This release is bigger than our usual releases: we’ve done a lot of internal improvement and upgrades. Please read the upgrade notes at the end of this post carefully before upgrading.
We are now shipping images with Ceph Nautilus. Run the Nautilus migration task found in the Maintenance tab to upgrade your cluster to Ceph Nautilus. We recommend all users to upgrade to Ceph Nautilus as it’s an awesome new release of Ceph!
As always: the migration task will handle all of the nitty-gritty details of the complex steps required to safely upgrade a cluster without any downtime.
Both our container and all Nautilus images are now based on Debian 10.0 (Buster). Yes, we are a week early with our release as Buster is officially scheduled to be released on July 6th, but no big changes are expected until then.
Update 2019-07-12: Buster has been released as scheduled. We’ve published a new docker container and server image as croit v1901.1 based on this official release.
SMB shares with Active Directory authentication
Our SMB servers can now join your Active Directory domain and you can use your AD users and groups to control access on a per-share basis.
Adding servers to croit is now easier than ever if your server supports IPMI. You can now create a DHCP network with the new type IPMI and croit will automatically detect and configure new IPMI devices. We even configure the server to boot from network via IPMI, so you never have to configure BIOS settings.
Better template support
We’ve started to use user-customizable templates for some config files in the last release. This release moves most config files that we generate to our template system: DHCP, Samba, ctdb, Ganesha, and the PXE boot menu. Customizing these templates gives you full control over the services we use internally, for example, it’s easy to add a custom entry to the boot menu or a custom DHCP option.
But you are probably using croit because you don’t want to fiddle with annoying template files. Don’t worry, we still aim to provide a UI for all commonly used settings, this change just gives more control to power users. Also, we ship a nifty template editor with auto-complete for all available variables to make editing templates as easy as possible.
Trouble finding a disk in your server? You can now easily turn on its identify LED. That works for all controllers that implement SGPIO, so virtually all SCSI controllers. But even RAID controllers not supporting this standard can be used as we build on libstoragemgmt which supports plugins.
RADOS namespace support
You can now set the RADOS namespace for CephFS folders and RBD images (requires Ceph Nautilus). Keys can also be restricted to namespaces. This feature is useful for multi-tenant environments: create folders or images in a namespace and give your users a Ceph key restricted to that namespace.
Last but not least
- We’ve updated our Nagios/Icinga script with finer granularity for the supported warnings
- Rolling restart for services: select multiple services and click restart
- croit no longer depends on ceph-deploy: more reliable initial cluster setup with unusual network configurations
- Dashboard now shows utilization per disk-type
- NFS shares can now be restricted to NFSv3 or NFSv4 only
- You can now set a password for logging in as root on the console
As always: ensure that you have a working backup (like our encrypted cloud backup) before upgrading the container.
There’s one additional step necessary when upgrading the container due to the upgrade to Debian Buster: please ensure that the MySQL database is shutdown correctly. The first command accomplishes that:
docker exec croit service mysql stop docker rm -f croit docker run --net=host --restart=always --volumes-from croit-data --name croit -d croit/croit:1907
Upgrading without running the first step might leave you with a container that fails to start the database. If you run into this: simply revert to the old container
S3 users must upgrade to Ceph Nautilus after upgrading to croit 1907 due to internal incompatibilities in Ceph, not upgrading Ceph after upgrading croit will disable some functions in our dashboard. S3 applications will continue normally.
The new Nautilus images come with Ceph iSCSI 3.x. Please reboot all iSCSI servers with the new image after upgrading to Ceph Nautilus. Some new features in iSCSI are only available after doing so, running mixed images will affect failover capabilities.