Preventing mass removal
- Dominant language
- Python
- Stars
- 21
- Forks
- 2
- PR merge metrics
- No merged PRs in 30d
Description
> avoid removing too many ZIMs at once => this is a bit blurry, but the scenario I have in mind is that should we publish an empty catalog for any reason (or should the program believe the catalog is empty for any reason), it would be a pity / concern to cause all ZIMs of all peers to be deleted within few hours ; maybe even deleting too many ZIMs at once is an indicator of an issue ; that been said, since we keep the ZIMs for `--keep` days, the impact is obviously not immediate ... should we simply have an alerting system? a summary file one can watch with details about files which are going to be removed in 1 week, 2 weeks, 4 weeks, 8 weeks, 12 weeks?
_Originally posted by @benoit74 in https://github.com/kiwix/seeder/pull/6#pullrequestreview-2575066969_
Contributor guide
Research direction
Start by reviewing the seeder's catalog-processing and ZIM-removal behavior, especially the `--keep` retention period. Define how the program should detect or prevent unusually large removals, and whether an alert or summary file is the intended outcome. Done should include an agreed safeguard and a documented way to verify pending deletions.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- backend
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100