Translating with DrWatson… this can take a few seconds the first time.
This is a genuine, complex translation. DrWatson protects commands, error codes, and log output while naturally translating the surrounding text. It’s translated once and saved.
Jonathan Leffler — — source: Informix-list mailing list archive (1991-1998)
>From: uunet!magellan.navsses.navy.mil!macia (Harland Macia)
>Subject: Informix-Star help/opinion needed
>Date: 23 Mar 93 15:01:08 -0500
>X-Informix-List-Id: <news.2886>
>Has anyone done anything like the following or any ideas?
>1. I have 5 different workstations running Online connected by I-Star.
>2. The databases on all 5 computers must always be the same.
>3. If one computer is (known to be) down the transactions should continue
> to the other 4 computers. I do not want I-Star rolling back the
> transaction everytime because the 2 phase commit thing doesn't work.
>4. If computer #5 was down and then brought back up at a later time then
> its database should be synchronized with the other 4.
>I will try to clarify any questions.
Item 3 means that item 2 cannot be maintained. One of those two conditions
is going to have to be relaxed. Item 4 is the hard part. The worst case
is not when a machine crashes; it is when a network connection breaks and
two sets of machines continue processing independently of the other set.
This is a very hard (post-doctoral research) type problem. Given enough
special conditions, something of the sort can be achieved, because I've
done it, but the special conditions were very stringent.
Yours,
Jonathan Leffler (johnl@obelix.informix.com) #include <disclaimer.h>
We use strictly necessary cookies to make this site work. With your
consent we’d also use optional cookies for analytics and marketing. You can accept all,
reject all, or choose. Read our Cookie Policy.