*Updated 5/11/20. We've added a Deployment API which allows you to investigate deployments in Opsgenie from the deployment tool of your choice, read on below for more details about the new API.
Deployment API allows you to investigate deployments from the tool of your choice
The Incident investigation view feature is powered by a connection between Opsgenie and Bitbucket that empowers teams to troubleshoot incidents related to code deployments, and quickly take corrective action from Bitbucket.
Helpful links:
Bitbucket to Opsgenie connection: https://docs.opsgenie.com/docs/bitbucket-connection
Bitbucket repo-to-service mapping: https://docs.opsgenie.com/docs/bitbucket-repository-to-service-mapping
Incident investigation view documentation: https://docs.opsgenie.com/docs/incident-investigation
How to get started
To begin using the Incident investigation view you need to connect your Bitbucket Cloud account with your Opsgenie account. Don’t have an Opsgenie account? You can start a free 14-day trial here.
In Opsgenie, navigate to Settings > Atlassian Apps. You'll see the Bitbucket logo, click it to begin.
*Please note that linking Bitbucket workspaces to an Opsgenie account can only be performed by Account admins, Owners and all Custom role users who have the appropriate rights to access the global settings page.
Then, connect your Bitbucket workspace to your Opsgenie account, click Connect Workspace.
Next, click Grant Access to allow Opsgenie to read your Bitbucket account information, access repositories, and keep both tools synchronized.
Once your workspaces are connected, navigate to the Service details page within Opsgenie to map the Bitbucket repos that are available under the connected Workspaces - click Add Mapping.
After you click “Add Mapping”, a pop-up window will appear where you can select the workspace and choose between two options for mapping repos. You can select all repos under the selected workspace to map within the service. Doing so will keep the workspace synchronized with the service and even new repos that are added later in Bitbucket will automatically map to the service on the Opsgenie side.
The second option allows you to select specific repos that exist under the selected workspace.
Click “Add mappings”, the mapped workspaces will be shown in the service details page.
Use the deployment tool of your choice
Opsgenie has added API support that allows you to use any deployment tool of your choice to perform deployments and track them within the Incident investigation view. The new deployment API allows you to investigate deployments from any tool in your stack, including but not limited to Circle CI, Jenkins Blue Ocean, Octopus Deploy, or your own in-house deployment tool of choice.
With the Deployment API, you can create deployment records inside the Incident investigation view, update the deployment status, and retrieve deployment information of previously tracked deployments.
With this deployment data now flowing into the Incident investigation view, it is much easier to explore the root causes of incidents related to deployment changes. You can also quickly determine the previous most stable deployments and perform the necessary rollback actions for quick incident resolution.
Start Investigating
Setup is complete. Now when an incident is created in Opsgenie you will see the impacted services on the incident details screen. These are the services that need to be investigated further to understand if one of their deployments or commits were a potential cause of the incident question.
Below the impacted services section is the new Potential Causes section on the incident details screen. Click on the “investigate” button to begin troubleshooting the incident and its impacted services in greater detail.
Explaining the deployment graph
At the top of the incident investigation screen is the deployment history graph. The graph shows us the deployments, and commits that have occurred for the impacted services for the incident in question in the last 24 hours. You can adjust the timeframe in the top right of the screen from 6 hours to 1 week in length.
Successful deployments are highlighted with green dots. The larger the dot the greater the number of files that were changed in a deployment. Failed deployments show in red, and previous incidents are shown with grey lines that signify the length of the incident. The current incident is highlighted with a red vertical line at the date and time the incident was created.
Click on a deployment to investigate it in more detail. The investigation view will populate the summary view below with the details of the selected deployment, including the related commits, the commit author, the total number of files changed, and the environment in which the deployment occurred. Upon hover, a “Select Deployment” button will appear, click it to highlight the deployment in question.
Click Add Potential Cause to append a particular deployment or commit to the open incident.
Back on the incident details screen, the potential cause is now shown below the impacted services section. Team members can now take action by clicking on the “deployment number”, in this example “Deployment #29” which will take you directly to Bitbucket to take corrective action on the issue.
We hope the new Incident investigation view helps you troubleshoot and resolve incidents related to code deployments more quickly. As always, we are working hard to add additional functionality and connections with other deployment tools. We look forward to your feedback in the comments below on how we can continue to create connections between Bitbucket and Opsgenie that make incident investigation and resolution more efficient for your teams.
Quick links:
Bitbucket to Opsgenie connection: https://docs.opsgenie.com/docs/bitbucket-connection
Bitbucket repo-to-service mapping: https://docs.opsgenie.com/docs/bitbucket-repository-to-service-mapping
Incident investigation view documentation: https://docs.opsgenie.com/docs/incident-investigation
Deployment API: https://docs.opsgenie.com/docs/deployment-api
Shaun Pinney
Product Marketing Manager, Jira Service Management
Atlassian
Massachusetts
4 comments