Answering your most frequently asked questions about Bitbucket 7.0 (part 1)

[#1]

Hello Community 👋

A few weeks ago, we team hosted a webinar called What’s New in Bitbucket 7.0. Many webinar attendees asked great questions, and we thought it would be beneficial to share our answers to the most frequently asked questions below:

Will there be an Enterprise release for the 7.x series?

Yes! We plan to have another Enterprise release sometime later this year. We will likely announce the exact version in June/July of this year.

For those that are new to Enterprise releases (ERs), an ER receives additional backported bug fixes and security updates from later feature releases throughout its two-year support window. If you manage a large, complex deployment that can only be upgraded once a year, we recommend targeting an ER. If your team has the flexibility to upgrade more frequently, we suggest doing so as often as your business allows.

What is an upgrade path? We are still 5.13.0.  Do we need an incremental upgrade first before 7?

It is possible to upgrade from 5.13.0 directly to 7.x.

However, it is important to check the supported platforms document to ensure your database and installed Git version is supported by Bitbucket 7.x first. You may need to upgrade either or both before making the switch.

Any ability to categorize tasks to collect peer review defect metrics? 

We don’t categorize tasks in pull requests.

If you think task categories can be useful for your workflow, please vote for the relevant feature suggestion. Don’t forget to leave a comment describing your use-case. If the suggestion doesn’t currently exist, feel free to create a new one. This will help us to make better feature prioritization in the future. Feature suggestions

Did I see a "Create Jira Issue" option?

Yes, for better tracking items of work you can easily create Jira issue from comments and tasks by clicking on “…” and then "Create Jira Issue". This will open issue creation dialog in Bitbucket, so you don’t have to switch tools.

Is the file limit removed in large PRs with the wild-card feature?

We have the same limitation for files-tree filtering.

Is there any ability to review document files (Word, Excel, PDF, etc)?

We don't support rendering any binary files other than images inside the pull request.

If you are interested in reviewing specific types of binary files, which are not supported at the moment, find the relevant suggestion and vote for it. Don’t forget to leave a comment describing your use-case. If the suggestion doesn’t currently exist, feel free to create a new one. This will help us to make better feature prioritization in the future. Feature suggestions

Does the search feature on the PR search only the changed code or any code in the codebase?

The search feature on the pull request page works only on files from that pull request.

Can you explain the drawbacks of removing 3-way merges from PRs? Is this configurable?

Bitbucket will still indicate if a pull request has conflicts, but it will no longer show conflicts in the diff itself.

3-way diffs not only imposed significant CPU, memory and I/O load on Bitbucket Server instance, but it also introduced major complexities that impacted other functionalities we developed.

We understand that some of you were making significant use of the 3-way diffs, so this decision was not taken lightly. We decided to switch to 2-way diff starting from Bitbucket 7.0. It is not a configurable setting. However, we believe that this change will benefit the great majority of you in the long run.

Are there any plans on more granular seat counts for BB Datacenter. We are close to needing more than 100 licenses, but the next seat count is 250 at a very large cost jump.

We don’t have plans to change Bitbucket DC license levels in the mid-term. However, it doesn’t mean that current license levels won’t ever change.

Do smart mirrors have git integration, e.g. sub-commands helping in setting up the different push/fetch URLs?

Smart Mirrors do have git integration. In fact, their primary goal is to clone git repositories from primary Bitbucket Server to off-load git requests. Regarding push and fetch URLs, mirrors have a different clone URLs for any given repository which cannot be changed but clients can setup different URLs for push and fetch, e.g. push URL can be set to primary server and fetch URL to Smart Mirror.

Is there an ability to download a specific file locally from a pull request, rather than cloning an entire branch?

Currently, there is no way of downloading a file directly from the pull request, but you can try a workaround:

Select the file in the file tree, then click on the file icon in the top right corner of the file header:

You will be navigated to the file view where you can click on the “Raw file” link to download the file:

If you are interested in downloading files directly from the pull request, find relevant suggestion and vote for it. Or you can create a new feature suggestion. Don’t forget to leave a comment describing your use-case. If the suggestion doesn’t currently exist, feel free to create a new one. This will help us to make better feature prioritization in the future. Feature suggestions

Are there plans to add functionality to allow users to download only the files that changed in a pull request?

You can try the workaround described in the previous answers, or retrieve a raw patch or raw diff for a set of commits or pull request via REST API.

You can get the raw diff of a pull request by adding “.diff” after the id in the REST request, like this:

http://…./rest/api/latest/projects/<PROJECT_KEY>/repos/<REPO_SLUG>/pull-requests/<PR_ID>.diff

Related to that you can get a “git patch” of a pull request if you add “.patch” after the id, like this:

http://…./rest/api/latest/projects/<PROJECT_KEY>/repos/<REPO_SLUG>/pull-requests/<PR_ID>.patch

If you are interested in the feature, find relevant suggestion and vote for it. Or you can create a new feature suggestion. Don’t forget to leave a comment describing your use-case. If the suggestion doesn’t currently exist, feel free to create a new one. This will help us to make better feature prioritization in the future. Feature suggestions

Is Bitbucket gonna be optimized for EFS at some point so that it's a multi-az NFS implementation?

AWS EFS, while suitable for some workloads, due to its extremely high latency for file operations such as open/stat/create is totally unsuitable for hosting Git repositories where performance is required. We do not believe this will change in the near future.

We are exploring other approaches to achieving a multi availability zone deployments in AWS.

Any improvements to "Default Approvers"? Would it be easier to manage with Groups?

Indeed, groups of default reviewers would be helpful when creating a pull request.

If you are interested in the Default approvers, vote for the feature. Don’t forget to leave a comment describing your use-case. This will help us to make better feature prioritization in the future.

https://jira.atlassian.com/browse/BSERV-2870

Are there plans to implement Zero Downtime Upgrades like Jira Data Center?

We hear this request quite often for Bitbucket DC. At the moment, we are exploring ways to enable rolling upgrades for the product.

If you are interested in ZDU, vote for the feature. Don’t forget to leave a comment describing your use-case and if you are willing to be contacted for a customer interview.

https://jira.atlassian.com/browse/BSERV-7894

Will there be support for deploying in a containerized environment like Kubernetes?

For Bitbucket we don’t have official support for Kubernetes. However, you can find official docker image in Docker hub.

If you are interested in the official support for Kubernetes, vote for the feature. Don’t forget to leave a comment describing your use-case. This will help us to make better feature prioritization in the future.

https://jira.atlassian.com/browse/BSERV-10913

As an admin, what gotchas should I expect when upgrading to 7.xx - what should I look out for in my testing prior to rolling it out live in my organization?

The upgrade path to Bitbucket 7.0 is the same as for any other release. Make sure Bitbucket runs on the supported platform (OS, DB, Java). Then upgrade on a staging environment to do general testing and pay extra attention to custom apps, if you have any. If everything is fine you can safely upgrade your production environment.

We have though Bitbucket Server upgrade guide for your convenience.

Is there a feature to see all pull requests across all repositories?

If you are looking for your pull requests across multiple repositories, they can be easily found on “Your work" page.

However, if you are looking for all pull requests in Bitbucket, there is no such page at the moment. However, you can use REST API to get the list of all pull requests easily.

You can get the list of all projects like this:

http://…./rest/api/1.0/projects

Then, you can get the list of repositories:

http://…/rest/api/1.0/projects/<PROJECT_KEY>/repos

Finally, the list of pull requests per repository:

http://…/rest/api/1.0/projects/<PROJECT_KEY>/repos/<REPO_SLUG>/pull-requests

If you are interested in the feature, find relevant suggestion and vote for it. Or you can create a new feature suggestion. Don’t forget to leave a comment describing your use-case. If the suggestion doesn’t currently exist, feel free to create a new one. This will help us to make better feature prioritization in the future. Feature suggestions

---

Part 1 - Answering your most frequently asked questions about Bitbucket 7.0 (part 1) 

Part 2 - Answering your most frequently asked questions about Bitbucket 7.0 (part 2)