I posted this last week but I never say it on line so either it was
blocked or ??

We use triggers to keep our co-location in sync on a hot basis.  We have
a somewhat more Complex situation in that ALL of our jbase files are
kept in sync this way.  To manage this, And in order to preclude
recursion, our common trigger creates a key describing the file, record
Key, and operation (write, clear, delete) and writes the record with
that key into a common file at The incurring site.  A separate process
is WAKED and writes the available records to the corresponding File at
the other site.  There a process which was started at port 4999 writes
the transported record(s) to the target File.  The trigger code is
written such that it checks the process port and if > 4500, ignores the
trigger trip and exits.

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf
Of LWhite
Sent: Monday, August 02, 2010 10:50 AM
To: jBASE
Subject: Re: Trying to avoid trigger recursion in a jBASE system

If a second post shows up I apologize. I had a mishap with the tab
key. :)

OK, Thanks for the replies. I am going to be looking up a lot of what
you mentioned.

Maybe if I try to explain the situation again you might see something
that I didn't properly express the first time.

I have a system with a file named IM (item master)

There are both basic and PROC programs accessing it.

Management has decided that we need to be able to have one company
write code and distribute it to all of the other entities.

We have done this before by using an include file that each entity
adjusted to fit their individual file names and order of attributes.
Example being that we all store an English part name in the IM but not
in the same attribute.

Operations has decided that we will have all companies move to the
same file structure. So every company will now have a MPF
(manufacturing product file) and that it will have the part name in
attribute 2. For some reason they will not continue using the include
file. I cannot explain why. I have not had it explained to me.

Global code is being worked on. We must use it and we must have the
file structure set up by a specific date.

However, They do not want us to touch every program that currently
accesses the files that have a global equivalent. Since they will be
moving to one process they want time to evaluate each portion. We
still have to deploy part of it now.

Our idea was to create triggers on the files to keep the local and
global files in sync. Triggers are not required by anything. Any idea
that someone believes is better will be looked at.

I am just a code monkey. I am doing what I have been told to do and we
are not given an option on the implementation of the code or the
structure.

I am going to go look up transaction journaling and activeMQ.
Thanks.

-- 
Please read the posting guidelines at:
http://groups.google.com/group/jBASE/web/Posting%20Guidelines

IMPORTANT: Type T24: at the start of the subject line for questions
specific to Globus/T24

To post, send email to [email protected]
To unsubscribe, send email to [email protected]
For more options, visit this group at
http://groups.google.com/group/jBASE?hl=en

-- 
Please read the posting guidelines at: 
http://groups.google.com/group/jBASE/web/Posting%20Guidelines

IMPORTANT: Type T24: at the start of the subject line for questions specific to 
Globus/T24

To post, send email to [email protected]
To unsubscribe, send email to [email protected]
For more options, visit this group at http://groups.google.com/group/jBASE?hl=en

Reply via email to