Restart onError sampling for replays after stopping
- 主要语言
- Python
- 星标
- 44.8k
- 派生
- 4.9k
- 平均合并
- 22 小时 21 分钟
- 30 天内合并 PR
- 586
描述
### Problem Statement
We received the following message from a customer:
One account in our team has multiple roles that can be switched. We hope that when the user switches roles, the errors that occur will be recorded in a new Replays.
The latest version does not have replay.setupOnce(). Simply using replay.stop() and replay.start() will not generate new Replays. I only want to use replaysOnErrorSampleRate, because if replaysSessionSampleRate or tracesSampleRate is used, replay.start() will be executed when switching users, and then a Replay without error records will be generated.
### Solution Brainstorm
Expose a public API to allow for restarting of onError sampling after replay has been stopped.
### Product Area
Replays
贡献指南
调研方向
Start by reviewing the replay lifecycle around replay.stop() and replay.start(), focusing on how replaysOnErrorSampleRate behaves after stopping. Define the public restart behavior so switching roles can create a new error-triggered Replay without enabling session or trace sampling. The issue names no files or tests, so those entry points must be located first.
由索引模型根据 Issue 内容生成。
评估
- 领域
- observability
- Issue 类型
- 功能
- 难度
- 5/5
- 预计耗时
- 一周以上
- 活跃度
- 停滞
- 描述清晰度
- 需要澄清
- 新手友好度
- 35/100