我的日志里充满了绝地武士的例外，即使是日志级别的信息我也不能忽略
加载缓冲区花费了2942毫秒：redis.clients.jedis.binaryjedis.ping（binaryjedis）上的java.lang.nullpointerexception。java:187)在redis.clients.jedis.jedisfactory.validateobject（jedisfactory。java:152)在org.apache.commons.pool2.impl.genericobjectpool.returnobject（genericobjectpool。java:534)在redis.clients.jedis.util.pool.returnresourceobject（池。java:68)在redis.clients.jedis.jedispool.returnresource（jedispool。java:251)在redis.客户.绝地.绝地.接近（绝地。java:3505)在com.puls.availability.services.availabilityserviceimpl.getavailabilitybufferpervertical（availabilityserviceimpl。java:360)在java.base/java.lang.thread.run（线程。java:834)redis.clients.jedis.binaryjedis.quit（binaryjedis）上的java.lang.nullpointerexception。java:255)在redis.clients.jedis.jedisfactory.destroyobject（jedisfactory。java:99)在org.apache.commons.pool2.impl.genericobjectpool.destroy（genericobjectpool。java:927)在org.apache.commons.pool2.impl.genericobjectpool.evict（genericobjectpool。java:768)在org.apache.commons.pool2.impl.basegenericobjectpool$exector.run（basegenericobjectpool。java:1138)在java.base/java.util.concurrent.executors$runnableadapter.call（executors。java:515)在java.base/java.util.concurrent.futuretask.runandreset（futuretask。java:305)在java.base/java.util.concurrent.scheduledthreadpoolexecutor$scheduledfuturetask.run（scheduledthreadpoolexecutor。java:305)位于java.base/java.util.concurrent.threadpoolexecutor.runworker（threadpoolexecutor。java:1128)在java.base/java.util.concurrent.threadpoolexecutor$worker.run（threadpoolexecutor。java:628)在java.base/java.lang.thread.run（线程。java:834)redis.clients.jedis.binaryjedis.quit（binaryjedis）上的java.lang.nullpointerexception。java:255)在redis.clients.jedis.jedisfactory.destroyobject（jedisfactory。java:99)在org.apache.commons.pool2.impl.genericobjectpool.destroy（genericobjectpool。java:927)在org.apache.commons.pool2.impl.genericobjectpool.evict（genericobjectpool。java:768)在org.apache.commons.pool2.impl.basegenericobjectpool$exector.run（basegenericobjectpool。java:1138)在java.base/java.util.concurrent.executors$runnableadapter.call（executors。java:515)在java.base/java.util.concurrent.futuretask.runandreset（futuretask。java:305)在java.base/java.util.concurrent.scheduledthreadpoolexecutor$scheduledfuturetask.run（scheduledthreadpoolexecutor。java:305)位于java.base/java.util.concurrent.threadpoolexecutor.runworker（threadpoolexecutor。java:1128)在java.base/java.util.concurrent.threadpoolexecutor$worker.run（threadpoolexecutor。java:628)在java.base/java.lang.thread.run（线程。java:834)redis.clients.jedis.binaryjedis.quit（binaryjedis）上的java.lang.nullpointerexception。java:255)在redis.clients.jedis.jedisfactory.destroyobject（jedisfactory。java:99)在org.apache.commons.pool2.impl.genericobjectpool.destroy（genericobjectpool。java:927)在org.apache.commons.pool2.impl.genericobjectpool.evict（genericobjectpool。java:768)在org.apache.commons.pool2.impl.basegenericobjectpool$exector.run（basegenericobjectpool。java:1138)在java.base/java.util.concurrent.executors$runnableadapter.call（executors。java:515)在java.base/java.util.concurrent.futuretask.runandreset（futuretask。java:305)在java.base/java.util.concurrent.scheduledthreadpoolexecutor$scheduledfuturetask.run（scheduledthreadpoolexecutor。java:305)位于java.base/java.util.concurrent.threadpoolexecutor.runworker（threadpoolexecutor。java:1128)在java.base/java.util.concurrent.threadpoolexecutor$worker.run（threadpoolexecutor。java:628)在java.base/java.lang.thread.run（线程。java:834)
我的池配置是：
final JedisPoolConfig poolConfig = new JedisPoolConfig();
poolConfig.setMaxTotal(128);
poolConfig.setMaxIdle(128);
poolConfig.setMinIdle(16);
poolConfig.setTestOnBorrow(true);
poolConfig.setTestOnReturn(true);
poolConfig.setTestWhileIdle(true);
poolConfig.setMinEvictableIdleTimeMillis(Duration.ofSeconds(60).toMillis());
poolConfig.setTimeBetweenEvictionRunsMillis(Duration.ofSeconds(30).toMillis());
poolConfig.setNumTestsPerEvictionRun(3);这是我的创作：
new JedisPool(poolConfig, host, port, 1000);它从哪里来？
1条答案
按热度按时间vq8itlhq1#
使这个问题难以解决的是例外的隐藏部分，
异常应该如下所示：
原始异常中缺少此行：
co.elastic.apm.agent.redis.RedisSpanUtils.createRedisSpan(RedisSpanUtils.java:)该异常发生在elastic的javaapm代理中，但由于该代理使用java的instrumentation api，因此没有跟踪该异常。在等待elastic解决问题时，您可以禁用代理监听redis请求。