Force Exit on Timer (1 Viewer)

[The_Doc_Man]

I am very much hoping that wasn't when the DoD decided that "Peoplesoft" could do everything for them. That was a Charlie Foxtrot.

That event from post #19 was a few years after the time that the DoD got that PeopleSoft idea. I was actually in the U.S. Navy Reserve meeting with some GS12 and GS13 civilian managers ("the powers that be") who decided to go ahead with the PeopleSoft development program even though I warned them there was a hidden "gotcha" waiting in the wings. It was going to be a trial program for the Reserves to see if we could switch the regular Navy over to it. I had two or three major misgivings because of the lavish claims made for what PS could do. The claims were great sounding but didn't stand up to logic. But I was a lowly contractor who didn't know what I was talking about, or so they said. So I stood back and let them run that project into the ground, to the tune of $10 Bn bucks for the overall effort before Congress pulled that plug.

Seeing what happened with the DIMHRS project, I had very little satisfaction when the project manager for the Navy's part of that fiasco got canned. I helped him carry his belongings to his car and, before he left the office for the last time, he said "I wish we had listened to you." I felt it was the wrong time for an "I told you so" because he clearly knew it. He was actually a fairly decent fellow. He (and quite a few others) just fell for the PeopleSoft sales pitch even though I instantly knew it was going to be a massive CF.

 

[Gasman]

Only now I know what that Charlie Foxtrot is?

 

[Mark_]

That event from post #19 was a few years after the time that the DoD got that PeopleSoft idea. I was actually in the U.S. Navy Reserve meeting with some GS12 and GS13 civilian managers ("the powers that be") who decided to go ahead with the PeopleSoft development program even though I warned them there was a hidden "gotcha" waiting in the wings. It was going to be a trial program for the Reserves to see if we could switch the regular Navy over to it. I had two or three major misgivings because of the lavish claims made for what PS could do. The claims were great sounding but didn't stand up to logic. But I was a lowly contractor who didn't know what I was talking about, or so they said. So I stood back and let them run that project into the ground, to the tune of $10 Bn bucks for the overall effort before Congress pulled that plug.

Seeing what happened with the DIMHRS project, I had very little satisfaction when the project manager for the Navy's part of that fiasco got canned. I helped him carry his belongings to his car and, before he left the office for the last time, he said "I wish we had listened to you." I felt it was the wrong time for an "I told you so" because he clearly knew it. He was actually a fairly decent fellow. He (and quite a few others) just fell for the PeopleSoft sales pitch even though I instantly knew it was going to be a massive CF.

When it first came out I asked ONE question that told me it wouldn't work; "Does this mean each UIC will get its own server to run it?". If they had set it up where each member's information was kept at their UIC and only summary data was pushed up, it would have worked. This is kinda how Facebook works.

Course they wouldn't do something like that... Had to have everything in one big box!

 

[The_Doc_Man]

@Mark_ : Talking about UICs...,

The U.S. Navy Reserve's personnel system was a central repository setup with each field reserve unit having a PC (not of server class, just a well-configured PC) that used various networking techniques to synchronize the Reserve Unit office with the master system, which was deemed "authoritative" - even though the field units each had a stand-alone system. They exchanged transactions to maintain synchronization - and in fact that was the way that the central Naval Reserve system talked to 18 other offices such as NSIPS, DFAS, NRPC, etc. Each UIC (that's Unit ID Code for the uninitiated) PC was NOT authoritative, because the PC was not backed up to a certain, rather strenuous standard and wasn't able to run Windows Server and couldn't other security-related things it needed to run to earn that "authoritative" status. So the field unit's machine was never fully authoritative.

The machine I primarily ran was that centralized authoritative USNR personnel machine. Over the years, it was hosted on VAX 8000-series CPUS, then after DEC got bought out, on a mixed bag of COMPAQ Alpha systems, and then when COMPAQ got bought out, by HP 8-processor Itanium CPUs based on the INTEL Longhorn hardware platform. The Itanium running OpenVMS was an absolute champ, though I understand that when running Windows, it wasn't so good.

The CLAIM made by PeopleSoft is that you would have a near-identical schema in each of several machines and that if you ran a query from ANYWHERE in essentially a STAR network (as opposed to a MESH structure), you could run this putative remote query just fine and PeopleSoft would transparently fetch the required data from the individually authoritative machines to run that query. Of course, it IS possible to do such a thing. However, "practical" was the biggie.

We had the hottest servers available at the time for the PS centralized servers and our daily operations - personnel updates and reports - took over 24 hours to create. A LOT more than 24 hours. To fix that, the schema became more centralized but never could manage to produce timely daily results because all of the handshaking that went on - at WIDE-area-network speeds. After a major hardware upgrade, PS STILL couldn't beat the 24-hour mark. I.e. daily synchronization took more than a day to run. And you COULD NOT skip a day's sync run because it was transactional. Had to be, for accountability purposes. And it was getting worse, not better, because initially PS wasn't participating in the daily sync with the 18 other offices that were national in scope.