GCP Monitoring

Google Cloud Platform (GCP) is a cloud services platform offering compute engine, object storage, database storage and other services to help businesses scale and grow without worrying about underlying infrastructure.

What You Can Monitor

Opsview Monitor's GCP monitoring tools provide an easy way to monitor performance metrics for:

Compute Engine Instances
Cloud Storage Buckets
Cloud SQL Database Instances

Host Templates

The following Host Templates are provided within this Opspack. Click the name of each Host Template to be taken to the relevant information page, including a full Service Check description and usage instructions.

Cloud - GCP - Compute

Service Check Name	Description	Default Thresholds (Warning, Critical)	UOM
GCP - Compute - CPU Usage	CPU usage of the Compute instance.	cpu_usage=70,90	%
GCP - Compute - CPU Reserved Cores	Number of CPU Reserved Cores.	N/A	N/A
GCP - Compute - CPU Usage Time	CPU Usage Time.	N/A	s
GCP - Compute - Disk Bytes	Bytes read and written from disk.	N/A	B
GCP - Compute - Throttled Disk Bytes	Percentage of throttled bytes read and written from disk.	N/A	%
GCP - Compute - Disk Operations	Read and write disk operations.	N/A	N/A
GCP - Compute - Throttled Disk Operations	Percentage of throttled read and write disk operations.	N/A	%
GCP - Compute - Instance Availability	Instance uptime percentage over the period checked.	availability=100:,80:	%
GCP - Compute - Network Bytes	Received and sent network bytes for the instance.	N/A	B
GCP - Compute - Network Packets	Received and sent network packets for the instance.	N/A	N/A
GCP - Compute - Firewall Dropped Bytes	Bytes dropped by the firewall for the instance.	N/A	B
GCP - Compute - Firewall Dropped Packets	Packets dropped by the firewall for the instance.	N/A	N/A

Service Check Name	Description	Default Thresholds (Warning, Critical)	UOM
GCP - Storage - Network Bytes	Bytes received and sent by the bucket.	N/A	B
GCP - Storage - API Requests	Number of API requests to the bucket.	N/A	N/A
GCP - Storage - Bucket Object Count	Total number of objects in the bucket. This value is measured once per day by Stackdriver, so changes will take 24hrs to appear.	N/A	N/A
GCP - Storage - Bucket Bytes	Total bytes in the bucket. This value is measured once per day by Stackdriver, so changes will take 24hrs to appear.	N/A	B
GCP - Storage - Bucket Byte Seconds	Total daily storage used by the bucket in byte-seconds. This value is measured once per day by Stackdriver, so changes will take 24hrs to appear.	N/A	Bs

Service Check Name	Description	Default Thresholds (Warning, Critical)	UOM
GCP - MySQL - Network Bytes	Network received and sent bytes for the MySQL instance.	N/A	B
GCP - MySQL - Disk Operations	Disk read and write operations for the MySQL instance.	N/A	N/A
GCP - MySQL - Database Availability	Database uptime percentage over the period checked for the MySQL instance.	availability=100:,80:	%
GCP - MySQL - Status	Current serving state of the MySQL instance.	N/A	N/A
GCP - MySQL - Memory Usage	Memory used by the MySQL instance.	memory_usage=70,90	%
GCP - MySQL - Disk Usage	Disk used by the MySQL instance.	disk_usage=70,90	%
GCP - MySQL - CPU Usage	CPU used by the MySQL instance.	cpu_usage=70,90	%
GCP - MySQL - Queries	Number of received and executed MySQL queries on the MySQL instance.	N/A	N/A
GCP - MySQL - Bytes	Number of bytes received and sent by the MySQL instance.	N/A	B
GCP - MySQL - Network Connections	Number of network connections to the MySQL instance.	N/A	N/A

Service Check Name	Description	Default Thresholds (Warning, Critical)	UOM
GCP - PostgreSQL - Network Bytes	Network received and sent bytes for the PostgreSQL instance.	N/A	B
GCP - PostgreSQL - Disk Operations	Disk read and write operations for the PostgreSQL instance.	N/A	N/A
GCP - PostgreSQL - Database Availability	Database uptime percentage over the period checked for the PostgreSQL instance.	availability=100:,80:	%
GCP - PostgreSQL - Status	Current serving state of the PostgreSQL instance.	N/A	N/A
GCP - PostgreSQL - Memory Usage	Memory used by the PostgreSQL instance.	memory_usage=70,90	%
GCP - PostgreSQL - Disk Usage	Disk used by the PostgreSQL instance.	disk_usage=70,90	%
GCP - PostgreSQL - CPU Usage	CPU used by the PostgreSQL instance.	cpu_usage=70,90	%
GCP - PostgreSQL - Transaction Count	Number of PostgreSQL transactions.	N/A	N/A
GCP - PostgreSQL - Replication Lag Bytes	The Replication Lag in bytes for the PostgreSQL instance.	N/A	B
GCP - PostgreSQL - Network Connections	Number of network connections to the PostgreSQL instance.	N/A	N/A

GCP Monitoring Prerequisites

Opsview Monitor Version 6.7.0 or newer

GCP Monitoring Setup

To monitor your GCP environment, you will need to: - Enable the Google Stackdriver Monitoring API. - Retrieve the Service Account Credentials. - Add the GCP project to Stackdriver Monitoring.

Enable the Google Stackdriver Monitoring API

Step 1:

Log into your GCP console: https://console.cloud.google.com/

Step 2:

Make sure you have selected the proper project in the top bar - next to the Google Cloud Platform logo:

From the menu on the left side of the screen select APIs & Services and Dashboard

Step 3:

Click the Enable APIs and Services button.

Step 4:

Using the search box, look for monitoring and from the list select Stackdriver Monitoring API

Step 5:

Click the Enable API button. It may take some time until API will be turned on.

Retrieve the Service Account Credentials

Step 1:

From the left menu choose IAM & admin and Service accounts, you will see the service account dashboard.

Step 2:

Click the Create Service Account button.

Step 3:

Populate the fields: Service account name, Service account ID and Service account description, and click the Create button to move on to the next step.

Step 4:

Click the Select a role field, and in the filter type monitoring viewer. From the available options select Monitoring Viewer and click the Continue button to move on to the next step.

Step 5:

Click the Create key button.

Step 6:

On the right side of the screen you will see a tab in which you should set the type of key to JSON, then click Create.

After you click Create you will be asked to save the file with credentials. Once done, you should then see a confirmation message:

Click Close and then click Done to finish creating your Service Account.

Add the GCP project to Stackdriver Monitoring

If you are using Stackdriver for the very first time, you will be asked to create your first workspace. With workspaces you can group your GCP projects into monitoring workspaces.

Step 1:

From the dropdown next to the Stackdriver logo, select Create Workspace.

Step 2:

Click within Google Cloud Platform project field, select the project that you want to monitor within that workspace and click the Create workspace button.

Step 3:

In the next screen, you can add additional GCP projects to the Stackdriver workspace. Once you have selected the desired projects, click Continue.

Step 4:

In the following screens you can: - Add your AWS projects to the Stackdriver workspace. If you don't want to add your AWS projects, simply click the Skip AWS Setup button. - Setup Stackdriver agents on your VMs. You can skip this step by clicking the Continue button. - Setup reports by email - choose the desired frequency and click the Continue button.

After you've performed all steps, you should see a Finished initial collection! notification:

You can then click the Launch monitoring button to complete your Stackdriver setup.

Importing this Opspack

Download the cloud-gcp.opspack file from the Releases section of this repository, and import it into your Opsview Monitor instance. Now you can add the Host Templates you want following the info links in the table at the top.

For more information, refer to Opsview Knowlege Center - Importing an Opspack.