设置超时 honor timeouts

Question

关于超时的显式设置

所有调用外部接口显式设置超时时间。比如：connectTimeout，readTimeout, writeTimeout。
Spring RestTemplate 的超时时间若不设置，则是自身没有超时（依赖于外部超时）。

MySQL 数据库连接，需要在连接属性上也配置好各种超时时间，避免默认设置。

1. 交易类的http超时时间，建议不超过3秒，最长不超过 15 秒。
2. 数据库超时时间，建议不超过3秒。例如：
   jdbc:mysql://xx.xx.xx.xx:3306/xx?characterEncoding=utf8&zeroDateTimeBehavior=convertToNull&useSSL=false&connectTimeout=3000&socketTimeout=3000
3. redis 连接的超时时间，不超过1秒。spring.redis.timeout=1000 # Connection timeout

话外之 honor timeouts

今天看到一个新闻：

OkHttp 4 正式版发布，从 Java 切换到 Kotlin
OkHttp 4 正式版发布了，此版本最大的变化就是项目从 Java 迁移到了 Kotlin。 就像官方介绍的，“此版本改变了一切，又没什么改变”，我们此前在 OkHttp 4 的 RC 3 版本更新中已经报导过，OkHttp 4.x 将实现语言从 Java 切换到了 Kotlin，用等效的 .kt 替换了 25K 行的 .java，这就是改变了一切的意思。 而“没什么改变”..

然后我就随手看了一下 github 的 OkHttp，然后又随手看到了 OkIO。

okio的readme

Timeouts. The streams provide access to the timeouts of the underlying I/O mechanism. Unlike the java.io socket streams, both read() and write() calls honor timeouts.

然后看到了honor timeouts，不懂啥意思，难道超时光荣不成？

honor单词的解释

fulfill (an obligation) or keep (an agreement).
accept (a bill) or pay (a check) when due.

源码

/**
 * Returns a sink that writes to `socket`. Prefer this over [sink]
 * because this method honors timeouts. When the socket
 * write times out, the socket is asynchronously closed by a watchdog thread.
 */
@Throws(IOException::class)
fun Socket.sink(): Sink {
  val timeout = SocketAsyncTimeout(this)
  val sink = OutputStreamSink(getOutputStream(), timeout)
  return timeout.sink(sink)
}

/**
 * Returns a source that reads from `socket`. Prefer this over [source]
 * because this method honors timeouts. When the socket
 * read times out, the socket is asynchronously closed by a watchdog thread.
 */
@Throws(IOException::class)
fun Socket.source(): Source {
  val timeout = SocketAsyncTimeout(this)
  val source = InputStreamSource(getInputStream(), timeout)
  return timeout.source(source)
}

谷歌搜索 honor timeouts 的一个 结果

The connection pool supports a timeout for setup, but the glue code that attaches it to asio doesn't actually honor this timeout. This means requests may hang forever behind connect. Even a series of requests repeat this workflow (because new connections aren't created in the pool because one is connecting forever).