AnsweredAssumed Answered

Updating file on a FileMaker server

Question asked by PeterThorton on Jun 5, 2012
Latest reply on Jun 6, 2012 by PeterThorton


Updating file on a FileMaker server

Your post


I have a question regarding deployment and maintenance of a database file on a FileMaker server. I'll try to describe my situation:

I'm fairly new to FileMaker, (and indeed to the company I work for at the moment). A couple of weeks ago I started developing a solution (single file) in FileMaker Pro Advanced 11, basically learning by doing. I've been working on a file stored locally on my computer. The solution is nowhere near finished, but some integral parts of the solution are working, and the users need to start using them.

This obviously calls for two versions of the file - a live version and a development version. (I've been saying this from the beginning, but they've already managed to input a lot of live data into my development file before I managed to persuade them).

It turns out the company has a FileMaker server, hosted on a machine within the company's LAN. My boss took my file today and deployed it on the server. The file is accessible throughout the network, if Ichoose File -> Open Remote in FileMaker, and type in the address like this:


FileMaker opens the file. So that's working fine. But there's still a problem of updating the file on the server.

Let's say I update the solution about once a week. That means the users have been using the file for a week, and the data on my developer version is out of date.

Since FileMaker bundles data layer, application logic and user interface into one unseparable package, the only way of updating the file I can think of is replacing the old file with the new one. This would mean replacing live data with my week-old copy, so obviously I first need to update my development version with the data from the live version.

Is there an easy way of doing this? I can access the file that's on the server, so I can obviously choose File -> Import, and import the new data. But as far as I know, I can only do this table by table, right? So far I have some 35 tables, so I'd have to do it 35 times, every time I deploy a new version. And the number of tables is likely to grow in the near future.

Is there an easier way of doing this? Like, for example, updating all the data in all the tables in one go? Or some way of updating the user interface portion of the file, without touching the data?

At this point it seems almost easier for me to actually work with the live file that's on the server. Some of you may suggest it, but this is out of question. For me to be changing layouts and scripts while the users are actually using them, is a recipe for disaster. I also need to work with a lot of testing data - creating fake orders, moving non-existent stock from warehouse to warehouse, and so on. If the testing data gets mixed with the live data, and nobody knows which one is which, we may as well shut down the servers and go back to pen and paper. I also need some tolerance for errors. If I test a script and it accidentally rewrites all the orders with useless data, it's no biggie if it's on a development file. But if it's on live database, I'm dead.