Does the CheckExists method determine whether there is a delay in the existence of a job?

[ZLVincent]

Describe the bug

The results obtained by calling the CheckExists method and the GetCurrentlyExecutingJobs method are different

Version used

3.5.0

Expected behavior

I saw that the implementation of CheckExists is to judge whether the Job exists from the JOB_DETAILS table of the data. However, GetCurrentlyExecutingJobs seems to get the running Jobs from the memory. I don't know if there is a delay in updating the database. If there is a delay, will there be inconsistent results between CheckExists and GetCurrentlyExecutingJobs?

[lahma]

They serve a different purpose. CheckExists will tell if job has been registered, it might not have any triggers associated and thus might not ever be even running. GetCurrentlyExecutingJobs reports jobs that are currently actively executing in the specific scheduler instance.

[lahma]

Two queries at different point of time can return different results, they are not part of the same transaction after all.

[ZLVincent]

ah...So these two should not be used together, Thanks a lot

[lahma]

Especially in clustered environment you probably won't get the right answer as some job could be running on different node which wouldn't be included in other node's currently executing list.

[ZLVincent]

Then in the cluster environment, how to get the currently executing list?

[lahma]

There's no API for that which would report entire cluster state.