|
View previous topic :: View next topic
|
| Author |
Message |
nkothako
Active User
Joined: 26 Nov 2008
Posts: 125
Location: chennai
|
|
|
|
when a file(called as notify file in our system) reaches record count of 100 can i trigger a job.
note : This file will be updated by our front end dataset trigger jobs when every a new file is cataloged. And it contains the newly received file details ex : file name, time, date, amount of file and etc.. |
|
| Back to top |
|
 |
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
Periodically (hourly, daily, whatever is appropriate) schedule a job to count the records in the file and set a particular condition code when the record count is 100 or greater. When the condition code is set, run the part of the job that is to be run at that time.
If you provide a better explanation of the process, someone may have a more useful suggestion.
Keep in mind that you need to get "all" of the accumulated records and empty the file for the next 100 entries. |
|
| Back to top |
|
|
nkothako
Active User
Joined: 26 Nov 2008
Posts: 125
Location: chennai
|
|
|
|
Dick,
My notify file is keepon updated by other front end jobs. When ever my notify file reaches record count 100 there is a specific job, which is other than front end job needs to be trigger.
note : I don't want to check the count of file through an other job periodically. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
How would anything know the count had gotten to 100 if something does not check periodically?
Is this already running in production or is this still an experiment? |
|
| Back to top |
|
|
nkothako
Active User
Joined: 26 Nov 2008
Posts: 125
Location: chennai
|
|
|
|
Apart from jcl, is there any way verifying the specific dataset when its updated
note : Through ispf or rexx or mvs or assembler.
This is purely production, not for an experiment.. this concept will save a huge no of unwanted runs of certain jobs in our exisiting environment. |
|
| Back to top |
|
|
dick scherrer
Moderator Emeritus
Joined: 23 Nov 2006
Posts: 19244
Location: Inside the Matrix
|
|
|
|
Hello,
| Quote: |
| Apart from jcl, is there any way verifying the specific dataset when its updated |
This is not something you can do "with jcl". Why do you care when the file was updated? You mentioned that you needed to know when the file had 100 records, not when it was updated.
| Quote: |
| This is purely production, not for an experiment.. |
There may be production jobs now running, but this process of identifying when there are 100 records in a file is trying to become an experiment 
You still have not posted what it is you are trying to accomplish. What you've posted is something you want to use as a solution, but you have not explained what the requirement really is. If you explain how you intend to save this huge number of unwanted runs, someone may be able to help. |
|
| Back to top |
|
|
Arun Raj
Moderator
Joined: 17 Oct 2006
Posts: 2481
Location: @my desk
|
|
|
|
| Quote: |
| this concept will save a huge no of unwanted runs of certain jobs in our exisiting environment |
nkothako,
Which concept?Can you explain this?
How do you make sure that the "updates" made to the dataset are permanent while the job is running? What if the job abends after writing 200 records and uncatalogs the file , but you have triggered some job with this "new concept"?. |
|
| Back to top |
|
|
|
|
