Thanks for the explanation, Dick. This is the direction I was going....guess
I was mostly looking for a clue as to how complicated it was going to get
before I invested hours and hours. Your explanation seems pretty
straightforward.

You've helped much!

Thanks!
D

************************
Diana Nichols
Webmistress
http://www.lavenderthreads.com
770.434.7374


-----Original Message-----
From: Dick Applebaum [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, November 14, 2001 9:46 AM
To: CF-Talk
Subject: Re: SQL -> CFHTTP -> SQL


Diana

Will the remote client allow you to place a program on their server?

If do you can:

   initiate a scheduled task every 5 minutes on your server

   query your local database for changes (last 5 min)

   build the SQL commands & data to update the remote DB

   optionally encrypt the data

   Serialize the above to a WDDX packet

   send (post) to the program (a stub program) on the remote site via cfhttp
   using SSL

The stub program

   receives a posted WDDX packet

   deserializes it to the original format (SQL commands & data)

   optionally decrypts it

   executes the above against the remote (local to the stub) db.

   serializes any SQL results recordset as a WDDX packet

   returns this to the caller (outputs the above WDDX packet)

With the same (or a separate) cfhttp request, you can query the
remote db and return its changes to the local server.

The net result of all this is that you can manipulate (update.
Select, etc.) the remote db as if it is local to you.

The stub program can be written in CF or any language that supports
WDDX (ASP, Perl...)

If the remote site wants to secure/limit the stub program, you can
furnish it to them, they can add whatever security or limits, and
install it themselves... so you need not have unnecessary access to
their db.

This is more complicated to explain than it is to implement.

I use a similar scheme all the time to manipulate clients' databases
on sites remote to me.

HTH

Dick


At 8:57 AM -0500 11/14/01, Diana Nichols wrote:
>I have a more complicated solution (VB progamming) - but am wondering if
>this CF solution will work.
>
>The scenario - I have a SQL table that needs to be replicated (changed data
>only) to
>a remote client's db every 5 minutes. (The data is credit applications.) At
>the same time, I need to check to see if they have new data for me
>(approvals, etc) and GET that.
>
>The simple thing would be SQL replication using a VPN, but they're
insisting
>on using SSL for the transmission security.
>
>I don't know how to make SQL use SSL, so here's my thought:
>
>We load a CF page into our browser on our server. They open a session
>connection with this page on theirs, also using a browser.
>
>We refresh the page every 5 minutes - many ways to program that (CF, JS,
>etc).
>
>Functions of the page:
>
>1) run a query for new data
>2) use CFHTTP to POST any new data to a form or CGI script on their end
>3) use CFHTTP to GET any new data on their end and return to our DB
>
>If we run the browsers continuously, we can maintain an open
session.....and
>we only have to re-establish the connection when one of the servers is
>re-started.
>
>Does this make sense, is it feasible, and am I overlooking something
>critical?
>
>TIA!!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Get the mailserver that powers this list at http://www.coolfusion.com
FAQ: http://www.thenetprofits.co.uk/coldfusion/faq
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists

Reply via email to