实现方式采用spring的bean后置装载的接口
代码：
public class ExecutorMetricWapper ``implements BeanPostProcessor { ``@Override`` ``public Object postProcessAfterInitialization(Object bean, String beanName) ``throws BeansException {`` ``if ((bean ``instanceof Executor)) {`` ``ExecutorServiceMetrics.monitor(Metrics.globalRegistry, (Executor) bean, beanName, Collections.emptyList());`` ``LOGGER.info(``"Executor beanName:{} Has been registered to monitor"``, beanName);`` ``return bean;`` ``}`` ``return bean;`` ``} ``private static final Logger LOGGER = LoggerFactory.getLogger(ExecutorMetricWapper.``class``);``} |
|---|
效果：
只是公司内的grafana监控图
指标的含义：
spring自带的ThreadPoolTaskExecutor线程池（至于spring的Executor为什么参数没有自定实现的Executor线程池全，这问题是由于实现机制的原因）
| 指标 | 含义 | 推荐expr |
|---|---|---|
| ```html | ||
| executor_idle_seconds_count |
| ```html
executor_idle_seconds_sum
``` | 任务耗时 | 同上 |
| ```html
executor_idle_seconds_max
``` | 任务空闲秒数（最大值） | -- |
| ```html
executor_execution_seconds_count
``` | 任务执行次数 | 执行器平均执行秒数:increase(executor_execution_seconds_sum[5m])/increase(executor_execution_seconds_count[5m]) |
| ```html
executor_execution_seconds_sum
``` | 任务耗时 | 同上 |
| ```html
executor_execution_seconds_max
``` | 任务执行秒数（最大值） | -- |
自定义的Executor线程池
| 指标 | 含义 | 推荐expr |
| ------------------------------------------ | ------------ | ------------------------------------------------------------------------------------------------------- |
| ```html
executor_completed_tasks_total
``` | 完成的任务 | 完成的任务/分钟：increase(executor_completed_tasks_total[1m]) |
| ```html
executor_idle_seconds_count
``` | 执行器闲置次数 | 执行器平均闲置秒数：increase(executor_idle_seconds_sum[5m])/increase(executor_idle_seconds_count[5m]) |
| ```html
executor_idle_seconds_sum
``` | 执行器耗时 | 同上 |
| ```html
executor_idle_seconds_max
``` | 执行器空闲秒数（最大值） | -- |
| ```html
executor_active_threads
``` | 执行器活跃的线程数 | -- |
| ```html
executor_seconds_count
``` | 执行器执行次数 | 执行器平均执行秒数:increase(executor_seconds_sum[5m])/increase(executor_seconds_count[5m]) |
| ```html
executor_seconds_sum
``` | 执行器耗时 | 同上 |
| ```html
executor_seconds_max
``` | 执行器执行秒数（最大值） | -- |
| ```html
executor_queued_tasks
``` | 排队的任务 | 队列占用率：100*(1-executor_queue_remaining_tasks/ (executor_queue_remaining_tasks + executor_queued_tasks )) |
| ```html
executor_queue_remaining_tasks
``` | 排队的剩余任务 | 同上 |
| ```html
executor_pool_size_threads
``` | 执行器的线程池大小 | -- |
| executor_rejected_tasks_total | 拒绝的任务 | 拒绝的任务/分钟：increase(executor_rejected_tasks_total[1m]) |
| executor_scheduled_cron_total | 执行器预定的cron | -- |
| executor_scheduled_once_total | 执行器预定一次大小 | -- |
| executor_scheduled_repetitively_total | 执行器重复执行次数 | -- |
参照：INFRA|springboot2.X统一配置面板的grpc统计模块
其他的监控也是类似，甚至可以自己来写实现
从这个问题主要是引出监控的事情，要是想要对一个线程池进行matrix监控，那么当前比较low的实现方法如下
| `@Service``public` `class` `ThreadPoolMonitor ``implements` `InitializingBean {` ` ``@Resource``(name = ``"consumerExecutor"``)`` ``ThreadPoolTaskExecutor consumerExecutor;`` ``@Resource`` ``ConnectionPool connectionPool;` ` ``private` `static` `final` `Iterable<Tag> consumerExecutorTAG = Collections.singletonList(Tag.of(``"thread.pool.name"``, ``"consumerExecutor"``));`` ``private` `static` `final` `Iterable<Tag> okHttpExecutorTAG = Collections.singletonList(Tag.of(``"thread.pool.name"``, ``"okHttpExecutor"``));`` ``private` `final` `ScheduledExecutorService scheduledExecutor = Executors.newSingleThreadScheduledExecutor();` ` ``private` `Runnable monitor = () -> {`` ``//这里需要捕获异常,尽管实际上不会产生异常,但是必须预防异常导致调度线程池线程失效的问题`` ``try` `{`` ``Metrics.gauge(``"thread.pool.core.size"``, consumerExecutorTAG, consumerExecutor, ThreadPoolTaskExecutor::getCorePoolSize);`` ``Metrics.gauge(``"thread.pool.largest.size"``, consumerExecutorTAG, consumerExecutor, e -> e.getThreadPoolExecutor().getLargestPoolSize());`` ``Metrics.gauge(``"thread.pool.max.size"``, consumerExecutorTAG, consumerExecutor, ThreadPoolTaskExecutor::getMaxPoolSize);`` ``Metrics.gauge(``"thread.pool.active.size"``, consumerExecutorTAG, consumerExecutor, ThreadPoolTaskExecutor::getActiveCount);`` ``Metrics.gauge(``"thread.pool.thread.count"``, consumerExecutorTAG, consumerExecutor, ThreadPoolTaskExecutor::getPoolSize);`` ``Metrics.gauge(``"thread.pool.queue.size"``, consumerExecutorTAG, consumerExecutor, e -> e.getThreadPoolExecutor().getQueue().size());` ` ``Metrics.gauge(``"thread.pool.connectionCount"``, okHttpExecutorTAG, connectionPool, ConnectionPool::connectionCount);`` ``Metrics.gauge(``"thread.pool.idleConnectionCount"``, okHttpExecutorTAG, connectionPool, ConnectionPool::idleConnectionCount);`` ``} ``catch` `(Exception e) {`` ``//ignore`` ``}`` ``};` ` ``@Override`` ``public` `void` `afterPropertiesSet() ``throws` `Exception {`` ``// 每5秒执行一次`` ``scheduledExecutor.scheduleWithFixedDelay(monitor, ``0``, ``5``, TimeUnit.SECONDS);`` ``}` `}` |
| ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
