multi instance IS 6.1.5 using same TN

, , ,
1

Hi wmusers,

has anyone faced the problem to use the same Trading Networks from more IS 6.1.5.
I have searched on advantage for any Docs related to that, but could not find anything!

Maybe someone of you has experience with that and could help me to find some doumentations.

kr
Werner

2

The TN docs and help touch on this a little. You can connect multiple IS instances to a single TN database. Each IS must be configured with the same values. Are you having an issue? Or just looking for how to set it up? Keep in mind that 6.1.5 is predominantly a Modeler/PRT version and that there is no TN 6.1.5 per se, nor does 6.1.5 change the TN DB schema AFAIK.

3

hi reamon,

thanks for your fast reply! I am only investigating on it, because we will have to run this setup within the next 2 month. Are there any other docs on that, except the “standard” TN docs and help?
kr

4

Not that I’m aware of. It’s not a hard thing to do. Just configure one IS for TN. Then do the next one exactly the same way. There are some properties that control whether or not the 2+ IS/TN instances talk to each other–“I just changed the DB, reload your cache” type of stuff. These properties are described in the TN admin help pages.

5

but isn’t it a problem if you have routing rules which executes a service on one of the IS? How could you specify on which Instance the service should be executed(5555,5556,…)?

6

Hi Schlahu,
Please go through the IS and TN, clustering guide.

In the TN Processing rule, you don’t specify the IS instance. You only give the service name to execute.

By having a DB based Repository, setting the TN properties and enabling the clustering, you can achieve the true clustering using the 2 or more IS and 1 TN DB.

In that case if you change any Rule or Profile or anything in TN DB using 1 IS, all the IS in the cluster gets refreshed automotically with the latest values from TN DB.

Also, when a document satiafies a rule, the corresponding service may execute on any 1 IS in the cluster based on load (if cluster is based on least pending request) or alternatively (if the clustering is based on Round-Robin).

7

You don’t. Once a connection is routed to a specific IS, it stays there to completion. How a connection is routed to which IS depends on how you have things set up. If clients connect directly to one of your instances, that’s the only instance they will ever see and if it is down, they cannot connect.

I presume you would have an external load balancer to route traffic to one of your IS instances.