[FEATURE] Add functionality to allow deferring deletion of shrunk index
- Dominant language
- Python
- Stars
- 3.1k
- Forks
- 627
- PR merge metrics
- No merged PRs in 30d
Description
Currently the shrink action only offers the option to immediately delete the original unshrunk index following a successful shrink. It may be desirable to defer deletions because it's more efficient to create the shrink index with 0 replicas as it allows for quicker processing (think applying best_compression, etc.) and create replicas after the processing is done only. Only after the shrunk index is processed and replicated should the original unshrunk index be deleted (to maintain redundancy across the whole process).
Cf. https://discuss.elastic.co/t/curator-delay-deleting-shrunk-indices/101437
Contributor guide
Research direction
Start by reading the shrink action described in this issue and the linked discussion about delaying deletion. Clarify how the shrunk index should be processed and replicated before the original unshrunk index is removed. Done means the workflow can defer deletion while preserving redundancy throughout processing.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- databases
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100