Errors like 403,404 while running FUAM pipelines

Hello Shreya_Barhate 

Since the child pipelines run successfully when you run them interactively, it suggests to me that it is related to API authorization on the identity runnig it. When you run a pipeline directly, the REST APIs execute under your identity context, and you may have all the required authorization - hence they succeed! 

When the E2E pipeline orchestrates it, it is using the connection identity, which is probably lacking the Admin API access in tenant settings. 

There could also be a case of concurrency, since you mention that not all errors are consistent. This is probably because the Admin APIs are throttled, and you're hitting those limits. 

I suggest you use the following:

 - a service principal enabled with admin API setting for running the pipelines 

 - try running child pipelines sequentially or with a low batchcount, with retries