Datastage Health check

[clarcombe]

Hello
I am looking to create a job that checks the health of Datastage before running a job. IBM support say I need to do this manually through the Admin console but I can't check that manually before each run.

Has anyone come up with any good ways to check DS and the server before running a job ?

Thanks

Colin

[PaulVL]

Well it really depends upon your level of comfort when saying "it is up".

I will assume a Linux back end host for this example.

1) You could look for dsrpcd pid.
- if that is not running you are dead in the water.
2) You could look for the java pids associated with websphere.
- webphere not running is also very bad for you.
3) You could run $DSHOME/bin/dsjob -lprojects
- this tests the engine, but not your authentication
4) You could run an istools.sh command to do a dummy extract (-preview) of the first job contained within a project and validate the return code on that.
- this will test each item you care about as long as you pass in credentials (best done via encrypted authentication file).



You do not want to do that at the beginning of each job submission.
Maybe once a day if your ecosystem is unstable.

An automated scheduled monitoring process is advised. Someone needs to be alerted if the environment goes down (does not happen often). As with many alerting mechanisms, having an external host reach into your DataStage box is a good strategy because if your DS box is down... so is your alert.

[ray.wurlod]

Operations Console dashboard will give you a quick overview, with Activities tab giving more detail. You can make use of the REST API (or queries against DSODB) too.