When Daysha recognized that DevOps was an inevitable outcome for organisations that need to be agile, our customers explicitly stated they wanted solutions not process preaching. We therefore entered into partnership with technology providers where our services complemented their technology to help our clients become agile. This blog focuses on workflow and Atlassian tooling to overcome the silo’ing of data between dev and ops.
When we decided to partner with Atlassian,we also selected Chef and AppDynamics all of whom helps underpin a continuous delivery process. Chef enables automations and supports the 2nd way. AppD shortens and amplifies the feedback loop from right to left which is also supportive of the 2nd way.. The 1st of the ‘three ways‘ of DevOps is that you must smooth the flow of work left to right and Atlassian tooling does this very elegantly.
It’s impossible to be functional in a DevOps sense unless you are sharing artifacts (user stories, code, configurations, test cases, bug reports etc.) from product owner to dev to test to production support.
When you are wrestling with email trails and versions of files that are out of date the second they are emailed its not easy to be on the same page as colleagues. We often recommend no tools at all when you are starting out on an Agile journey. But once you are deploying code at velocity you need tools that smooth the flow of data bi-directionally.
The effort required to create or find an artifact has to be zero if everyone in the workflow is to be collaborative. If a DevOps engineer has to stop what they are doing to create an artifact it wont happen. The artifact has to be generated as a consequence of adhering to a process and easily accessed by all.
Sharing data can also be a state of mind and we offer training as a means to help dev and ops work more closely.This training is about changing attitudes and behaviour. The third way relates to a reality that often its organisation structures that can impede the smooth passage of work through a software development lifecycle (SDLC) process. We speak to this topic in other blogs.
When we engage teams that want to move towards Continuous Delivery, many of the functions have their own silo’d tools and don’t see value in sharing data. It can often be expensive to buy tools for people who are only viewing. This approach from your tools vendor would indicate if DevOps is a marketing concept or a commercial thumbs up in the form of a pricing policy.
Atlassian understands this key concept and provides links between artifacts but each user in the workflow has a unique pane on his or her primary artifact and a link to colleagues work that relates to the specific instance of code in BitBucket (GIT for grown ups) task in Jira, the textual description in a Confluence space, or the test case in any modern test tool such as Selenium.
Teams that share data between dev and ops produce better quality code faster. In turn this lowers the cost and risk of missing a window of opportunity in e-commerce or consumer facing applications.
Atlassian offers its tools for $10 for the first 10 users and it donates the $10 to charity. They operate a try and buy model and we support implementation rollout of their products.
POSTED IN: DevOps tools