Azure Monitoring and Alerts

General

During the Azure deployment, we set up a few monitors and alerts to monitor the cluster servers and cluster health:

• lbcluster-deallocateNode-alerts: raises an alert when a cluster server is shut down (deallocate).

• lbcluster-deleteNode-alerts: raises an alert when a cluster server is deleted.

• lbcluster-maintenace-alerts: raises an alert when a maintenance notification is received for any of the cluster servers.

• lbcluster-serverDown-alerts: raises an alert when a server is down.

• Lightbits cluster event log alerts:

  • Warning
  • Error
  • Critical

Alerts

You can view the alerts in the managed resource group of the cluster under the Alerts section, or under the managed application in the Alerts section.

Lightbits Event Log

The Lightbits cluster has an event log. This event log is stored in the AppInsights that is created as part of the cluster deployment (and then click on the Logs menu).

In any case, any important event will raise an alert with the appropriate severity (Warning, Error, or Critical).

In the KQL, run the following Query:

traces | where message contains "EventName" | extend SplitMessage = split(message, ',')

​

| extend Time = replace_string(tostring(SplitMessage[1]), ''Time':', '')

​

| extend EventName = replace_string(tostring(SplitMessage[4]), ''EventName':', '')

​

| extend EventStatus = replace_string(tostring(SplitMessage[8]), ''Status':', '')

​

| extend ServerName1 = replace_string(tostring(SplitMessage[12]), ''Name':', '')

​

| extend ServerName = replace_string(ServerName1, '}}', '')

​

project Time, EventName, EventStatus, ServerName, SplitMessagesort by Time

​

sort by Time