Member Avatar for k9huey

Briefcase application
Access 2003 two way replication

Are there any open source databases that allow two way replication exactly the way Access 2003 handles it?

I have a desktop app that I would also like to have live on the web but I have to use Open Source software like MySQL. The more I look into all of this the more I am forced to continue using Access because no other database can handle replication the way that Access 2003 does.

IT will not give me a windows based server so I can not use Access or MSSQL.

I need the replication because it is a briefcase application meaning that I have service guys that take the info on the road with them and it is hard to connect to the office server at 30,000 feet or in the middle of a customers work floor.

Does anyone know of a way to make this work????
Ken

  • 2 Contributors
  • 2 Replies
  • 213 Views
  • 6 Days Discussion Span
  • Latest Post Latest Post by k9huey

Dani AI

Generated

As described, the requirement is exactly the old Access “briefcase” workflow: offline, two‑way merges, conflict tracking and later reconciliation. Access/Jet provided that out of the box (Briefcase + Replication Manager and built‑in conflict tables), which is why it feels hard to replace. (flylib.com)

A quick, maintainable way forward is to pick one of two proven open‑source paths rather than trying to force MySQL native replication or lots of ad‑hoc dumps.

  • SymmetricDS — a Java sync engine that adds change tracking and push/pull multi‑master sync to relational DBs (MySQL included). It is built for many remote nodes, intermittent connections, and conflict handling; it fits a 10+ laptop topology. (symmetricds.org)

  • CouchDB + PouchDB — an “offline‑first” document database approach. PouchDB runs on the client (browser/mobile) and syncs bi‑directionally with CouchDB, with built‑in replication protocol and conflict detection. Good fit if you can move to a document model. (couchdb.apache.org)

Practical checklist and minimal design rules you can apply now:

  1. Choose model (relational → SymmetricDS; document → CouchDB/PouchDB).
  2. Use global IDs (UUIDs), tombstones for deletes, and a last_modified/version field for merge heuristics.
  3. Define conflict policy up front (server wins, LWW, or custom merge).
  4. Test with simulated disconnected edits and recovery.

Example relational fragment:

CREATE TABLE items (
  id CHAR(36) PRIMARY KEY,
  payload TEXT,
  last_modified TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
  replica_version BIGINT DEFAULT 0
);

Do not treat MySQL circular master‑master as a client‑sync solution: native MySQL replication expects networked servers and is not designed to merge frequent disconnected client edits; scripts like mysqldump can work as a short hack but will be brittle at scale. For a reliable, supportable system, evaluate SymmetricDS (if you must keep MySQL) or CouchDB+PouchDB (if you can move to documents). (dev.mysql.com)

Member Avatar for phpstan

I am not familiar with Access 2003, so I do not know how it handles replication.
If you have something in the database records that let you know the newest records, then you should be able to use mysqldump with the "where clause" to dump the updated records from one of the databases so they can be added to the other database.

Member Avatar for k9huey

Thank you phpstan for the replay.
I currently just let access handle all of that and was hoping for some type of automated system to replace it with. I will look more into the mysql dump.

I have seen sites talking about using XML to handle it but they do not go into any type of explanation as to how to do it. MySQL does have a Master to Master replication process but it seems limited to 9 masters and I need at minimum 15. It also seems that this type of scenario I am trying to setup is NOT what they had in mind for a proper use of it.
I have been looking for a while now for some type of system or written process to handle this exact thing for me but have had no luck. This seems like a common thing to me but I guess I am wrong.

Does no one use offline access to databases?
You can NOT be connected to the internet 100% of the time.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.