The short video, "Getting Started", will get you started with Cloud Monitoring.
To use Cloud Monitoring with one of your projects, do the following:
- Visit the Cloud Monitoring Console, select the project you wish to monitor, and select Continue. Alternatively, from the Cloud Platform Console menu, select Operations > Monitoring.
- If this is the first time you have used Cloud Monitoring with your project, select Enable Monitoring.
You are now looking at the Cloud Monitoring Console. The information shown will vary depending on the Google services you are using and the monitoring features you have set up, but it will look something like the following:
Monitoring Google App Engine projects
The preceding screenshot displays some sections for App Engine projects:
App Engine shows a list of your App Engine modules along with the number of incidents for each module. Selecting any of the modules here will take you to the detailed dashboard for that module.
Uptime Checks > GAE Module shows a list of configured uptime checks for your App Engine applications. An App Engine check consists of the module and a path, so in the preceding screenshot you see default / indicating that the check is monitoring the top level URL for module default. See Uptime checks to see how the checks are configured.
To get more detailed App Engine monitoring information select the App Engine section title or select Services > App Engine in the Cloud Monitoring Console top menu.
Monitoring Google Compute Engine projects
The preceding screenshot includes the section Uptime Checks > Instance, which shows a list of uptime checks for your Compute Engine VM instances. To configure these checks, see Uptime checks.
For more Compute Engine instance information, try the following:
- Select Infrastructure > Instances in the Cloud Monitoring Console top menu for an overview of your instances.
- Select an individual instance name for details about that instance.
- Install the monitoring agent and optional third-party plugins on your VM instances to get additional metrics.
The Event Log
The Event Log is a list of events that have occurred in your project. You can see the log in the Events section of the Cloud Monitoring Console, or you can select Alerting > Event Log from the top menu bar. You can filter the Event Log to see only events of a particular type or events from a particular service. You can also add your own message as an event, to keep a record not captured by the monitoring software.
Event Log entries can note changes related to the following:
- user accounts
- alerting policies
- metric values
- code deployments
- Google Cloud status changes, such as outages
- Identity and Access Management roles
- security groups
- incident status
- resource additions, removals, and tag changes
- manually-created notices
The short video, "Uptime Checks", will get you started with uptime checks.
Uptime checks let you quickly verify the health of any web page, instance, or group of resources. Each configured check is regularly contacted from a variety of locations around the world. Uptime checks can be used as conditions in alerting policy definitions.
You can display your checks and their status by selecting Alerting > Uptime Checks on the Cloud Monitoring Console top menu. You'll also find Uptime Checks sections on the Cloud Monitoring Console dashboard and on pages devoted to specific resources. For uptime checks that cover a group of resources, you can expand the check to show the status of the individual members of the group.
Create an uptime check
You might be prompted to create an uptime check when Cloud Monitoring detects the creation of a resource. You can create an uptime check at any time by following these steps:
Select Add Check in the Uptime Checks section of the Cloud Monitoring Console.
Select a Check Type: HTTP, HTTPS, TCP, or UDP.
Select a Resource Type:
Choose Instance to check a single Compute Engine instance or a group of instances.
Choose App Engine to check an App Engine module.
Choose URL to check an arbitrary web-accessible point.
Depending on the type chosen, fill in the requested information about the resource.
Fill in Path, or accept the default path.
Select a time interval for Check every.
When the Test button at the bottom of the panel is enabled, select it to test your check.
If you have problems configuring your check, review the Advanced Options section.
When you create an uptime check, you are prompted to create an alerting policy for the check. Unless you create an alerting policy, you will have to look at the Uptime Checks section of the Cloud Monitoring Console to know if your check is healthy. Select Create Alerting Policy to create a policy using a form that is already populated for the check—you only have to supply a notification method. See Alerting Policies for more information.
Dashboards and charts
The short video, "Custom Dashboards", will get you started with dashboards and charts.
You can display the metrics collected by Cloud Monitoring in your own charts and dashboards. A dashboard is a screen that can include one or more charts. Here is a custom dashboard with one chart displaying the volume of network traffic coming into a Compute Engine VM instance:
Create a dashboard and chart
Follow these instructions to create a custom dashboard and chart:
Start from the Cloud Monitoring Console.
Select Dashboards > Create... from the top menu. You will see the text "Untitled Dashboard" above two time scales and a button labeled Add Chart.
Select and replace the text "Untitled Dashboard" with whatever name you wish. In the example it is "My custom dashboard."
Select Add Chart on the dashboard.
Choose a Resource Type from the drop-down list. In the example it is Instance—a Compute Engine VM instance.
Choose a Metric Type from the drop-down list. In the example it is "Network Inbound Traffic (GCE Monitoring)." The listed metrics will be appropriate for your resource type and might depend on other factors, such as whether you installed the Stackdriver Agent.
Optionally, change Title to be a name for the chart if you want it to be different from the automatically generated title of "Instance - Network Inbound Traffic (GCE Monitoring)"
Under Advanced Options, you can aggregate the time series across resources to view a single line representing the average, median, 95th percentile, sum, etc. You can also add a threshold line to provide a visual indication of normality.
You can display your dashboard from any page in the Cloud Monitoring Console by selecting it from the Dashboards menu at the top of the window.
Edit your charts and dashboards by selecting the icons at the top right of the charts and dashboards. Use drag-and-drop to reorder the charts in the dashboard.
View logs associated with charts
If you notice an interesting event in one of your charts, you can go directly to the relevant project logs for more information.
Choose Settings > View Logs in the chart. Cloud Monitoring creates an advanced logs filter identifying the log entries relevant to your chart. You are then redirected to the Google Cloud Logging Logs Viewer, where the logs filter is used to display the log entries. You can edit the logs filter in the Logs Viewer to further refine the set of log entries displayed. See Advanced Logs Filters for a description of the filter language. Changing the logs filter does not affect your chart.
In some circumstances, Cloud Monitoring might be unable to find any relevant logs for your chart, in which case View Logs might not be available or you might not see any log entries in the Logs Viewer. In other circumstances, the logs filter might match too many logs, making it impractical for the Logs Viewer to display all the log entries. In that case, you should edit the filter in the Logs Viewer to reduce the number of matching logs.
Groups let you easily monitor collections of related resources. Cloud Monitoring relies on metadata such as names and tags to create groups, so you have flexibility in defining a set of rules to aggregate your resources. For example, you can create a group for "all VM instances that are running Cassandra databases," or "all VM instances behind a load-balancer instance."
Create a group
Select Add Group within the Groups section of the Cloud Monitoring Console. If you have added groups before, you can also select Groups > Create... from the top menu list.
Enter a name or phrase to identify the group.
Choose whether you want the group members to match Any or All of the rules you will specify next. If there is only one rule, it doesn't matter which match you choose.
Choose the type of value that you want to test from the dropdown list. Based on the type, fill in the condition that has to be met by members of the group.
To add another rule, select Add Criteria and repeat the previous step.
If this group is a cluster, check the box to highlight when a node's performance or configuration differs from other members in the cluster.
Select Save Group.
You can edit or delete a group by selecting Groups > Groups Overview from the top menu list and then selecting your group. In the same panel, you can create subgroups of existing groups by selecting + Subgroups.
The short video, "Alerting Policies", will get you started with alerting policies.
You can set up alerting policies to define conditions that determine whether or not your cloud services and platforms are operating normally. Cloud Monitoring provides many different kinds of metrics and health checks that you can use in the policies.
When an alerting policy's conditions are violated, an incident is created and displayed on the Cloud Monitoring Console in the Incident section. Responders can acknowledge receipt of the notification and can close the incident when it has been taken care of.
You can also associate notifications with alerting policies, so that individuals or systems are notified when something is wrong.
Create a policy
To create an alerting policy, follow these steps:
Select Create Alerting Policy within the Incident section of the Cloud Monitoring Console. You can also select Alerting > Create Alerting Policy from the top menu list.
Enter a unique policy name or phrase to identify the policy.
Fill in the Add Condition section. You are defining a condition which, if met, will trigger an alert:
- Select the kind of measurement you want in the condition: Metric Threshold, Metric Rate of Change, Metric Absence, Group Aggregate Threshold, Process Health, or Uptime Check Health. An explanation of the measurement appears in the Overview pane. Select Next.
- Fill in the Target pane with the resource type and name. Select Next.
- Fill in the Configuration pane. You will be able to choose metrics and conditions from drop-down lists.
- Select Save Condition
If you want to add another condition, select + add condition, which brings up the Add Condition pane again. You can add up to six conditions.
If you have entered two or more conditions, fill out the Policy Triggers section. The policy trigger determines if you want to trigger on on any of the conditions or only if all are met.
Fill out the Notifications section:
- Select a notification method in the drop-down list. Select Add Notification.
- Fill in the information for that notification method.
- If you want more notifications, select Add Notification to enter each one.
If you wish, fill out the Documentation section. This text will be included in your notifications. For example, you might explain what the recipient should do.
Select Save Policy.
Try out the features of Cloud Monitoring using your own projects.
If you use Compute Engine, see Monitoring Agent. Its use is optional, but encouraged.
See Available Metrics for a list of the metrics available in the Cloud Monitoring Console. See Supported Metrics for details about the Google Cloud Platform service metrics. If you want to create your own Cloud Monitoring metrics, see Custom metrics.
For more information on logging and its relation to monitoring, see Google Cloud Logging.