Google Cloud Platform (GCP) plugin

For more information about what this plugin does and the data streams it retrieves, see:

Google Cloud Platform

Monitor your GCP environment, including GKE, Hosts and more.

To add a data source click on the + next to Data Sources on the left-hand menu in SquaredUp. Search for the data source and click on it to open the Configure data source page.

You can also add a data source by clicking Add data source on the Settings > Data Sources page, but pre-built dashboards are not added when using this method.

Before you start

In the GCP console select your project and check that the following are enabled:

To view cost information in SquaredUp using the Cost data stream check the following:

GCP Service Account Configuration

  1. Create a new Service Account, or edit an existing account.
    See GCP - Creating and managing service accounts
  2. Ensure the account has the role Viewer by adding the role Basic > Viewer.
    To view cost information, the account also needs to have BigQuery Data Viewer role.
    See GCP - Grant a single role
  3. Create a new key for the Service Account using the key type JSON. Download the JSON file as you will need to copy information from this JSON file when adding the data source next.
    See GCP - Creating service account keys

    Make sure to store the key file securely, because it can be used to authenticate as your service account.

Configuring the data source

  1. Open the JSON file that you downloaded when creating the key.
  2. Copy and paste the clientEmail from the JSON file into the data source form.
  3. Copy and paste the private_key from the JSON file into the data source form (everything between the quotes).
  4. Copy and paste the project_ids from the JSON file into the data source form.
  5. Optionally, to use the Cost data stream, copy and paste in the billingProjectId, billingDataSetName and billingTableName.

  6. Restrict access to this data source:
    You can enable this option if you only want certain users or groups to have access to the data source, or the permission to link it to new workspaces. See data source access control for more information.

  7. Click Test and add to validate the data source configuration. SquaredUp will now attempt to connect to SquaredUp using the provided authentication method.

Next steps

Data streams

The following data streams are installed with this plugin.

Data stream
Description
Builds
Builds for the selected trigger.
Cost
Cost for the selected objects
CPU Number
Number of vCPUs reserved on the host of the instance
CPU Usage/sec
Delta vCPU usage for all vCPUs, in vCPU-seconds
CPU Utilization
CPU utilization of selected host
Disk Max Read Bytes/sec
Disk's maximum per-second read throughput over a period of time
Disk Max Read Operations/sec
Disk's maximum per-second read requests count over a period of time
Disk Max Write Bytes/sec
Disk's maximum per-second write throughput over a period of time
Disk Max Write Operations/sec
Disk's maximum per-second write throughput over a period of time
Disk Status
Status for the selected disks
Disk Total Read Bytes Count
Count of bytes read from disk
Disk Total Read Operations Count
Count of disk read IO operations
Disk Total Write Bytes Count
Count of bytes written to disk
Disk Total Write Operations Count
Count of disk write IO operations
Execution times
Distribution of workflow execution times
Finished execution count
Count of finished executions for the workflow
GKE Status
GKE status for the selected resources
Host Status
Status for the selected hosts
Instance Total Uptime
Elapsed time since the VM was started, in seconds
Instance Uptime
Delta of how long the VM has been running, in seconds
Memory Total RAM
The total amount of memory in the VM
Memory Used RAM
Memory currently used in the VM
Network Total In
Count of bytes received from the network
Network Total Out
Count of bytes sent over the network
Oldest unacked message age
Age (in seconds) of the oldest unacknowledged message (a.k.a. backlog message) in a subscription
Publish message size
Distribution of publish message sizes (in bytes)
Publish requests
Cumulative count of publish requests
SQL Instance Status
Status for the selected SQL instances
Unacked messages
Number of unacknowledged messages (a.k.a. backlog messages) in a subscription.
Workflow Executions (most recent)
Most recent executions for the selected workflow

Configurable data streams

The following data streams have configurable Parameters.

Data stream
Description
Parameters
BigQuery
Returns the result of a BigQuery query.
  1. Query:
    Enter a BigQuery query.
  2. Project ID:
    Specify the project that you would like to run the BigQuery against.
Firestore Documents
Firestore databases can be queried for document data by specifying a Document Path
  1. Document Path:
    Specify the Document Path to a document or a collection. The path should be relative to the database (i.e. not include the project or database name). For example mycollection or mycollection/mydoc.
Firestore Structured Query
Firestore databases can be queried for document data by specifying a Firestore query.
  1. Structured Query:
    Enter the structured query JSON (See Firestore StructuredQuery schema for detailed information). For example: 
    {
    "from": {
        "collectionId": "users"
    },
    "where": {
        "fieldFilter": {
            "field": {
                "fieldPath": "firstName"
            },
            "op": "EQUAL",
            "value": {
                "stringValue": "Jo"
            }
        }
    }
}
Monitor Metric
Gets data for a specified monitor metric.
  1. Metric Name:
    Select the desired metric from the dropdown. The dropdown options will automatically update depending on the object that you have scoped to. For example, scope to a Host and select CPU usage from the Metric Name dropdown.
MQL Query
The GCP data source allows you to set up configurable data streams for any kind of query for Monitoring Query Language (MQL).
You are able to query any object from your Google Cloud, even if they are not indexed in the SquaredUp graph.
See Google Cloud: Using Monitoring Query Language for more information about MQL.
  1. Project:
    Select the project that you would like to run the MQL against from the dropdown.

    You will not need to select a project if you have already selected one in the scope.

  2. MQL Query:
    Enter a MQL query.

    Mustache parameters are only supported if you have scoped to a GCP object and selected the + MQL Query. Mustache parameters are not supported if you have scoped to a project or the data source instance itself.

    See MQL query examples below for more information about queries. Supports mustache parameters supplied in an array

    This data stream supplies scoped objects in an array for mustache parameters. When there are multiple objects in scope this data source will send the query once with all the objects in an array.

    When the scoped objects are supplied in an array the normal mustache syntax, for example {{name}}, must be contained between {{#.}} and {{/.}} (the full-stop indicates that the whole object should be used, in this case the array of objects in scope).

    For example, a query where clause might look like:

    | where ComputerName in (  {{#.}}    '{{name}}',    {{/.}} '' )

    • The {{#.}} and {{/.}} indicate that what is contained within is expanded for each element in the array of objects.
    • You can use properties of objects and write them in between curly braces e.g {{name}} to use them as mustache parameters. For example, if objects of type "host" have a property called name, you can use {{name}}. This will resolve {{name}} to the value of the name property of the different "host" objects used in the scope.
    • '{{name}}', means that the name property is expanded inside single-quotes with a trailing comma.
    • The trailing single quotes '' are necessary to stop the query being rejected because a trailing comma is disallowed.
    • Whenever you use mustache parameters, you need to use a scope of objects that contain the property you're referencing.
  3. Automatically apply dashboard timeframe:
    By default, Automatically apply dashboard timeframe will be selected.

    If you do not wish to use a dashboard timeframe, you can set the timeframe in the MQL query by adding within to the query. If you deselect Automatically apply dashboard timeframe and do not specify a timeframe in the MQL query, the query will fail. To add a timeframe to a query, simply type | within (timeframe) at the end of your query. For example, to add a timeframe of 24 hours you enter: | within (24h)

MQL query examples

MQL query not using dashboard timeframe - CPU usage time in the last 20 minutes in the selected project:

fetch gce_instance | metric 'compute.googleapis.com/instance/cpu/usage_time' | within (20m)

MQL query not using dashboard timeframe - authentication events count in the last 24 hours:

fetch iam_service_account
| metric 'iam.googleapis.com/service_account/authn_events_count'
| align rate(1m)
| every 1m | within (24h)

MQL query using dashboard timeframe - uploaded bytes in BigQuery data set:

fetch bigquery_dataset
| metric 'bigquery.googleapis.com/storage/uploaded_bytes'
| align rate(1m)
| every 1m

MQL query scoped to Hosts and using mustache syntax - CPU usage time using dashboard timeframe:

fetch gce_instance
            | metric 'compute.googleapis.com/instance/cpu/usage_time'
            | filter (metadata.system_labels.name == '{{#.}}{{name}}{{/.}}')
            | align rate(1m)
            | every 1m

Tips for using the GCP Cost data stream

The Cost data stream is part of the GCP data source and uses the GCP BigQuery API. The GCP data source needs some additional configuration to enable the Cost data stream see How to add a Google Cloud Platform data source

  • The GCP Cost data stream shows daily data.
  • By default the Cost data stream splits costs by service.
  • We recommend scopes used with the GCP Cost data stream contain only one type of object, for example, only accounts, projects or hosts. Mixed scopes may include duplicated data in the results, because a query is sent for each type of object in the scope, and the results are combined. For example, resulting data would show the cost of a host, as well as the cost of a project, even if the host is part of that project. It is useful to create a scope for your account(s), a scope of project(s) and a scope for hosts.

Use Cases for the GCP Cost data stream

Was this article helpful?

Have more questions or facing an issue?
Submit a ticket