Hot reload for worker runtime parameters
- Dominant language
- Rust
- Stars
- 32
- Forks
- 5
- Avg merge
- 15h 21m
- Merged PRs (30d)
- 21
Description
### Problem
Worker runtime parameters are currently read at startup only. In practice, changes to scheduling interval, concurrency, timeout, retry, and rate-limit settings usually require a worker restart to take effect. That adds operational overhead and makes online tuning harder.
### Request
Support hot reload for worker runtime parameters so configuration changes can take effect without restarting workers.
### Suggested scope
- Runtime knobs: `poll_interval`, `claim_batch_size`, `deadline_duration`, `priority_aging_interval`, concurrency, rate limiting, and similar settings.
- Declarative worker-owned business config: include it if it is safe to update online.
### Out of scope
- Task code
- Execution logic
- One-off task inputs
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by tracing where workers read poll_interval, claim_batch_size, deadline_duration, priority_aging_interval, concurrency, rate limiting, and related settings. Determine how configuration changes could reach running workers and which declarative business configuration is safe to update online; done means supported runtime changes take effect without a restart, while task code and one-off inputs remain unchanged.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- backend
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 38/100