Re: Why a fork?

I used the Daffodil Replicator long before they released their enterprise version. I'd gotten fairly involved with it, even paying for a gold support contract with Daffodil, in hopes that they would fix the issues I was having. Instead, they released an 'enterprise' version of their replicator, and no longer seemed very interested in fixing/improving their open-source project. The forum posts were seldom truly resolved, and seemed to have dropped off, and the mailing list was filled with 95% junk mail. Contact regarding bug-fixes through our support contract was responded to with solicitations for the 'enterprise' version. A patch to fix four or five bugs was submitted through the sourceforge project page, but received no response at all. Consequently, I've fixed the bugs I've come across, which were numerous, and made speed enhancements and packaged it for debian-derived systems for our own use.

Some of the enhancements include:

• Speed - this version runs orders of magnitude faster
• linux server mode - for headless environments
• bug fixes - too many to count.
• packaging - deb package and init/wrapper scripts for *nix systems

I've down-graded the Development Status from stable back to beta, because I haven't tested it on anything other than postgres, a little mysql and a little sqlserver since making changes to it. In its present form on the daffodil project site, however, it is my belief that it is misrepresented as Production/Stable, as their version doesn't seem to work for postgres databases at all, has problems in any database with multi-field primary keys, etc. I believe that this is extremely valuable technology, and there didn't seem at the time to be another open-source product that did what this one did. I've fixed it, improved it and re-released it.

I guess the gist is, after finding all the problems that I did with Daffodil's open-source version, I had trouble trusting their enterprise version. Thus, I fixed the open-source version for myself, and then thought other people might be able to use it as well.

Brian

Thanks for your answer. I did notice the lack of forward progress on their open-source version. I'm glad you picked this up. There is definitely a need for heterogeneous database replication. I need one myself for Oracle-PostgreSQL.

I downloaded your version and tried it on Windows. I had an error right at the beginning -- sorry I did not save it -- so at that time I was just evaluating various solutions quickly to eliminate the worst. Though in the past a colleague made Daffodil Replicator OS work between two Oracle instances so I want to try your version again.

The error I was getting was when I was creating the publication.

I'll give it another try and let you know. If I can get it to work, I'll definitely contribute any enhancements I may have to do.