Closed LiJun990105 closed 4 months ago
服务器下线时,如果正常执行spring 自带的destroy()方法,执行器会向xxljob admin 发送一个http请求,最终执行com.xxl.job.admin.core.thread.JobRegistryHelper#registryRemove方法,该方法会在registry表中把当前节点去掉,但是不会更新group表,而实际调度是根据group表中的address_list来选择pod。group表中的地址信息是30s更新一次。所以最快也要30s。如果destroy()方法没有执行,只能等xxljob admin 90s没有收到心跳,再去registry表中把当前节点去掉,再更新group表。建议1、com.xxl.job.admin.core.thread.JobRegistryHelper#registryRemove方法中同步更新group表。 2、如果多次因为网络调度失败,将当前节点排除。
Please answer some questions before submitting your issue. Thanks!
Which version of XXL-JOB do you using?
Expected behavior
Actual behavior
Steps to reproduce the behavior
Other information