This depends on what you want to do: If you want to run the master package and have re-startability in the master package ( I mean when a child package failed then in next run of master package execution will be started from that child package), then you just need one checkpoint in the master package. but if you want to have re-startability in each package ( I mean for example in the package1 you may have 10 tasks, and if task #3 failed in the next execution you want to start from task #3 not from the start of package), then you need to have a checkpoint for each package.http://www.rad.pasfu.com

Check points are implemented on Package level not the Project level so every package file should have its own check point file. Look at it this way: each package has its own set of variables and check point is all about variable values of a package NOT the project. One good article you can look into for this is: http://www.sqlservercentral.com/articles/Integration+Services+(SSIS)/65473/ Hope this helps.Please mark the post as answered if it answers your question

hi i created a package in ssis..in that project i am having 3 package and 1 master package.now i want to create a check point file for this..i am newly for this check point process.. so i m having some question: shall i create 1 xml file for check point and shall i use same xml file for all 4 package means whether it will work???

Refer this wherein it clearly defines re-start points in check-pointing. An excerpt is "The completion of any child containers of transacted container is not recorded in the checkpoint file. Therefore, when the package is restarted, the transacted container and its child containers run again." http://btsbee.wordpress.com/

This depends on what you want to do: If you want to run the master package and have re-startability in the master package ( I mean when a child package failed then in next run of master package execution will be started from that child package), then you just need one checkpoint in the master package. but if you want to have re-startability in each package ( I mean for example in the package1 you may have 10 tasks, and if task #3 failed in the next execution you want to start from task #3 not from the start of package), then you need to have a checkpoint for each package.http://www.rad.pasfu.com

Check points are implemented on Package level not the Project level so every package file should have its own check point file. Look at it this way: each package has its own set of variables and check point is all about variable values of a package NOT the project. One good article you can look into for this is: http://www.sqlservercentral.com/articles/Integration+Services+(SSIS)/65473/ Hope this helps.Please mark the post as answered if it answers your question