PDA

View Full Version : Scheduling jobs



kettle_anonymous
04-21-2006, 06:37 AM
Hi,


I'm trying to use KETTLE with a MySQL DB.
The only missing point on Kettle (for me) is a scheduler...



I 'm tying 2 ways to schedule my Kettle jobs :
- Using the EVENT function of MYSQL, but i don't know how to "connect" it with Kettle
- Looking on the Web, i heard about using the Windows Scheduler, but i don't know how to use it with Kettle too.



If anyone/matt can help me, it will be nice.



thkx



Rom

MattCasters
04-21-2006, 06:47 AM
Hi Rom,

Did you try looking in the Kitchen or Pan manual?

Thanks,

Matt

kettle_anonymous
04-23-2006, 11:47 PM
Thank you for this very fast answer, I'll try it now.

Rom

kettle_anonymous
04-24-2006, 01:08 AM
Thanks a lot, scheduling works using pan or kitchen.

Rom

sam
01-08-2010, 03:13 PM
In the Start node of Job, there is a way to specify scheduling.
I could not locate a documentation for how this works.
Could please explain how it works, if any supporting software is needed, what happens if the previous job-run is still running when it is time to run the next schedule, etc? (or point to some documentation)?

marki
01-11-2010, 01:36 AM
Hi

I recently needed to do something similar myself and found that you have to use Kitchen to perform the scheduling.

I did find a piece in the Kitchen user guide but basically if you are using a windows environment you need to create a windows batch file which contains the following command:

kitchen.bat /rep:"The Name of your Repository" /job:"The Name of the Job you wish to Start" /dir:/location /user:XXX /pass:YYY /level:Basic

Simply substitute the name of your repositoty within double quotes after the /rep: parameter, the job you want to initiate in double quotes after /job:, its location in the repository /dir:, the username and password to access the repository after /user: and /pass:, and finally the type of logging you want after the /level:.

You then use the windows task scheduler to schedule the windows batch file you have created.

It works pretty well and I have had a job running on a daily basis for a month or so now via this method.

If you do not want to access the job via the repository you can access a file and this is covered in the Kitchen User Guide as is doing the same thing in a unix environment.

Hope this helps

Mark