This error message "Cannot complete this action" mean you CAML query statement is wrong somewhere!! Just only check your query :thumbsup:

wow that got complicated submit button > client script- if valid postback So I would just add the onClientClick="return validate_Form();" And write a simple function first in javascript/jquery test if the fields are empty test the field data regex the email address, if validation completes, then return true to fire the postback Then you can get fancy, and add an ajax call to check for duplicate email address, I would use onblurr for that, so it only fires once, and if a duplicate exist, focus the email field again.

Yes that is correct. I'm not familiar with the SQL Server flavour of doing things, but when defining a trigger you can specify a list of columns as opposed to the entire table, that the trigger is fired on. Glad that has helped you out. :) Chris Meech I am Canadian. [heard in a local bar] In theory there is no difference between theory and practice. In practice there is. [Yogi Berra]

OK, thanks a lot. Sounds like a good idea, will give it a go tomorrow. Thanks again for your help :)

Liqz wrote: Another lesson learned Same here, I learned it the hard way ... ;) MVP 2010 - are they mad?

OK Gideon that's great, thanks for your time and advice :)

Just in case anyone is interested, I solved the problem by replacing the SQL Express Server with the full version. All seems to be working correctly now.

Abhishek Sur wrote: But If I was in your situation, I would have invoked an AJAX call to the server, and get if it is available or not. Ah, didn't realise that I could do this (I am a newbie to ASP.NET :D ) Thanks again for your help.

Got it, thanks for the tip, all working now :)

Christian Graus wrote: I would not use them for anything. OK thanks, I appreciate the advice. The only other way I can think of to make the connection string available to my class library is to pass it into the constructor when instantiating objects. Is this the method you would use or do you have any other suggestions?

Thanks.....any luck ?

Great, just what I was looking for - thanks :D

Ah I see, understood :) Thanks for taking the time to explain it so well.

While this is what I call a "technical" issue I have not seen it become a problem in a practical sense (I also currently don't work on transactional systems either). I find that master file information is rarely updated once the system is set up and if I enforce a daily restart it pretty much eliminates the problem. I also tells the users it is a potential issues so they are aware of it. Never underestimate the power of human stupidity RAH

Abhijit Jana wrote: Then you should marked that post as Good Answer And you get 5 from me for understanding him courtesy. Manas Bhardwaj Please remember to rate helpful or unhelpful answers, it lets us and people reading the forums know if our answers are any good.

Ok great, thanks for the help :)

You should store it in Session variable. Regards, _Anil jhgdgf

Liqz wrote: I thought of having, in the database table, an owner (ID?) field and an owner type field (this seems the best way so far). You thought correctly, it enables you to lookup the type based on the OwnerID without storing duplicates :) It is dangerous to instantiate something that you do not know - since it might, for example, delete all records in a table on startup, as initialization. Even if you're sure that nothing can go wrong, then it would still be slow, as you already stated. Liqz wrote: what is the best way to determine the owner? i.e. what type of object. If you can't ask the owner-object itself, then you'll have to resort to keeping track of that information elsewhere. Like tracking it in a database :thumbsup: I are troll :)

Nevermind people, got it sorted thanks :)

Normally, I'd reply that any working solution isn't bad, but this is a dirty hack and not a solution. The thing is that you notice that 5 seconds isn't always enough, so you extend the period to 10 seconds. That's kinda cool when the task usually finishes within half a second, leaving you to wait for 9,5 seconds every dang time. How often do you come across a place where the application is waiting, no disk-activity and no cpu-activity? My train of thought departed from the application. What do you do when you can't change the code to send a message to signal an event? It didn't even cross my mind that one could check the object that the application is working on, I just kept thinking on the same track - where the problem arises. In short, thanks for showing a better option than the quick-and-dirty 'solution' :) I are troll :)