crmsh - News
crmsh 2.0 released and moving to Github
Item posted by Kristoffer Grönlund <krig> on Thu 03 Apr 2014 03:57:02 PM UTC.
Today, I have two major announcements to make: crmsh is moving to a new location, and I'm releasing the next major version of the crm shell!
Find us at crmsh.github.io
Since the rest of the High-Availability stack is being developed over at Github, we thought it would make things easier to move crmsh over there as well. This means we're not only moving the website and issue tracker, we're also switching from Mercurial to git.
From this release forward, you will find everything crmsh-related at http://crmsh.github.io, and the source code at https://github.com/crmsh/crmsh.
Here are the new URLs related to crmsh:
- Website: http://crmsh.github.io/
- Documentation: http://crmsh.github.io/documentation.html
- Source repository: https://github.com/crmsh/crmsh/
- Issue tracker: https://github.com/crmsh/crmsh/issues/
Not everything has moved quite yet, but the source code and web site are in place.
New stable release: crmsh 2.0
Secondly, we are proud to finally release crmsh 2.0! This is the version of crmsh I have been developing since I became a maintainer last year, and there are a lot of new and improved features in this release.
For a more complete list of changes since the previous version, please refer to the changelog:
Packages for several popular Linux distributions (updated soon):
Zip archive of the tagged release:
Here is a short list of some of the biggest changes and features in crmsh 2.0:
- More stable than ever before! Many bugs and issues have been fixed, with plenty of help from the community. At the same time, this is a major release with many new features. Testing and pull requests are more than welcome!
- Cluster management commands. We've added a couple of new sub-levels that help with the installation and management of the cluster, as well as maintaining and synchronizing the corosync configuration across nodes. There are now commands for starting and stopping the cluster services, as well as cluster scripts that make the installation and configuration of cluster-controlled resources a one-line command.
- Cleaner CLI syntax. The parser for the configure syntax of crmsh has been rewritten, allowing for cleaner syntax, better error detection and improved error messages.
- Tab completion everywhere. Now tab completion works not only in the interactive mode, but directly from bash. In addition, the completion back end has been completely rewritten and many more commands now have full completion. It's not quite every single command yet, but we're getting there.
- New and improved configuration. The new configuration file is installed in /etc/crm/crm.conf by default or per user if desired, and allows for a much more flexible configuration of crmsh.
- Cluster health evaluation. As part of the cluster script functionality, there is now a cluster health command which analyses and reports on low disk space, problems with network configuration, firewall configuration issues and more. The best part of the cluster health command is that it can work without a configured cluster, providing a checklist of issues to amend before setting up a new cluster.
- And wait, there's more! There is now not only an extensive regression test suite but a growing set of unit tests as well, support for many new features in Pacemaker 1.1.11 such as resource sets in location constraints, anonymous shadow CIBs makes it easier to avoid race conditions in scripts, full syntax highlighting for the built-in help, the assist sub-command helps with more advanced configurations... the list goes on.
Big thanks to everyone who have helped with bug fixes, comments and contributions for this release!
// Kristoffer
Powered by Savane 3.14-8aba.
Corresponding source code