Blog Planview

Votre parcours vers l’agilité métier

Produits & Outils

Atlassian’s Jira Cloud API Change – All you need to know (and how Value Stream Integration means you don’t need to worry)

Publié le Par Patrick Anderson
Atlassian’s Jira Cloud API Change – All you need to know (and how Value Stream Integration means you don’t need to worry)

Software tool vendors, such as Atlassian, can amend their APIs at any moment for multiple reasons. As business needs change rapidly and unexpectedly, developers must integrate new application components into existing architecture to ensure the software is as innovative, functional and responsive as possible for their customers.

What this means is that the tools that your specialist teams use and collaborate with to plan, build and deliver software are constantly vulnerable to disruption. The flow of work across your value streams from ideation and operation can go down overnight, bringing your operations and customers’ products and services down in a heartbeat. In a fiercely competitive digital climate, downtime is everything. So how do we mitigate this disruption?

Atlassian’s latest change to Jira Cloud is a fine example of the myriad issues that API changes can cause. As one of the most popular agile planning tools on the market, Atlassian Jira’s latest API change – while improving the product and ensuring its compliance with GDPR – is likely to impact the software delivery value streams of many organizations across many industries. Here, we answer your questions on the impact of API changes and how model-based integration can ensure your software products remain up and running.

What is it happening with Jira Cloud?

Throughout 2018 and 2019, Atlassian has been undertaking a number of changes to their products and APIs to improve user privacy in accordance with GDPR (European General Data Protection Regulation).  Currently, we know that they will be rolling out changes to Atlassian Cloud product APIs to consolidate how personal data about Atlassian product users is accessed by API consumers. The main change centers around user identification, with “username” changing to a user “account id”. You can read more about these changes on the Atlassian website.

It’s important to note that Zephyr Cloud will also be affected by these changes, but the company has yet to confirm the date that its corresponding updates will go live.  For this reason, Zephyr users should not upgrade their Tasktop instance at this time. Instead, we are asking our customers with Zephyr Cloud integrations to contact us at their earliest convenience to discuss the right time to upgrade.

How does this change affect me?

Atlassian’s change to the Jira Cloud username field would cause any existing integrations between Jira Cloud and other tools to stop working when retrieving an artifact (such as story or defect). However, through close coordination with Atlassian, Tasktop has released updated versions of our products, Tasktop Integration Hub and Tasktop Sync, to mitigate these changes and ensure integrations continue to run uninterrupted.

Without Tasktop, this toolchain fragmentation disrupts the flow of product-critical data (such as feature, epic, story, defect etc.) between teams, and undermines the data integrity between the tools. With no one source of truth, organizations have poor end-to-end visibility and traceability across their value stream. This is a big problem for identifying bottlenecks and optimization opportunities across process and causes compliance issues in heavily-regulated industries such as finance, insurance, government and healthcare.

When do these changes take place?

Atlassian has stated that the deprecation period for this change started as of October 31 and will be completed by April 29.

How does Tasktop help?

Tasktop is always working closely with our wide network of tool partners to track upcoming changes to ensure that our integrations between tools are robust and unaffected by API changes. Through our constant dialogue with leading tool vendors and our exclusive access to their APIs, we can ensure our connectors between tools are always working. Or in other words, our large-scale integrations are robust, reliable and something you never have to worry about.

How is Tasktop’s approach different to other solutions on the market?

Tasktop’s approach to Value Stream Integration is unique for three key reasons:

1. We integrate through models

A model is a shared definition of a unit of work for the purpose of reporting and measurement that extracts the most important data in each system of record and distills it into a standard shape and form, namely a fixed set of fields and possible values. Models standardize the information exchanged by integrations into archetypes of business value, which express the organizational consensus of the data package that needs to flow between tools and be measured by stakeholders. Models are defined upfront and in advance on the integration platform to prevent ad-hoc, subjective or arbitrary data selection and field mapping. As data flows between two repositories, it goes through the model. A good analogy to help understand how models work is the act of translating between people who speak different languages. You can learn more on model-based integration in the below blog, with links to an infographic and white paper for a deeper dive.

https://blog.tasktop.com/blog/model-based-integration/

2. The Integration Factory

A key ingredient to Tasktop’s secret sauce is our fabled Integration Factory, a unique testing infrastructure. Maintaining integrations is a moving target that can quickly become a huge resource drain for your organization – that’s why we created the Integration Factory.

The Integration Factory is a set of technologies, an approach, a methodology, a repeatable process for creating robust, high quality integrations/connectors. It is a proprietary knowledge database on how all tools can be integrated across your entire value stream. We run tests against running instances of the tools with testing methods that are so exhaustive that we’ve found errors in new versions of endpoint products before their vendors have.

The Test Infrastructure performs over half a million tests per day keeping customers current with all new releases, while maintaining the integrity of existing tool integrations. What this means is that multiple versions of the same tool can be connected via Tasktop and that the frequently updated APIs of SaaS and On Prem tools will be accounted for without breaking your integration, since Tasktop tests all needed version combinations. We support 50+ leading tools and 350+ versions thereof.

Rapid, high quality continuous software delivery at scale hinges on the stability and speed of multiple integrations, which in turn rely on powerful 24/7 automated testing. You can attempt to create your own DIY test factory in house, but it would take at least a decade (if not much longer), and a lot of money, to build up the knowledge and expertise that Tasktop has invested into the system.

Moreover, our well of knowledge into vendor tools has been gradually garnered over the years through our symbiotic relationships with tool vendors to acquire a deep understanding of their products. The factory also requires round-the-clock maintenance from people who are trained in integration, something your IT teams may not have the time, nor desire to do, with their own jobs to focus on.

3. Our Domain Knowledge

APIs are the only way to integrate between tools. Through APIs, we gain access to a tool’s capabilities and artifact structure, allowing us to flow data from one tool to another. APIs, however, are only part of the answer. APIs are historically poorly documented and incomplete:

  • Data structures, how and when calls can be made (for statement systems), and the side effects of operations are often all excluded from documentation
  • Poor error handling and error messages are common
  • Edge cases and bugs are rarely documented
  • Lack of documentation means there’s a great deal of trial and error in figuring out how to handle these issues
  • Vendor customer support staff are often unaware of these issues and how to use their API, so finding resolution requires access to vendor’s development team

To manage this complexity requires expert understanding of the tools being integrated and how they’re used, as well as ‘domain understanding’ – i.e., adding a layer of domain expertise on how to structure the data from the practitioners who will be involved and benefit from the integration. Tasktop invests a lot of resources into the research that we bake into the Integration Factory. 

Click image to download e-book.

What do I do now?

If you’re a Tasktop customer using Jira Cloud or Zephyr Cloud, all you need to do is follow these instructions by April 29th.  Tasktop admins can access the upgrade through their my.tasktop.com account or by contacting Tasktop Support. Our product documentation provides detailed instructions on upgrading and updating extension scripts related to user identification.

If you’re not a Tasktop customer, chat to us today to see how Value Stream Integration can ensure that your software delivery value streams are never interrupted by API changes, evolving workflow, new tools or tool versions and more.

Articles similaires

Rédaction du contenu Patrick Anderson

Patrick est directeur principal du contenu chez Tasktop et supervise les programmes de contenu et de leadership éclairé de l'entreprise. En dehors du bureau, vous le trouverez en train de lire, d'écrire, de taper quelques basses (mal), de divaguer dans la nature et de suivre son équipe de football (soccer) anglaise, West Ham United.