 My name is Gautam Pacharavi and I'm the current PTL for the OpenStack Manila team. I'd like to give you a high-level overview of the project and give you an update about the things that we've accomplished in the recent Victoria release and our plans for Wallaby. So what is Manila? Manila is a service that seeks to provide OpenStack users the ability to provision and manage the life cycles of POSIX compliant shared and distributed file systems. It's inherently multi-tenant and secure. It is capable of providing hard network and data path isolation guarantees with the help of tenant-dedicated shared servers. Tenants from the get-go can determine who has access to a shared file system and this access can be revoked at any time in real time. Tenants can integrate their own authentication domains, so think Kerberos, Active Directory, or LDAP. Further, tenant resources are scalable and elastic, so they can be growing and shrinking shared file systems instantaneously and easily. Manila supports several NAS protocols like NFS, CFFS, CIFS, clusterFS, HDFS and so on and it has drivers for over 35 storage systems or solutions. It can make intelligent placement decisions to ensure that you're making optimum use of your shared storage. Manila also provides a flexible model to expose storage system service catalogs to end users in a discoverable and programmable way. So some statistics now. The Victoria cycle was the 11th production-ready release of Manila. This release continues to be supported by all major distributions and deployment tools. The Manila API has growing support in SDKs. So we'll soon have support in the OpenStack SDK and the Ansible Collections OpenStack. We have a healthy community maintaining this project. Members come from a variety of Linux distribution companies, storage system companies and we also have a lot of independent contributors, many of whom are students or new contributors that are making their first open-source contributions. There's growing adoption of Manila in public clouds as well as in high-performance computing. So let's preview a few things that the project team accomplished in the Victoria cycle. So we added support for shared server migration. This is a two-phase design and administrators can use this feature to facilitate cold and live migration of shared servers and they can go within storage pools or even across storage pools and back ends. And this feature has been implemented in the container and then add up storage drivers and more drivers are to follow in the next few releases. So the share application feature is now generally available. We've added this feature as experimental in the Mitaka cycle and over many cycles we've actually committed several improvements and many of which are now being well tested and well used over this time. So we no longer consider these APIs experimental. You don't need to include the experimental header to have access to these APIs and you can use these APIs to plan your load balancing or disaster recovery strategies. We had several driver feature improvements including in the container driver we added support for shared migration. We added support for adaptive QoS and share server transfer limits and then adapt driver. The Dell EMC Unity driver now supports a new driver filter and snapshots are fully supported in this FFS driver. Several client enhancements were made as well. We continue to improve on our OSC integration. The OSC client now supports interacting with shares, snapshots, access rules, share types, quotas and resize. We continue to play the catch up game and complete the parity there with the Python monologue client. We also added support for user messages in the UI. So users don't need to leave the UI in order to triage asynchronous failures that can happen and that can be reattempted. We made several improvements to testing and continuous integration throughout this cycle and I think this reflects in the number of bug fixes that we committed during this cycle because we added new test cases to several existing file system protocols or file system management modes like for example the hard multi-tenancy mode and for exclusively testing the admin interactions against various shared backends and stuff. So we also made many improvements in the Manila CSI land although the release cycle is not coordinated with the rest of OpenStack. It follows the Kubernetes release cycle but what coincided with the Victoria cycle has been the introduction of new Helm charts, a new OpenShift operator and support for OpenStack availability zones and also for shoving in any runtime configuration to make intelligent decisions while mounting shares onto the Kubernetes node plugins and for also share metadata to be added to tag the provision resources and so on. So all of this can also be used against older versions of OpenStack and that's the way that the driver has been implemented. So we just recently concluded our project technical gathering for the upcoming Wallaby cycle and so we have a fair idea of the things we want to accomplish in the current release cycle. First off is Verti-OFS. This is a novel file system attachment protocol that's been developed within the Linux kernel and it's aimed at virtual machines. So now that there is sufficient mainstream adoption for the kernel it's time to integrate that into OpenStack and so with this release we aim to provide file system attachments to NOVA VMs and with the help of NOVA APIs you could do what you could be doing with block devices today. So let's say you can execute OpenStack our FS attach or detach and expect NOVA to interact with Manila to gather all of the attachment info arbitrate the security and the access rules and so on mount the file system via the host kernel make it available to the guest virtual machine. So this should greatly enhance the user experience for Manila and NOVA users of shared file systems and it also provides a more secure way of accessing shared file system drivers in Manila that do not support the hard multi-tenancy guarantees the network path multi-tenancy guarantees that some of them do. So we're also looking to enhance support for the CFFS drivers in the upcoming cycle we will be adding support for enhanced snapshot cloning something that Ceph is back porting to the Ceph Nautilus release upstream and we're also adding support for the upcoming releases of Ceph such as Ceph Octopus and Ceph Pacific. We will also be making several RBAC and security improvements we'll be supporting the reader user admin role as well as refreshing policies to support the user scope feature that's been added to Keystone in the past several cycles. We're also planning to drop the use of rootwrap and provide a more secure and flexible way of privilege escalation via a Oslo-Privce app in this cycle. We also plan to enhance security services to be mutable so users can make any day-to-day changes to their security services or even add or remove security services on existing share networks. And we're trying to make the metadata APIs consistent across all user-facing resources in Manila and of course we'll continue to keep the momentum on OpenStack Client OpenStack SDK and as I said before we have several new contributors in the form of students that are looking to get involved with OpenStack or OpenSource and we're helping them help us land this important piece. And we're also looking to continue making UI improvements where we're going to be doing a version catch-up with the Manila API and in the CSI drivers we're looking to add support to share resize and also try to re-architect the driver to be a multi-protocol driver so that way it makes things easier for day-to-management and for observability and other concerns. So we have a lot to accomplish in terms of features and bug fixes and so we'd greatly benefit from your help. So should you be interested to contribute we'd love to have help in several areas like code, maintainer ship and documentation. We enjoy bringing new contributors on board and we're changing some of the process to make it easier to become a core reviewer. So please get in touch with us if you're interested. Alongside, there are a couple of useful links here for unfinished work that's important to the project team. So if you're willing to help, these are great places to start. That said, thank you so much for listening. We really appreciate your contributions, help and support in keeping us motivated and for making Manila better with each release.