crawlab-team
crawlab 重启之后，重启前 pending 的任务无法继续执行(After crawlab restart, tasks pending before restart cannot continue)
Bug 描述 当我重启crawlab之后，在重启之前处于pending的任务无法成功调度，一直处于pending状态。
复现步骤 该 Bug 复现步骤如下
- 保证当前tasks列表中有pending的任务，且node为空
- 重启master和worker节点
- 重启成功后pending的任务会一直pending
期望结果 重启crawlab之后，之前处于pending的任务可以继续调度执行。
Describe the bug When I restarted crawlab, the tasks that were in pending before the restart could not be successfully scheduled and stayed in pending state.
To Reproduce Steps to reproduce the behavior:
- Ensure that there is a pending task in the current tasks list and that the node is empty.
- Restart the master and worker nodes.
- After a successful restart, the pending tasks will remain pending.
Expected behavior After restarting crawlab, tasks that were previously in pending can continue to be scheduled for execution.
@ma-pony hey! I would appreciate if you provide relevant translation for your issue in ENG. I'm not sure if Google Translator handle it properly. Thanks)
@ma-pony hey! I would appreciate if you provide relevant translation for your issue in ENG. I'm not sure if Google Translator handle it properly. Thanks)
Thanks for the suggestion！
@ma-pony thanks for your translation. Have the same issues. In my case I have > 400 scheduled tasks and lot of them staying in pending state and never executed. Seems like Redis queues don't clear in proper way.
@ma-pony thanks for your translation. Have the same issues. In my case I have > 400 scheduled tasks and lot of them staying in pending state and never executed. Seems like Redis queues don't clear in proper way.
I have 6k+ scheduled tasks...
@ma-pony 6k+ scheduled tasks it is impressive.
What kind of Crawlab installation you using Docker based or Kubernetes and what kind of infrastructure you using.
Could you please share some details?
@ma-pony
6k+ scheduled tasksit is impressive. What kind of Crawlab installation you using Docker based or Kubernetes and what kind of infrastructure you using. Could you please share some details?
https://docs.crawlab.cn/en/guide/installation/kubernetes.html
four 8C16G machines, deployed using k8s, one master node and six worker node， with worker configured with 4C16G and master configured with 6C12G