PostHog/posthog

Feature Flag Hash Key Overrides table can grow out of control

Open

#16,885 opened on Aug 2, 2023

View on GitHub
 (5 comments) (0 reactions) (0 assignees)Python (988 forks)batch import
backlogbugfeature/feature-flagsgood first issueteam/feature-flagsteam/flags-platform

Repository metrics

Stars
 (16,704 stars)
PR merge metrics
 (Avg merge 12h 15m) (1,000 merged PRs in 30d)

Description

Bug description

The hashkeyoverrides table can grow seriously out of control because it keeps all rows for stale flags. We cascade delete, but since flags are soft deleted, this table doesn't clean itself up automatically, which makes it hard to keep the size manageable

Additional context

We should run a nightly/weekly job , or whenever a flag is deleted, to remove stale rows from this table.

Thank you for your bug report – we love squashing them!

Contributor guide

Research direction
The issue requires implementing a cleanup mechanism for the hashkeyoverrides table. Currently, flags are soft deleted but the overrides table retains rows. A scheduled job (e.g., using Celery) should delete rows where the associated flag is soft deleted. Alternatively, a post delete hook could be added. Review the existing model definitions in posthog/models/feature flag.py and the soft deletion logic. Ensure the job is idempotent and efficient for large tables.
Tech stack
pythonpostgresql
Domain
backenddatabase
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
Django ORMdatabase cleanup strategies
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60