Skip to main content
Announcements
See what Drew Clarke has to say about the Qlik Talend Cloud launch! READ THE BLOG
cancel
Showing results for 
Search instead for 
Did you mean: 
borrisbeck
Contributor
Contributor

Auto resubmit of talend orchestration job based on error message

Hi

I have more than 20 jobs in a talend orchestration job, often days the job fails due to connection issues. I'm trying to automate the re submission process. The orchestration job already has tcheckpoint enabled between the jobs and all the jobs are linked through "onsubjobok" trigger. To automate the re submission process I'm thinking of using a "tlogcatcher"  and parse the error and resubmit the orchestration based on the error message in the same orchestration job itself. but I'm not sure about howto call the same orchestration job within that job itself? even if i use another job to do this, will it restart from the point of failure on auto re submission as the orchestration runs from TAC ?

 

I have read some article on using the DB tables to log each job success/failure and ignore jobs which already ran on restart, at this point I don't want to go on that route, as I feel the tcheckpoint is for that reason.

 

FYI - I use enterprise edition.

 

thanks

Mj

Labels (2)
1 Solution

Accepted Solutions
Anonymous
Not applicable

If you have implemented the tCheckPoint components correctly, restarting the job should just pick up from where you have left off. Have you tried manually restarting to test this? If you have and it works, then all you need to do is restart the job. The tCheckPoint components are described here by the guy (Alan Black) who created them (https://www.talendbyexample.com/talend-tcheckpoint-component.html).

 

To restart the job automatically might be tricky. You need to know that the job failed first. For this the only way I can think of handling this is to use a table in a DB. You *could* use a flat file I guess, but I would opt for a DB. You will then need to have a job that is scheduled to run every 5 or so minutes to check this DB/file. If the job finds that your orchestration job has failed, you can use the Talend Metaservelt API (https://help.talend.com/reader/wt3LuroKJjOAVRRWmlnFsw/FGEUR1zvb4mnG6REP5s8~g). This is described in a brief tutorial here (https://help.talend.com/reader/YcEtLnE30CM8SeOxtXyHYQ/SoQ_WNncoZj9Yg3t2xu2mA). It is a little tricky to get used to, but once you have worked it out, it is VERY useful

View solution in original post

1 Reply
Anonymous
Not applicable

If you have implemented the tCheckPoint components correctly, restarting the job should just pick up from where you have left off. Have you tried manually restarting to test this? If you have and it works, then all you need to do is restart the job. The tCheckPoint components are described here by the guy (Alan Black) who created them (https://www.talendbyexample.com/talend-tcheckpoint-component.html).

 

To restart the job automatically might be tricky. You need to know that the job failed first. For this the only way I can think of handling this is to use a table in a DB. You *could* use a flat file I guess, but I would opt for a DB. You will then need to have a job that is scheduled to run every 5 or so minutes to check this DB/file. If the job finds that your orchestration job has failed, you can use the Talend Metaservelt API (https://help.talend.com/reader/wt3LuroKJjOAVRRWmlnFsw/FGEUR1zvb4mnG6REP5s8~g). This is described in a brief tutorial here (https://help.talend.com/reader/YcEtLnE30CM8SeOxtXyHYQ/SoQ_WNncoZj9Yg3t2xu2mA). It is a little tricky to get used to, but once you have worked it out, it is VERY useful