Jump to content

digitaljunkie

Members
  • Posts

    98
  • Joined

  • Last visited

Profile Information

  • Slogan
    I want to learn...

digitaljunkie's Achievements

Collaborator

Collaborator (7/14)

  • First Post
  • Collaborator
  • Conversation Starter
  • Week One Done
  • One Month Later

Recent Badges

0

Reputation

  1. if your new to join tables a good start woud be here... As your new to join tables a good start would be here... It should give you a better understanding of there purpose. Watch the (conveniently) free section on relationships. Click this link and whatch the tutorial section on RELATIONSHIPS Many to many relationships part 1 and 2.
  2. No because the join table simply create the relationship and contains anything you want to be unique for that particular relationship. If the book is already in the book table then simply create a relationship through the join table from the media contact Jane to the book. If you want to send her that book you can at that point deal with that in the script at point of creating the relationship. But you said a completely new book indicating it is new to the system, although either way its still the same, after creating the join record the script says do you want to send this book to Jane now? yes or no If yes then do this.... in no then exit script, leaving the relationship built and ready to use later on.
  3. OK this is where scripting comes into play for you. You would have a new book script this then goes off to the books table creates a new record and lets you fill in the details. You click continue and the script then asks do you want to assign this book to media contact Jane. You say yes and a record is created in the join table.
  4. I think what you want to do is have all the books in one table and the media contacts in another table, then you can relate the right books to the right people based upon a common denominator or 'interest' that way you create a related portal on the contacts layout which will only show the books related to each contact. You will need to factor in a few things though, a join table in order to relate the book and contact in many instances as I don't know if your notes are contact specific or book specific or even both. Creating a join table will enable both unique instances to have notes etc. I may be wrong here but it does sound like your maybe over complicating the process slightly.... [Media Contacts] > [Join Table] < [books] John > Unique Record < Jungle Book Mary > Unique Record < Dummies Guide Jane > Unique Record < Jungle Book Jane > Unique Record < Haynes Manual As you see jane now has unique (join) table record for 2 different books. If Janes ID has a matching foreign key ID in the Unique record then the foreign key ID for the Book is present so your join table tells you this... A record here is related to Jane (Foriegn Key) and at the same time is related to Jungle Book (Foreign Key) But Also at the same Time John (FK) is related to Jungle Book (FK) This then lets you represent the data in both tables or not depending on what you want to do. It is what is commonly known as a many to many relationship, where many relationships occur in one or more places and relate the same data or records across many places or to many opposing records that they are related to. Personally as you are taking the project on this is how I would approach it so that you future proof this example for later. Now like I say, I may be completely off point here but it sounds like your saying the current state of the database is unrelated like a spreadsheet so you need to split the data out and make it related in order to restore some form of sensibility and not have to show 167 fields per record just becuase the data is unrelated and doesn't know what where and when.
  5. Hi Steven, Could you elaborate on the R/W issue? I.E is it purely a performance issue or a data integrity issue? Also does this mean the Databases on a RAID volume or FMS in general? I was under the impression it was a performance issue when drives were mirrored because the software had to process the duplication process over two or more volumes. I did check with FM and SoftRAID regarding this as Hardware RAID cards are very expensive, I know that Apples own RAID software isn't supposed to up to much but SoftRAID comes in highly recommended around the way.
  6. If it helps I have a FileMaker Server deployed running on a G5 OS X Server (Running SoftRAID) as can't justify an Xserve, that servers out Mail, AFS and VPN Services, I do use ACL's to limit access down to only users and strictly no guest access anywhere, Also I have one folder shared out as a SharePoint via ACL's not the root drive. Its been fine for the last 2/3 years, only issues I can see if you share out the databases or access them directly. Or if something else falls over on the server that will take down your FMS etc etc. TBH yes it is bad practice but cost is a weighing factor, but then a Mac Mini (Intel) is fine for FMS I use these on clients sites for just FMS. Also If your planning HTTP access to the server for FTP/SFTP/HTTP/HTTPS etc etc then steer well away from having FMS on the same server. These services are the most common form of services utilized by hackers and viruses etc. Having your FM Database on a server that is open to the big wide world for other purposes other than to access itself is really asking for trouble, especially as that access would normally be from unknown sources and users. It comes down to best practice you either can afford financially to follow it or not, if its your own choice then personally I would recommend separate servers but that isn't always possible. Especially if your having to justify the cost to someone else or even your own bank roll. How important is your data? you can never be to safe or security paranoid. But on the flip side servers were designed to serve. OS X Server itself ships with the ability to serve/host a multitude of services that people will argue are best left to dedicated servers but theres nothing to stop you from doing so from one machine as long as you fully understand both sides of the argument and the possible problems encountered by doing so. The reason these services ship together is purely for convenience not because you can. I wouldn't run a Web Server on the same server as my Mail or File Server but then thats because web serving gets me paranoid. But if it was internal and I could control its access to only the people I wanted then why not? I guess what I am saying is for a moment ignore the papers and manuals and take considerable time to study and fully understand you EXACT requirements and how they are likely to change/increase in the future. Then understand what you can do. Then understand what you should do. Then understand what you shouldn't do. From that you can make a better decission as to what you are going to do and how. But always understand the alternatives to you and whether they are better and achievable. If so ask yourself why you aren't doing them. Remember people say things for a reason and its all experience in my mind, but thats not always going to help you if you CAN'T do it. Hope my little blah blah blab helps/ hehe
  7. Hi I want to find record occurances of (instances) of a particular criteria e.g... I have 500 contacts and want to find instances that include bob in the name field. But this is what I want to do... Search for Bob goto 1st instance but also be able to browse the previous or next records surrounding this instance that don't necessarily have bob in them but then I want to click to the next or previous instance of bob and again browse the surrounding records. Any ideas?
  8. Well kind of, after lengthy research on my behalf it seems that even though FMS performs a consistency check this is definately not to gospel or acurate as with any form of corruption in the extreme or most severest of instances may not be noticed until much further down the line, sometimes months of development later. Corruption it would appear can take two basic forms 1) Data the data becomes corrupt. 2) Structural Object Based Corruption, where the actual data wrapper or platform structure becomes corrupt, as filemaker is file (object) based in that everything usually is contained in one file (Data, Structure, Scripts, Model, Definitions etc etc) there is a lot more room for unknown corruption. Even myself have experienced corruption from a basic VPN disconnect whilst in scripts. It is in this instance that best practice dictates you should roll back to your last available backup or take the database(s) offline and check them locally in FMP itself. As I am relatively new to FMS I was unware that it consistency checking when automated with no human verification will always lead to possible omitted errors, as far a I see it no device is intelligent enough to check consistency of itself unless it has a 100% pure copy to check itself against. Thinking on reflection this stands to reason anyhow that FMS can only check consistency against what it knows to be a consistent structure many things may 'seem' consistent but actually not be, for example missing relationships, corrupt tables or fields etc etc. So in future if a remote server restarts without my intervention I want to or at least have someone check over the hosted FileMaker files on that machine before allowing users back in. Hopefully I have summed up what others have spoken about in more detail in a readable and correct way?
  9. Hi Steven, Thanks for that. Also is there CL command for permanently stopping the daemon from auto-starting on a restart.
  10. ok found it you need to log into FMS Admin from the local server itself, the tool is under the Server>Local FileMaker Server Administration Menu item (In case anyone else needs to know) I am pretty sure its a wrapper for a command line fix, if anyone knows how to do this via SSH would be good as would be ten times quicker than cracking open a VPN + RDC session to each remote server and doing it.
  11. At work we are going to be receiving XML files from a supplier via FTP (Push to our server) I need to think about best practice here... I need this received XML file to be accessible to remote databases in order to pull into filemaker. I was thinking about applescripting a remote mac to pull down the xml then instruct FileMaker to import it but I'm still a little unsure of the exact/best process to do this.
  12. I recently read about how its bad practice to have FMS autostart on restart or login. Two questions to follow this up if I may... 1) How do I stop a FileMaker Server from doing so if I set it up to do so initially, as I can't find this in the FMS preferences or settings. 2) Where can I read more about this or the post? or could someone explain why its bad practice to do so. Thanks
  13. never seen this in OS 9 itself, in fact from memory there is no safe mode in OS 9 so this must either be FMS related or possibly something else log at what extensions are installed and running have you upgraded or installed anything recently?
  14. ACL (Access Control List) it overwrites poisix permission when sharing files or sharepoints in OS X Server. Basically I can add a group of people to an ACL and say "Let this group read/write" to this sharepoint. I pull down backups of a file in order to clone it off for a new client or solution. Unfortunately my boss at work likes to take a copy of the file offsite with him to do some ahem "Developing" (lol) then he will say use this file from now on becuase my changes are soo much more significant to out solution (slight sarcasm). But the backups are important a couple of developers regularly need to pull down a backup off the server to their local machines etc etc.
  15. what would you peeps say is best practice, I need to push and pull copies from os x server, would sharing the folder using ACL's affect the working of those files under FMS? Or what is a better way?
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.