Get started with Event Monitoring

Create an alert group with yourself in it, and an alert that emails you when memory use on your nodes goes above 80%.

You need at least one node in the Running state, in the project and region selected in MyAccount. The alert form lists running resources only.

You also need create permission in Event Monitoring. With it, MyAccount shows Create Alert and Create Alert Group. For the permission each action needs, see Troubleshooting Event Monitoring.

Open Event Monitoring

In the MyAccount side menu, click Event Monitoring. The Monitoring Dashboard opens, with Problems by severity at the top. A problem is something an alert has found, such as memory above its threshold on a node. Below the counts are the open problems that began in the last 30 minutes, the period the dashboard opens on. When there are none, the list reads No problems found.

Click Manage Alerts. The list of your alerts opens. Alongside the alerts you create, it holds the default alerts E2E adds to each new resource it monitors.

Make an alert group

An alert group is the list of people an alert notifies. Make one with yourself in it.

1. On Manage Alerts, click Manage Alert Groups, then Create Alert Group. The Create Alert Group form opens.
2. In Alert Group Name, enter a name of three to 50 characters, using letters, digits, underscores, or hyphens, such as ops-team.
3. Under Select User(s), click Select. The Users list opens with the people you can add. For who appears there, see Who you can add.
4. Tick your own name, and click Add.
5. Click Create.

MyAccount shows Alert Group created successfully. and returns to Manage Alert Groups, where your group is listed with your name under Users.

Create an alert on your nodes

Now create an alert that watches memory on all your running nodes at once.

1. In the side menu, click Event Monitoring, then Manage Alerts.
2. Click Create Alert. The Create Alert form opens.
3. In Service for alert, choose Node.
4. In Associated Resources, choose All, or choose the nodes you want to watch. The list shows only your running nodes.
5. In Alert trigger type, choose Consumed Memory, the share of each node’s memory in use.
6. In Operator, choose Greater than.
7. In Threshold value (%), enter 80. The field takes a number from 10 to 100.
8. Leave Severity on High. Severity only labels the problem; see Severity.
9. In Alert Group(s), choose the group you made. With every field filled in, an Alert Summary sentence appears and describes your alert. Read it to check that the alert does what you want.
10. Click Create. MyAccount shows Alert created successfully. and returns to Manage Alerts.

See your alert

Your alert is one row on Manage Alerts, however many nodes it watches. Alert Condition reads “Alert when Memory is greater than 80 %”, Service reads Node, and Alert Group(s) shows your group. To see the nodes the alert covers, point at Node.

When memory on one of those nodes goes above 80%, you get an email about that node. It goes to the address you sign in to MyAccount with. The email comes once, when the problem starts. The problem stays open while memory is above 80%. The Monitoring Dashboard lists it whenever the period you choose covers the time it began.

What to do next

• Read the monitoring dashboard: find the open problems, narrow them by severity or resource, and see each resource’s graphs.
• Create an alert: every choice in the alert form, for nodes, load balancers, DBaaS instances, and Kubernetes nodes.
• Manage alert groups: add your colleagues, and choose who gets each alert’s notifications.
• Alerts and problems: what severity means, the default alerts E2E adds, and who gets notifications by email and on WhatsApp.
• Troubleshooting Event Monitoring: help with any message the alert form shows, and with alert emails.