5 tips to help you keep your Bitbucket and Bamboo instances available

[#1]

Whether you’re managing your code with Bitbucket or building and deploying with Bamboo, your team relies on uptime and availability to get the job done and remain on track to meet your goals. That’s why we’ve built, and are continuing to build, features into our developer tools to increase and optimize high availability. Downtime not only keeps your team from accessing code but can also keep you from accessing other business-critical digital assets if you use Git to store them.

Read on to discover 5 tips to help you keep your Bitbucket and Bamboo instances available.

1. Cluster your Bitbucket instances

2. Install Bitbucket Mesh

3. Leverage Smart Mirroring

4. Take advantage of cold standby

5. Plan for your Bamboo 9 upgrade

Tip 1: Cluster your Bitbucket instances

One of the advantages of Data Center is the option to deploy on a single node or a cluster. If you are running your Bitbucket instance on a single node, consider clustering as the best way to gain high availability. We designed clustering for enterprises with large or mission-critical Data Center deployments that require continuous uptime, instant scalability, and performance under high load. With a clustered deployment, if one node in your cluster goes down, the others take on the load, ensuring your users have uninterrupted access to Bitbucket. Learn more about Clustering with Bitbucket Data Center.

Tip 2: Install Bitbucket Mesh

Bitbucket Mesh, released as part of Bitbucket 8.0, is a distributed, replicated, and horizontally scalable Git repository storage system. We designed Mesh to increase performance and improve the resilience of Bitbucket while making high availability more widely available for Bitbucket Data Center. The below diagram shows how Mesh ensures your Git is no longer dependent on the shared file system which previously could become a single point of failure, making your instance more resilient.

When you migrate your repositories to Mesh, they are replicated to multiple Mesh nodes. This ensures that in the case of a loss of a Mesh node, others will immediately cover it, keeping repositories available for end-users as if nothing happened. When the Mesh node comes back online, it will rejoin the pool, giving you peace of mind. Learn more about Bitbucket Mesh.

Tip 3: Leverage Smart Mirroring

If you’ve experienced availability issues, we recommend taking advantage of Mirrors and Mirror Farms for your Bitbucket instance. Smart Mirroring allows users to set up live Mirrors that host replicas of repositories in remote locations. These Mirrors automatically keep any repository hosted on them in sync with the primary Bitbucket Data Center instance. If a primary Bitbucket instance goes down, the Mirror likely will not, keeping your repositories available for read operations.

If you want to go even further and improve the high availability of Mirrors, then Mirror Farms is your choice. Not only they can help to offload heavy CI/CD load from the primary Bitbucket Data Center, but they also have much higher resilience compared to a single Mirror. Mirror Farms increase CI/CD capacity by allowing users to cluster Mirrors into “Farms” grouped behind a load balancer. Learn more about Mirror Farms.

Looking for the best in high availability for Bitbucket? Be sure to download the latest version to take advantage of performance improvements and Bitbucket Mesh!

Tip 4: Take advantage of cold standby

We built Bamboo Data Center to include build resiliency and cold standby high availability, ensuring smoother recovery from an outage. With Bamboo’s server instances lack of high availability, many users experienced hours of downtime and difficult deployments.

With Bamboo Data Center, in the event of an outage, cold standby disaster recovery simplifies the recovery process and minimizes downtime. The request router detects that an active instance went down, starts a cold instance, and redirects all traffic to the cold instance. Thanks to the shared state, the newly started instance continues the work, keeping your team on track. Further, builds that were in process during the server restart continue to build. The agent waits for the server to be back online and then delivers build results, logs, and artifacts. This ensures that users don’t lose the results of long-running builds. Learn more about Bamboo for Data Center.

Tip 5: Plan for your Bamboo 9 upgrade

Bamboo is planning its next platform release, Bamboo 9, which will establish the groundwork for coming performance development throughout the 9.x releases, including warm standby. This will develop a “primary-secondary” architecture to be used for even further high availability development down the road. With this, server crashes will no longer disrupt builds and critical path deployments, securing your team’s work.

I hope these tips helped you take advantage and get the most out of your Bitbucket and Bamboo instances!

[Pramodh M]

Thank you so much @Mel Policicchio 

This helps a lot for all Administrators!!

[Mel Policicchio]

So glad to hear it, @Pramodh M =)