that provided those evening activities) there is only so much you can do. My hope is that next year the CaN budget will receive some attention so that the “community” activities aren’t shortchanged.

The BEST part of the conference were the great, well attended sessions. Although the conference was smaller in the number of sessions offered, it still felt “large” because the facilities were “right sized” for the event. Overall I’d give the conference a B+ grade.

It will be interesting to see how the San Antonio event turns out. It is structured much differently in terms of speaker/volunteer compensation and size. If successful, it could point the way back to the restoration of some of the benefits cut last year (we hope!).

If you missed the Common Conference then you really MISSED a great conference. Make sure you register and get in on the fun in Minneapolis next spring!

Alas!  That is were I find myself now.  *Months* ago I committed to one presentation for the Common Conference in Orlando on May 3rd. Then I found out that the one was actually three.  Then I committed to deliver a totally new module in my time and attendance system. Now I find myself heading out on a two day trip to volunteer for more development work (gratis) and wanting to add yet another consulting trip in the next couple of weeks.  What is wrong with me???  I haven’t even touched my new presentation yet and have no idea of the amount of code I’ll have to crank out in order to get it done.

On top of that, I am *blogging* when I should be working!!!  I have to stop right now and…

Adding to the mix is the need for checking to see if the clocks are online, updating the data en-mass and clearing data that has been retrieved without losing anything.  Still working through that as well.  The UI is pretty much done but I am not 100% happy with it.  The difficulty is viewing the time clock “punches” in a way that is easy to view.  If someone punches in and out 8 times in one day, how do you present that?  As a horizontal row or as a set of vertical transactions?   Best way to handle it is with agile, iterative steps I think.  See how it works and then improve it.

Lots of chatter about Flex over on the midrange.com list today.  I have seen some demos and sooner or later I’ll take the plunge.  But right now, jQuery and Web 2.0 techniques are my bread and butter for now.  I need to get the application done, and sell a few dozen, before I even think about moving to a new development platform.  All of the data I/O is using Ajax at the moment.  Seems like the most flexible way to go.

I have about a week or two left with this project and then I need to return to ASAAP for some additional polishing work and then jump into the IVR replacement.  Nice to have work, but it would be better to have some income (still working on that!).

You know that I have a Java based Absence Management System called ASAAP .  I have recently been contracted to develop an automated time clock system and make it work seamlessly with a payroll package.  I naively figured it would be a snap.  This being the 21st Century, I figured I’d just find a electronic time clock  that had a JDBC compliant interface and within a couple of weeks hook it up to ASAAP which already has payroll posting capabilities.  After poking around the Internet, attempting to contact sales people and gather information, I settled on a clock that had a “polling” package to collect the data on a scheduled basis and also push the employee information up to the clock.

After receiving the clock and having a chance to try it out, I was less than enamored with the user interface and my attempts to interface with the DB were frustrated by the fact the DB was an MS Access DB and the JDBC-ODBC bridge is slow and unreliable.  What I wanted was something that allowed secure access to the data over a network.  So, back to square one.

The company I had purchased the timeclock from had a “barebones” configuration that basically had nothing but a scripting api that accessed clock functions in a embedded Linux distro called Busybox. So, I went with the “start from scratch” approach and have been learning (and re-learning) Linux commands and their particular implementation in Busybox.

This tiny footprint Linux distro is pretty cool and the folks on the Busybox mailing list have been very tolerant of this noob as I have bumbled my way through the learning curve.  At this point I have a functioning application that will take input from the keypad or the Proximity badge scan (RFID), look up the employee in the database, display the employee name, prompt them for a punch in or out and store that information in the database.

Next step is to work on data import/export.  Unfortunately the Busybox distro I am working with has a relatively old version of Sqlite3 which doesn’t support JDBC, so my plan is to use the HTTP server and CGI to post data to and retrieve data from the DB.  THAT should be fun!

So I’ll keep you posted on the progress (or lack thereof).  All in all, Busybox provides enough to work with to build this application, I just need the smarts to make it work.