Symantec Management Platform (Notification Server)

Hello,

I've a little question...

What is the fundamental difference between these two jobs ?

In what case use one or the other job ?

Thanks a lot.

Greg

A server job is running on the server itself. There are special server tasks you can only run in a server job (f.e. sending an email).

If you add any client tasks to a server job, the system will ask you for "input data" (the clients to run the client task) and there is some configuration when to continue.

just try one ;)

I was just about to ask the same question.  Is there somewhere that gives more details on the differences?

Client jobs are what you use 99% of the time. They are the kind of jobs you assign to a client, get pushed down to the client's agent, and run on that client.

Server jobs do not target any specific client. They run on whatever server you're connecting to the console on. Usually you use these for things that don't have anything to do with clients. For instance there are "run script on server" tasks you can use to do things like back up data on the server, etc. You just run these jobs without having to pick a specific client to run on (see below for exceptions).

You CAN add client tasks (like a regular "run script" task) to server jobs. Since you don't target the job on a client at run time, it will either prompt you when you run it to pick clients for the task (default) or you can set up a static list of clients (or a filter) when you add the task. Then when you run the job, it always runs those client tasks on the same set of clients you initially specified. This is useful for doing tasks on a known set of computers or a filter without having to re-choose them every time.

Thanks Shawn for your answer,

In my customer case, they want to script in the job, some actions on the client computer, and then on the NS. Typically, they want to delete some kind of computers that they just deploy with DS, because they doesn't need to be manage. So when the deployment is finished (OS + softs) on the target computer, at the end of the job, they want to remove the Symantec Agent and delete the resource from the NS Database automatically.

I find a solution to not use the server job, but it's 'homemade'.

These two job types are identical with one exception. Server Jobs guarantee that the exact same task sequence and execution path is followed for all nodes. For example, the logic for a job specifies that the job stops if one of the tasks fails. When that task fails or times out in one node, that job stops for all of the nodes.

 http://www.symantec.com/docs/HOWTO63326