MagicStack / MagicStack/asyncpg

Pool created with min_size loses inactive connections and never re-acquires them

未关闭
#1,268 0 条评论 4 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

主要语言
Python
星标
8.1k
派生
468
PR 合并指标
30 天内没有已合并 PR

描述

I love `asyncpg`, been using it for years. However recently came across some unexpected behavior with v 0.30.0. It goes a little something like this.

Initialize a pool with a `min_size`:

```python
import asyncio
import asyncpg

pool = await asyncpg.create_pool(min_size=10, max_size=20, **kwargs) # assume kwargs has connection info
print(f'Total: {pool.get_size()}')
# Total: 10
```

All good. But now let's say that nothing happens for over 5 minutes. You know, because `max_inactive_connection_lifetime=300.0` by default. And check again:

```python
await asyncio.sleep(305)
print(f'Total: {pool.get_size()}')
# Total: 0
```

Huh? Why isn't it still 10? No matter what I set for `max_inactive_connection_lifetime` this behavior happens immediately after that time, as long as the time is non-zero. If it's set to `0` then the behavior correctly doesn't happen.

I don't know if this is intentional or not, but as a developer, when I open a database pool with a minimum number of connections I expect that the pool will maintain that minimum, _even if they are automatically closed or timeout_.

贡献指南

这个仓库没有索引到贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

调研方向

从 asyncpg.create_pool 入手,检查空闲连接关闭时 max_inactive_connection_lifetime 如何与 min_size 交互。使用提供的 sleep 和 pool.get_size() 检查来复现该问题。当配置了 min_size 的 pool 在空闲连接超时后仍能保持或重新获得该最小值时,即表示完成。

由索引模型根据 Issue 内容生成。

评估

技术栈
postgresql, python
领域
databases
Issue 类型
缺陷
难度
3/5
预计耗时
1-2 天
活跃度
停滞
描述清晰度
基本清楚
新手友好度
45/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。