Hi Ralph, The database (as are the 6 other database I run) was set up using the "Absolute Incrementation Using Two Recursive Links" method, as its run on a network and as far as I can see, still conforms to your instructions summarised in box on page 140 of your book
P1F2 (actually my P3F1) is edit order field 76 and P1F2 (reverse field - actually my P3F83) is edit order field 83 P1F3 & P1F4 (Actually P3F84 & P3F85) are linked as per instructions and both Indexed with my Index 9 which only contains the reverse field my P3F83 P3F1 formula set to update when created record is saved and P3F83 is set to update on any change I mentioned new records are added daily but should also have mentioned that child panel records are often edited on a daily basis also, and this is the 1st time the problem has occurred. As I say the re-indexing which I probably hadn't done for a month or so appears to have instantly fixed the problem and it hasn't as yet re-occurred again. Any other thoughts - not a major if the index recovery fixes it should it ever re-occur again. BTW - referring to your late breaking news on page 140 - did Lew actually change the structure so the two link method was no longer neccessary on a network - just curious Regards Colin ----- Original Message ----- From: "Colin Roberts" <[EMAIL PROTECTED]> To: "Dataperfect Users Discussion Group" <[email protected]> Sent: Monday, 21 July 2008 21:36 Subject: Re: [Dataperf] Re: Recursive link and index > Hi Ralph, > > Changed from auto-incremental fields to your recursive link methods many > years ago (more than I care to remember :-) and when set up your rules were > followed. There is a chance that maybe the edit order may have been changed, > and I'll need to get out your book and check tomorrow. Haven't got access to > your book at present, but from memory I think I remember the edit order > between the two recursive link fields was important and I'm pretty sure the > edit order relationship between those two fields hasn't been altered even if > the edit order of the panel may have been modified as new fields have been > added over time. > Database is extensively used and new records added daily - this is the 1st > time I've ever had such a problem and re-indexing appears to have fixed it > (at this stage?). > > I'll recheck your rules still apply tomorrow, and get back with my findings > > Thanks for responding so quickly > > Regards > > Colin > > > ----- Original Message ----- > From: "Ralph Alvy" <[EMAIL PROTECTED]> > To: <[email protected]> > Sent: Monday, 21 July 2008 17:51 > Subject: [Dataperf] Re: Recursive link and index > > > > Colin Roberts wrote: > > > Long estblished Database - 35000 parent records and 85000 child records. > > > > > > Edited a child record (one of 20 child records of that particular parent > > > record) via the parent record and found that when saved, the child > record > > > picked up a new recursive number on saving. Was able to remove non > updatable > > > field designation and put indexed recursive number back to what it was > and > > > all looked OK. Editing of some other child records sometimes did the > same > > > thing (created a new number) and sometimes it saved OK retaining the > > > originally allocated number. There was never any error message or bomb > out. > > > Re-indexing database seems to have got rid of the issue, which I've not > ever > > > experience B4 - and not sure exactly why or how a saved record can > generate > > > a new number when edited - no other user was accessing database at same > > > time > > > > Did you make sure the Edit Order followed the rules outlined in my book? > > > > Did you make sure the Update Conditions are following those rules too? > > > > Ralph > > > > _______________________________________________ > > Dataperf mailing list > > [email protected] > > http://lists.dataperfect.nl/mailman/listinfo/dataperf > > _______________________________________________ > Dataperf mailing list > [email protected] > http://lists.dataperfect.nl/mailman/listinfo/dataperf _______________________________________________ Dataperf mailing list [email protected] http://lists.dataperfect.nl/mailman/listinfo/dataperf
