Pin individual events for longer periods than 90 day default retention
- Dominant language
- Python
- Stars
- 44.8k
- Forks
- 4.9k
- Avg merge
- 22h 21m
- Merged PRs (30d)
- 586
Description
### Problem Statement
Currently, Sentry does not allow the retention of issues or events for more than 90 days. This limitation poses a significant problem for teams maintaining desktop applications with long lifecycles.
In our case, these applications are used in Game Development, VFX, and Production workflows, where software versions remain fixed for extended periods—often years—to ensure consistent and reproducible results. Because of this, some bugs and crashes are difficult to reproduce or resolve within a 90-day window.
Furthermore, our tools serve a specialized user base, meaning issues occur infrequently. Losing historical events due to retention limits severely impacts our ability to investigate, diagnose, and improve software reliability.
### Solution Brainstorm
A pin system could allow users to retain individual events or issues for longer than the standard 90-day retention period — possibly even with an undefined duration.
This would enable teams to keep critical or hard-to-reproduce events beyond the automatic deletion window, which is especially valuable for long-lived desktop applications where issues may only surface or be addressed after several months.
To ensure fair usage and prevent storage abuse, these pins could be managed under a quota system, for example:
- 1000 pins per year, or
- 100 pins per month, or
- Up to 10 pinned events per issue
This approach would give teams the flexibility to preserve essential debugging data without significantly impacting Sentry’s storage model.
### Product Area
Unknown
Contributor guide
Research direction
No files, tests, or entry points are identified in the issue. Start by locating the existing 90-day retention implementation and the data model for events or issues, then clarify the pinning and quota requirements before defining what an implementation and its tests would need to demonstrate.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- observability-sre
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100