Alarm related

This section explains common problems related to alarms

Q1:The alarm strategy has been configured, and the monitoring indicators are abnormal but no alarm has been sended?

May be the following situations:

  1. View object mount:The object (usually a machine) needs to be mounted under the node (or its descendants) that configures the policy
  2. View monitoring data:Confirm that the monitoring data corresponding to the monitoring strategy is valuable and will definitely trigger the threshold
  3. Incorrect threshold setting:Check the alarm function settings of the alarm strategy, confirm whether the alarm trigger condition is met, and confirm whether the effective time of the strategy is met
  4. Policy is blocked:Check the alarm policy blocking list to confirm whether the policy is blocked
  5. Notify gateway problems:On the alarm history page, check whether there is an alarm event that has been generated. If there is an alarm event, the problem is to notify the gateway
  6. Issue of strategy
    • Run curl '127.0.0.1: 5800 /api /portal /stras /effective? All = 1' to get a list of all strategies
    • If the list does not have this strategy, check the monapi logs to see if there are any error messages. If there are errors, follow the prompts
    • If so, check the judge_instance field, find out which instance of the judge the strategy is distributed to, and log in to the machine where the judge is located, Run curl 127.0.0.1: 5840 /api /judge /stra /: id to see if it has been sent to judge
  7. judge解析策略异常:Check the judge WARNING.log and ERROR.log logs to check if there is any error message for this strategy
  8. judge没有收到数据:Modify the judge log level to DEBUG, tail -f DEBUG.log | grep monitoring indicator No log output, indicating that the data has not reached the step of judge
  9. 查看数据链路:Check the transfer log and analyze why the data is not transferred from the transfer to the judge module
Last modified April 15, 2020: add en content/en/docs/faq (#11) (dfa721f)