[ClusterLabs] [Announce] Release Candidate 3 for Pacemaker 1.1.13 now available

Andrew Beekhof andrew at beekhof.net
Fri May 15 00:54:43 EDT 2015


Announcing the third (and hopefully final) release candidate for Pacemaker 1.1.13

  https://github.com/ClusterLabs/pacemaker/releases/Pacemaker-1.1.13-rc2

In this RC we have some fixes for users of the now deprecated Corosync 1.x plugin from Yan, consistent behaviour for operations with on-fail=ignore from Ken, and improvements to David’s Docker integration.
Additionally, Pacemaker no longer crashes if a stonith agent’s metadata is unparsable.

Any assistance testing these changes would be much appreciated.
Please keep the bug reports coming in!


To build `rpm` packages for testing:

1. Clone the current sources:

     # git clone --depth 0 git://github.com/ClusterLabs/pacemaker.git
     # cd pacemaker

1. Install dependancies (if you haven't already)

     [Fedora] # sudo yum install -y yum-utils
     [Others] # make rpm-dep

1. Build Pacemaker

     # make rc

1. Copy and deploy as needed

## Details

Changesets: 79
Diff:       45 files changed, 1253 insertions(+), 436 deletions(-)

## Highlights
### Changes since Pacemaker-1.1.13-rc2

  + cib: Ability to upgrade cib validation schema in legacy mode
  + cib: Always generate digests for cib diffs in legacy mode
  + cib: Enable cib legacy mode for plugin-based clusters
  + cib: No need to enforce cib digests for v2 diffs in legacy mode
  + cib: Revert d153b86 to instantly get cib synchronized in legacy mode
  + common: Increment current and age for lib common as a result of APIs being added
  + corosync:  Bug cl#5232 - Somewhat gracefully handle nodes with invalid UUIDs
  + crmd/pengine: handle on-fail=ignore properly
  + crmd: don't update fail count twice for same failure
  + crmd: include remote nodes when forcing cluster wide resource reprobe
  + crmd: Prevent the old version of DC from being fenced when it shuts down for rolling-upgrade
  + crmd: report operation rc as advertised instead of status
  + crmd: respect start-failure-is-fatal even for artifically injected events
  + fence_legacy: Avoid passing 'port' to cluster-glue agents
  + fencing: Correctly make args for fencing agents
  + fencing: Gracefully handle invalid metadata from agents
  + HealthSMART: OCF_RESKEY_drives could contain multiple values
  + lrmd: Hint to child processes that using sd_notify is not required
  + lrmd: move isolation wrappers to hidden directory
  + mainloop: resolves race condition mainloop poll involving modification of ipc connections
  + make targetted reprobe for remote node work, crm_resource -C -N <remote node>
  + pengine: cl#5130 - Choose nodes capable of running all the colocated utilization resources
  + pengine: only kill process group on timeout when on-fail does not equal block.
  + pengine: Prevent use-after-free in sort_rsc_process_order()
  + pengine: process rsc parameters during unpack
  + pengine: restart master instances on instance attribute changes
  + services: Increment lib service's current version to indicate a new interface is being exported
  + xml: Do not dump deleted attributes
  + xml: Fix upgrade-1.3.xsl to correctly transform ACL rules with "attribute"





More information about the Users mailing list