Thanks for your question. You highlighted two different cases, the first was FAILED objects and the second was the existing set of objects before replication was enabled.
For FAILED objects, your option for re-replicating and moving them to a COMPLETED state is to copy those objects within the source bucket. You can copy the object back to the same key, creating a new version and re-replicating the object. A lifecycle policy could then be used to expire previous/noncurrent versions of objects if desired. When copying the objects manually, there is no way to change the replication status of the object, it is only updated via replication.
For existing objects, you can activate replication of those objects using the guidance on this page: https://docs.aws.amazon.com/AmazonS3/latest/dev/replication-what-is-isnot-replicated.html
Adding S3 Bucket Policy Cause S3 Replication FailedAccepted Answerasked 6 months ago
How to remove replication statusasked a year ago
S3 replication path customisedasked 2 months ago
Does S3 Cross Account replication happens over Public Internet or Private Network of AWS backboneasked 8 days ago
AWS S3 Cross Replication - FAILED replication status for prefixasked 3 years ago
How to implement S3 replication for existing objects in CDK?asked a month ago
Why do we need S3 bucket versioning enabled in order to do replication?Accepted Answerasked 2 years ago
S3 Hudi Replication and Failoverasked 6 months ago
Manually set S3 replication statusAccepted Answerasked 2 years ago
Does AWS S3 CRR/SRR replication order match the order of object creation?asked 3 years ago