Atlassian wants to make it easier for IT teams to review of how they responded to major software outages with the addition of a postmortem feature for its Jira Ops incident management tool.
JiraOps was unveiled at the vendor’s user conference in Barcelona in September; it’s essentially a version of the Jira issue tracking tool tailored for use by IT operations teams.
Atlassian describes Jira Ops as a “single place to respond to incidents, connect incident point solutions, and execute a consistent process for every incident.”
Since the September announcement, Atlassian has added a number of functions, including those to streamline the creation of postmortem documentation – a process often overlooked after an incident's been resolved, according to the Australian software firm.
To encourage this analysis, Jira Ops users are prompted to create a postmortem document in Atlassian’s content collaboration tool, Confluence. Also provided are templates that contain fields for key information such as a summary of what happened, the root cause and lessons learned.
In addition, an event timeline created in JiraOps during incident response is automatically posted into the postmortem document. The timeline includes important events such as status changes, Slack messages and posts on Atlassian’s incident communications tool, Statuspage.
Collating this information manually could take hours, Atlassian said.
“Bringing all that [information] into Confluence for the postmortem saves a lot of time for a process that can otherwise be quite time-consuming and difficult to manage,” said Danny Olinsky, head of product marketing for Atlassian incident management solutions and co-founder of Statuspage.
After a postmortem analysis has been carried out and documented, Jira Software issues can be kicked-off from Confluence to “track action items and improve for the next time around,” said Olinksy.
“This tool is consistent with Atlassian’s strategy of building a lightweight, yet comprehensive, digital pipeline, from source control through operations including service management,” said Charles Betz, lead DevOps analyst at Forrester.
By extracting relevant documentation from other systems and pushing these findings into product backlogs, the postmortem feature “improves both efficiency and confidence in the overall delivery process,” Betz said.
Olinsky added that a simplified postmortem process will encourage more businesses to communicate more openly about system outages that affect customers. This includes explaining what was done to remediate a systems failure as well as the cause.
“We think that if we can surface [incident response information] easier and better for teams, then they will take advantage of it and do more postmortems," Olinsky said.
"We expect that would lead to more public communication…, so they are not just telling customers when there is an incident, but also doing the follow-up after to say, ‘Here's what happened, here's what we are doing about.’"
JiraOps is currently available for early access customers and will be rolled out fully in the first half of 2019.
In a related announcement, Atlassian unveiled automated actions for OpsGenie, the incident alerting tool acquired in September.
The new feature enables integration with AWS Systems Manager (the management console for AWS cloud resources such as EC2 instances), automating certain responses to a software outage, such as restarting a server.