Book a Demo

Author Topic: Version 1310 crashed when transforming huge package  (Read 3137 times)

Richard Freggi

  • EA User
  • **
  • Posts: 498
  • Karma: +18/-7
    • View Profile
Version 1310 crashed when transforming huge package
« on: April 21, 2020, 07:17:13 pm »
I have a package with > 4300 classes and their attributes and associations.

I want to convert it to Tables and generate DDL.  But my version which is normally very stable, stops conversion with an error message after an hour or so of trying to convert.  This happens regardless if the EA database is *eap or *feap.  Any ideas on how to do it?  p.s. I think if I only convert a few hundred classes at a time I would lose the associations of the classes that are NOT being converted, right?  My PC is supposed to be a workstation and I have 16GB memory, Windows 10 professional.

Any advice welcome!  Thanks!
« Last Edit: April 21, 2020, 07:19:15 pm by Richard Freggi »

Geert Bellekens

  • EA Guru
  • *****
  • Posts: 13523
  • Karma: +574/-33
  • Make EA work for YOU!
    • View Profile
    • Enterprise Architect Consultant and Value Added Reseller
Re: Version 1310 crashed when transforming huge package
« Reply #1 on: April 21, 2020, 08:58:31 pm »
Have you tried moving the model to a "real" database?

SQL Server Express or Developer Edition are free and often work more stable than .eap or .feap files.

And of course make sure you run the project integrity check before trying anything like this.

Geert

Uffe

  • EA Practitioner
  • ***
  • Posts: 1859
  • Karma: +133/-14
  • Flutes: 1; Clarinets: 1; Saxes: 5 and counting
    • View Profile
Re: Version 1310 crashed when transforming huge package
« Reply #2 on: April 22, 2020, 11:08:25 pm »
I think if I only convert a few hundred classes at a time I would lose the associations of the classes that are NOT being converted, right? 
I think that might actually work. Transforms have a bit of black magic in them, and it might be that as long as you target the same package, the second transform will actually merge its results into those of the first. Definitely worth a try.

/U
My theories are always correct, just apply them to the right reality.