[ClusterLabs] pcs 0.11.8 released
Michal Pospíšil (he / him)
mpospisi at redhat.com
Fri Jul 12 12:10:47 UTC 2024
I am happy to announce the latest release of pcs, version 0.11.8.
Source code is available at:
https://github.com/ClusterLabs/pcs/archive/refs/tags/v0.11.8.tar.gz
or
https://github.com/ClusterLabs/pcs/archive/refs/tags/v0.11.8.zip
Same as the new pcs-0.12 release, this pcs-0.11 release is deprecating
features that are planned to be removed in Pacemaker 3 to ease the
transition to the new major version.
This release brings the new pcs status wait/query commands which
allow for better scripting usage. Amongst many other scripting uses,
they can be used to check for any resource state in comparison with the
'--wait' option in resource commands which have hardcoded expected
states.
There is also now a possibility to export resource and stonith defaults
and operation defaults and many more new checks and bug fixes.
Complete change log for this release:
### Added
- Support for output formats `json` and `cmd` to resources/stonith
defaults and resource/stonith op defaults config commands
([RHEL-16231])
- Add lib commands `cib_options.resource_defaults_config` and
`cib_options.operation_defaults_config` to API v2
- Add lib command `cluster.wait_for_pcmk_idle` to API v2
- Add lib command `status.resources_status` to API v2
- Add `pcs status wait` CLI command ([RHEL-25854])
- Add `pcs status query resource` CLI commands ([RHEL-21051])
### Fixed
- Do not put empty uid/gid corosync configuration options to an uidgid
file when not specified in `pcs cluster uidgid add` command. Empty
options cause corosync start failure. ([ghissue#772])
- Do not allow fencing levels other than 1..9 ([RHEL-2977])
- Adjust OCF metadata processing to support the latest pacemaker changes
([RHEL-27492])
- Use different process creation method for multiprocessing module in
order to avoid deadlock on process termination. ([ghissue#780],
[RHEL-28749])
- Do not wrap pcs output to terminal width if pcs's stdout is redirected
([RHEL-36514])
- Report an error when an invalid resource-discovery is specified
([RHEL-7701])
- 'pcs booth destroy' now works for nodes without a cluster (such as
arbitrators) ([RHEL-7737])
- Validate SBD_DELAY_START and SBD_STARTMODE options ([RHEL-17962])
### Deprecated
- Pcs produces warnings about features planned to be removed in
pacemaker 3 [1]:
- score in order constraints
- using rkt in bundles
- upstart and nagios resources
- some date-spec and duration options in rules
Thanks / congratulations to everyone who contributed to this release,
including Ivan Devat, Michal Pospisil, Miroslav Lisik, Peter Romancik
and Tomas Jelinek.
Cheers,
Michal
[1]: https://projects.clusterlabs.org/w/projects/pacemaker/pacemaker_3.0_changes/pacemaker_3.0_configuration_changes/
[ghissue#772]: https://github.com/ClusterLabs/pcs/issues/772
[ghissue#780]: https://github.com/ClusterLabs/pcs/issues/780
[RHEL-2977]: https://issues.redhat.com/browse/RHEL-2977
[RHEL-7701]: https://issues.redhat.com/browse/RHEL-7701
[RHEL-7737]: https://issues.redhat.com/browse/RHEL-7737
[RHEL-16231]: https://issues.redhat.com/browse/RHEL-16231
[RHEL-17962]: https://issues.redhat.com/browse/RHEL-17962
[RHEL-21051]: https://issues.redhat.com/browse/RHEL-21051
[RHEL-25854]: https://issues.redhat.com/browse/RHEL-25854
[RHEL-27492]: https://issues.redhat.com/browse/RHEL-27492
[RHEL-28749]: https://issues.redhat.com/browse/RHEL-28749
[RHEL-36514]: https://issues.redhat.com/browse/RHEL-36514
More information about the Users
mailing list