`supabase start` waits about 10 seconds for Docker's first health check after services are ready
维护者通常 1 天内回复
评估
- 难度
- 3/5
- 预计耗时
- 1-2 天
- 新手友好度
- 70/100
- Issue 类型
- 缺陷
- 描述清晰度
- 基本清楚
- 活跃度
- 活跃
- 技术栈
- docker, typescript
调研方向
Start by reading StartHealthcheckSpec and docker-create-args.ts, the two locations named in the issue, and check how Docker Engine versions are handled. Verify how health-check options are passed for auth, realtime, storage, and kong. Done means supported engines use a 1-second start interval while retaining the 10-second steady-state interval, and older engines omit the flag.
由索引模型根据 Issue 内容生成。
描述
Existing issues
- I have searched the existing issues.
Affected area
Local development
Problem to solve
supabase start finishes about 10 seconds after every service is ready, because it waits for Docker's first health check and Docker runs that one full interval after the container starts.
auth, realtime, storage and kong use --health-interval 10s with no start interval. The CLI polls every second, but the container reports starting until the first check runs.
Timings from supabase start restarting an existing project, on 3.0.0-beta.17, Docker 29.8.1 and WSL2. I ran each container's own health check command inside it every 0.3 s to see when it first passed:
| Container | Created | Health check first passes | Docker's first health check |
|---|---|---|---|
| db | 0.7 s | 0.7 s | ~10.8 s |
| kong | 1.3 s | 1.8 s | ~11.6 s |
| auth | 1.5 s | 1.6 s | ~11.8 s |
| realtime | 1.7 s | 4.1 s | ~12.0 s |
| storage | 2.0 s | 3.0 s | ~12.1 s |
The command exited at 13.9 s, with everything passing by 4.1 s. This is one run with studio,imgproxy,vector,logflare,edge-runtime,postgres-meta,supavisor,mailpit,postgrest excluded. A first start with an empty volume showed the same gap (54.1 s and 63.8 s, including image pulls).
--ignore-health-check doesn't skip the wait.
Proposed solution
Pass Docker's --health-start-interval (for example 1s) with --health-start-period on these services, so the first checks run soon after start while the steady-state interval stays at 10 seconds. StartHealthcheckSpec and docker-create-args.ts would each need one new field.
The flag needs Docker Engine 25 or later (API 1.44), so older engines would need it left out. I haven't tested other container runtimes.
Alternatives considered
Lowering intervalSeconds is simpler and works on every engine, but it keeps running the checks at that rate for the life of the stack.
Additional context
Postgres already moved to a connection from the host in #7362, which took about 10 seconds off a restart compared with 2.122.0 (23.5 s in the same test). This is the remaining part of the same wait.
Happy to open the PR if this gets the open-for-contribution label.
This was written by Opus 5.5.
- 主要语言
- TypeScript
- 星标
- 2.4k
- 派生
- 540
- 平均合并
- 17 小时 3 分钟
- 30 天内合并 PR
- 390
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
supabase/cli 的其他 Issue
-
db schema declarative sync: no way to fail (non-zero exit) when the generated migration is destructive可能重新可做 关联的 PR 已关闭且未合并。 未关闭✨ Feature supabase/cli
难度 2/5 1-3 小时 新手友好度 84/100
维护者通常 1 天内回复
-
🐛 Bug supabase/cli
难度 4/5 3-5 天 新手友好度 48/100
维护者通常 1 天内回复
-
stack: every Edge Function request starts a new worker, so the same function never reuses one and each call pays a cold start可能已有人在做 @7ttp 于 7 天前认领。 未关闭🐛 Bug supabase/cli
难度 3/5 1-2 天 新手友好度 58/100
维护者通常 1 天内回复
-
🐛 Bug supabase/cli
难度 4/5 3-5 天 新手友好度 48/100
维护者通常 1 天内回复
-
Explicit local function bundling falls back to API bundling when Docker is unavailable可能已有人在做 @7ttp 于 6 天前认领。 未关闭🐛 Bug open-for-contribution supabase/cli
难度 3/5 1-2 天 新手友好度 68/100
维护者通常 1 天内回复
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 75/100
维护者通常 1 天内回复
-
难度 1/5 1 小时以内 新手友好度 83/100
维护者通常 1 天内回复
-
🔊 a11y
难度 2/5 1-3 小时 新手友好度 82/100
betagouv/mon-entreprise#4726 ·
维护者通常 9 天内回复
-
难度 2/5 1-3 小时 新手友好度 68/100
Deltares/fews-pi-requests#357 ·
维护者通常 1 天内回复
-
bug confirmed js v6
难度 2/5 1-3 小时 新手友好度 85/100
维护者通常 1 天内回复