You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Octopus Deploy will eventually stop supporting Nevermore. We will archive this git repository and retire the CI/CD infrastructure that supports it.
At this stage we have not set an exact date for this. We have some internal projects that still rely on Nevermore and it will remain supported until we transition fully away. Once this is done we will update this notice and provide a specific date. Roughly, we expect to complete the process and archive the repository at some point in 2026.
If you are currently using Nevermore, we recommend you start planning to migrate away from it as soon as reasonably possible.
Alternatively, Nevermore has always been Open Source under the Apache-2 license and you may choose to fork the repository.
Why sunset Nevermore?
Why Nevermore explains the rationale behind Nevermore; Octopus Deploy was built on top of a Document-based persistence model using RavenDB, and we wanted a way to keep the benefits of a Document-based system while gaining the rock-solid stability and support of Microsoft SQL server. Nevermore solved that problem, and during the decade of its existence it has served us well.
Octopus Deploy today is a larger product. We have many more customers, some of which have far larger data sets. We employ more software engineers, and the tradeoffs that are best for us to make today are different to the ones that were best 10 years ago. As most smaller companies do, historically we valued raw engineering velocity; In today's world other concerns like stability, performance, data integrity and the ability to onboard new software engineers rise in importance.
The technology landscape has changed dramatically too. When Nevermore was created .NET Core did not exist. Libraries like Microsoft Entity Framework were immature and had many significant problems. Today, we run Octopus Cloud using .NET 8 on Linux in Kubernetes, and Entity Framework is mature, robust and featureful.
As outlined in Defining a target architecture, we wanted to migrate Octopus Server away from Nevermore and over to Entity Framework. Over the course of approximately a year we completed this work; The release of Octopus 2024.4 completed the transition, no longer using Nevermore in any capacity.
Octopus has various other internal projects and systems built with Nevermore. A number of these were also migrated to Entity Framework, and we reached a point where internally there were no longer any major products or systems that used Nevermore. This meant there were no software development teams with the incentives or resources to do a good job of supporting Nevermore in the long term, leading to our decision to sunset it.
Questions or comments
Please reply to this issue if you have any further questions or comments regarding the Nevermore sunset process, we will do our best to answer them effectively.
The text was updated successfully, but these errors were encountered:
Summary
Octopus Deploy will eventually stop supporting Nevermore. We will archive this git repository and retire the CI/CD infrastructure that supports it.
At this stage we have not set an exact date for this. We have some internal projects that still rely on Nevermore and it will remain supported until we transition fully away. Once this is done we will update this notice and provide a specific date. Roughly, we expect to complete the process and archive the repository at some point in 2026.
If you are currently using Nevermore, we recommend you start planning to migrate away from it as soon as reasonably possible.
Alternatively, Nevermore has always been Open Source under the Apache-2 license and you may choose to fork the repository.
Why sunset Nevermore?
Why Nevermore explains the rationale behind Nevermore; Octopus Deploy was built on top of a Document-based persistence model using RavenDB, and we wanted a way to keep the benefits of a Document-based system while gaining the rock-solid stability and support of Microsoft SQL server. Nevermore solved that problem, and during the decade of its existence it has served us well.
Octopus Deploy today is a larger product. We have many more customers, some of which have far larger data sets. We employ more software engineers, and the tradeoffs that are best for us to make today are different to the ones that were best 10 years ago. As most smaller companies do, historically we valued raw engineering velocity; In today's world other concerns like stability, performance, data integrity and the ability to onboard new software engineers rise in importance.
The technology landscape has changed dramatically too. When Nevermore was created .NET Core did not exist. Libraries like Microsoft Entity Framework were immature and had many significant problems. Today, we run Octopus Cloud using .NET 8 on Linux in Kubernetes, and Entity Framework is mature, robust and featureful.
As outlined in Defining a target architecture, we wanted to migrate Octopus Server away from Nevermore and over to Entity Framework. Over the course of approximately a year we completed this work; The release of Octopus 2024.4 completed the transition, no longer using Nevermore in any capacity.
Octopus has various other internal projects and systems built with Nevermore. A number of these were also migrated to Entity Framework, and we reached a point where internally there were no longer any major products or systems that used Nevermore. This meant there were no software development teams with the incentives or resources to do a good job of supporting Nevermore in the long term, leading to our decision to sunset it.
Questions or comments
Please reply to this issue if you have any further questions or comments regarding the Nevermore sunset process, we will do our best to answer them effectively.
The text was updated successfully, but these errors were encountered: