Now you can specify the priority class for the Ceph daemon pods with a new setting in the CephCluster CR for priorit圜lassName. Since K8s 1.14, priority classes became available for ensuring system-critical pods would have priority during scheduling and eviction. In Rook v1.2, the operator will start a new “crash collector” daemon that will collect the crash information and report it to the Ceph mgr where the data can be analyzed. New in Ceph Nautilus is the ability to collect telemetry around daemon crashes and view them centrally from the Ceph dashboard. While in an ideal world nothing would ever crash, we know that in a complex system things you need the ability to troubleshoot when things go wrong. The new features in this release may not be ground-breaking, but they are moving the needle to the next level to provide more configuration options for your storage. Ceph continues providing block, file, and object storage to Kubernetes applications. The Ceph operator has been the core of Rook storage since the beginning. Hearing about your production usage will be helpful either way! Ceph Or if you prefer to keep your usage confidential we will certainly not share it publicly. If you are willing to have your usage published we will be happy to include your company on our Adopter’s doc. We’d like to know what size of clusters you are running, how long you have been running in production, what problems Rook has helped you solve, and more.Ī link to a survey will be sent out soon via the Rook Slack and Twitter accounts. As we work towards CNCF graduation, we need to collect data about the extent to which Rook is being run in production clusters. There is nothing to hide, which increases the level of security that much more! Production Usageĭo you run Rook in production? We would love to hear from you! Often with open source and upstream releases there are many users running our software and we aren’t even aware of it. Open source allows them to dig into all areas of the code. Keeping your data secure is always on our mind and this was a great validation of that mindset.Ī reassuring comment they made during the review is how refreshing it is to review open source projects. We are happy to say that no major concerns were found. It was very helpful to have another set of eyes on our code with security in mind. Even with all that effort on v1.1, there are plenty of new features in the v1.2 release to get excited about for the Ceph and EdgeFS storage providers! Securityīeyond all the features and stabilization, we have completed a security review with Trail of Bits, who also performed the security review for Kubernetes last year.
The Rook v1.1 release came with many new features and we have focused a lot of effort on stabilizing and improving that release. The statistics show how fast the community is growing around Rook since the v1.1 release just three months ago: It was great to meet many of you at KubeCon San Diego last month and hear all your positive feedback. Storage continues to be a critical area for Kubernetes administrators and we are excited about the momentum that continues for Rook to fill that need. Rook v1.2: More Storage Operator Enhancements!