FREECHAIN

[Don Coleman]

> -----Original Message-----
> From: Steve Wiltsie [mailto:swiltsie at micro-mui.com]
> Sent: Monday, July 02, 2007 2:58 PM
> To: Don Coleman
> Subject: Re: FREECHAIN
> 
> Don,
> 
> What is the possibility that you have:
> 
> 1.  a 'write' after the 'delete' statement?
> 2.  a unusual ascii character coming in from one of the alien files that
> 'confuses' filePro's normal freechain maintenance process?
> 3.  a 'bad' record in the file that will trip up the normal freechain
> maintenance process no matter what you do with the other records?   Does
> the
> main file ever get cleared completely and started over?
> 
> Just a couple of thoughts.
> 
> Steve Wiltsie
> microCONCEPTS Div. multi/user/inc.        Since 1981
> (812) 474-0127                          Fax: (812) 474-0147


Steve:

Your second scenario makes me wonder.  

An unusual ascii character ...

I ran into this before 1-2 years ago and was able to locate the problem
character although I don't remember how I did or remember the specifics.  I
know Ken B. suggested a program called TR.EXE and provided the syntax for
having it scan the file to remove this character, and copy the remaining
contents to a new file.  This new file then was the file imported.  The
syntax Ken provided was :

tr -d "\0" <qs1data2.txt > qs1data.txt

I seem to remember the "\0" being some type of null character but I may be
mistaken.  This file can become rather large with 50-60 fields and up to
20,000 lines so finding the problem character is a little like finding the
needle in the haystack.



Don Coleman
Donald G. Coleman, Consultant
402 Andrew Circle
Indiana, PA 15701
dcoleman at dgcreact.com
(724) 349-6302
> 
> ----- Original Message -----
> From: "Don Coleman" <dcoleman at dgcreact.com>
> To: "'Filepro-List at Lists. Celestial. Com'"
> <filepro-list at lists.celestial.com>
> Sent: Monday, July 02, 2007 1:44 PM
> Subject: FREECHAIN
> 
> 
> >I posted this early last week and had a number of people reply stating
> this
> > is unusual and I should be concerned about its cause.  I know I have not
> > posted the code, primarily because I don't expect a contributor to this
> > list
> > to sift through it all.  In fact, this client has now advocated a set
> time
> > each day to get everyone off their system for 10 minutes after lunch and
> > run
> > freechn.exe plus do several file index rebuilds as a preventive when
> they
> > are less busy to hopefully eliminate the need to do it near the end of
> the
> > day when they need every available minute.  I always thought the freechn
> > had
> > to be done with everyone out of the file like Mile S., however I see
> today
> > that Ken states this is unnecessary.  I assume when fP automatically
> > rebuilds the freechain and it never seems to finish, that automatic
> > rebuild
> > should be stopped before doing it manually.  Perhaps one of the reasons
> I
> > thought everyone needed to be out of the file was because the freechain
> > rebuild never seemed to finish, yet of everyone exits the file it
> finishes
> > in one minute or so.
> >
> > Below again is the flow of data in and out of my fP app.
> >
> > My application runs an ascii file import process every 10 minutes
> > importing
> > data from one alien application.  My app. also has a socket interface
> > connected to a second alien application importing data from that
> > application.  This client is in the process of transitioning from one
> > clinical prescription system to another, hence the two systems with two
> > different methods of bringing data into my fP app.  This transition has
> > been
> > underway for 11 months and will take another 6-12 months to complete
> (I'm
> > told).
> >
> > Approximately 40,000 records are brought into my fP application (a
> > workflow
> > application which tracks each Rx from the time into the building until
> it
> > leaves the building) from the two interfaces combined.  Each completely
> > processed and shipped Rx is then archived to a mirror file each night
> > beginning at 12:15AM.  This occurs after the network share (drive) is
> > disabled and all users are off the system.  In my archive script I
> rebuild
> > the freechain in my main Rx file before the archive procedure starts and
> > again as the last step before the script exits.  Problems occur
> (freechain
> > errors either during the archive process itself or shortly thereafter
> when
> > they begin processing prescriptions again) if I don't do this both at
> both
> > the beginning and end of the script.  During the day with 40-60 users on
> > the
> > system certain Rx's will be "voided" which copies the Rx record to the
> > mirror file and deletes it out of the active Rx file.  Depending on the
> > day
> > this may be enough to cause a freechain rebuild by itself.  The customer
> > also periodically needs to print multiple delivery sheets (packing
> slips)
> > and manually archive those orders after printing them during the day.
> If
> > they don't disable both interfaces, archiving those records will surely
> > cause a freechain rebuild.  I always assumed it was because I had two
> > interfaces assigning data to free records in the file (which were most
> > likely "used" records the previous day but have been archived and are
> now
> > free records) and at the same time was archiving data which was creating
> > "new" free records.
> >
> > Anyone who can provide some guidance on this would be appreciated by
> both
> > myself and the client.  Before this installation I had probably not
> needed
> > to run freechain on a system 5 times in my life.  At this site it may
> well
> > happen 5 times in a week.
> >
> > Don Coleman
> > Donald G. Coleman, Consultant
> > 402 Andrew Circle
> > Indiana, PA 15701
> > dcoleman at dgcreact.com
> > (724) 349-6302
> >
> >
> > _______________________________________________
> > Filepro-list mailing list
> > Filepro-list at lists.celestial.com
> > http://mailman.celestial.com/mailman/listinfo/filepro-list
> >
> >
> 
>