[Debian-ha-commits] [pacemaker] annotated tag upstream/1.1.14_rc4 created (now f41cc43)

Christoph Berg myon at debian.org
Wed Jan 6 20:15:28 UTC 2016


This is an automated email from the git hooks/post-receive script.

myon pushed a change to annotated tag upstream/1.1.14_rc4
in repository pacemaker.

        at  f41cc43   (tag)
   tagging  c2defa3797e38ca2508f49279abd7891002cfbc5 (commit)
  replaces  upstream/1.1.13
 tagged by  Ferenc Wágner
        on  Fri Dec 25 21:03:16 2015 +0100

- Log -----------------------------------------------------------------
Upstream version 1.1.14~rc4

Andrew Beekhof (230):
      Merge pull request #671 from johnruemker/host_list_semicolon
      Feature: crm_resource: Support debugging promote and demote operations
      Fix: crm_report: Allow log file discovery to be disabled
      Merge pull request #679 from HideoYamauchi/master
      Merge pull request #680 from kgaillot/mon
      Merge ../1.1
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Test: tools: Update regression tests for ACL fixes
      Feature: fencing: Allow complex configurations to apply to multiple hosts
      Fix: Fencing: Correctly handle changes to the fencing topology
      Merge pull request #705 from kgaillot/mon
      Merge branch 'rikkotec-debian-ha/patches'
      Merge pull request #699 from kgaillot/cts
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      CTS: Fix installation helper
      Merge ../1.1
      Merge pull request #714 from gao-yan/patchset-process-digest
      Merge pull request #719 from kgaillot/stonith
      Refactor: Tools: Start splitting up crm_resource into manageable chunks
      Merge pull request #721 from gao-yan/no-unit-assert
      Merge pull request #720 from gao-yan/dbus-reply-timeout
      Fix: PE: Ensure recurring monitor operations are cancelled when clone instances are de-allocated
      Fix: PE: Skip unrunnable actions when one-or-more is in effect
      Refactor: PE: More easily allow us to determine who changed an action's flag
      Test: PE: Skip unrunnable actions when one-or-more is in effect
      Merge pull request #722 from jnpkrn/ML-references
      Feature: PE: Reduce noise when sanitized policy engine files are replayed
      Fix: PE: Keep the size of the secure list small in the cib
      Feature: crm_report: Enable password sanitizing by default
      Fix: crmd: Prevent use-of-NULL
      Fix: PE: Prevent use of uninitialized data
      Fix: crm_simulate: Correctly invoke the policy engine library
      Merge branch 'remote_cancel_pending' of https://github.com/davidvossel/pacemaker
      Merge pull request #727 from davidvossel/service_rsc_collision
      Merge pull request #724 from jnpkrn/docbook-related
      Merge pull request #728 from davidvossel/lrmd-tls-to
      Feature: PE: Include a list of all and available nodes with notifications
      Fix: crmd: Pre-emptively fail in-flight actions when lrmd connections fail
      Fix: fencing: Correct the all_topology_devices_found() implementation
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Merge pull request #723 from kgaillot/peer-cache
      Merge pull request #718 from kgaillot/cleanup
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Merge pull request #711 from kgaillot/log
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Log: Core: Re-instate two log messages as debug
      Merge pull request #730 from marcan/master
      Fix: PE: Exclude nodes which don't match any exclusive discovery rules
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Merge pull request #734 from davidvossel/dup_recurring_conn_monitors
      Fix: crmd: Record actions in the CIB as failed if we cannot execute them
      Merge pull request #735 from davidvossel/dup_recurring_conn_monitors
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Merge pull request #732 from gao-yan/load-stopped-loop
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Log: crmd: Better logging of incomplete actions
      Feature: crm_simulate: Optionally include action IDs in the dot graph
      Fix: crmd: All pseduo ops must be allowed to complete
      Log: crmd: Scheduled actions are less interesting
      Fix: crmd: No need to re-process actions we've already marked as failed
      Log: crmd: Some entries can just go to the application logfile
      Fix: crmd: Prevent segfault when erasing resources from disconnected remote nodes
      Merge pull request #738 from gao-yan/memory-leak-expand_notification_data
      Feature: PE: Record the execution date in PE files so we can re-use it in crm_simulate
      Merge pull request #737 from gao-yan/xpath-delete-all
      Merge pull request #739 from jnpkrn/obsolete-extra-rgmanager
      Merge pull request #740 from davidvossel/group-guest-nodes
      Merge pull request #729 from kgaillot/stonith
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Fix: PE: Add the execution date to the correct xml blob
      Merge pull request #745 from davidvossel/lrmd-segfault-st
      Merge pull request #743 from kgaillot/cleanup
      Log: lrmd: Indicate the original action where appropriate
      Fix: PE: Ignore comment blocks when unpacking the cib
      Merge ../1.1
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Test: cts: Extra patterns to ignore
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Fix: Dbus: Older distros didn't define DBUS_TIMEOUT_USE_DEFAULT
      Doc: Update the ping metadata
      Merge pull request #757 from kgaillot/cleanup
      Merge pull request #750 from gao-yan/op-force-restart
      Merge pull request #746 from liu4480/master
      Merge pull request #753 from kgaillot/stonith
      Merge pull request #733 from jnpkrn/asciidoc.reference
      Merge pull request #731 from jnpkrn/pcmk-intro
      Update pacemaker-intro.txt
      Merge pull request #758 from kgaillot/stonith
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Merge pull request #760 from gao-yan/update-version-1.1.13
      Merge pull request #759 from gao-yan/fencing-broadcast-uname
      Test: cts: Logging format changed
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Doc: Ansible script for deploying a 3-node cluster with docker
      Fix: pengine: Correctly bypass fencing for resources that do not require it
      Build: Bump library versions
      Merge pull request #761 from kgaillot/cleanup
      Merge pull request #762 from davidvossel/clone-min
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Fix: crm_mon: Memory leaks
      Fix: Fencing: Repair cut and paste error
      Fix: xml: Prevent use-of-NULL in crm_xml_dump()
      Log: lrmd: Improved logs
      Merge pull request #765 from kgaillot/cleanup
      Merge pull request #763 from kgaillot/cman-cleanup
      Fix: systemd: Track pending operations so they can be safely cancelled
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Fix: systemd: Ensure pending structs are correctly unreferenced
      Fix: upstart: Ensure pending structs are correctly unreferenced
      Merge pull request #766 from kgaillot/cleanup
      Fix: dbus: Remove redundant ref/unref of pending call records
      Use the print function instead of the directive
      Build: tools: Drop dead files
      Test: CTS: Fix whitespace for better python3 compatability
      Test: CTS: Use list() for better python3 compatability
      Test: CTS: apply is deprecated in py2 and gone in py3
      Test: CTS: int() is enough and long() is gone in py3
      Test: CTS: Version independant check for a string type
      Test: CTS: Avoid using the has_key() built-in as its not available in py3
      Build: Ensure we track libraries as they are added
      Fix: crmd: Avoid cancelling transitions when we tell nodes to shutdown
      Fix: crm_resource: Correctly clean up failcounts for inactive anonymous clones
      Log: xml: Reduce severity of noisy log message
      Merge pull request #768 from jnpkrn/drop-legacy-SysVStartPriority
      Merge pull request #767 from HideoYamauchi/master
      Doc: PE: Document operation roles
      Merge pull request #771 from gao-yan/ipc-buffer-size
      Fix: cman: Corosync invents node names of the form Node{id} for peers not in its node list, wait for the real ones
      Fix: crm_resource: Compensate for -C and -p being called with the child resource for clones
      Feature: crm_resource: Indicate common reasons why a resource may not start after a cleanup
      Merge pull request #769 from gao-yan/cast-64bit-time_t
      Merge pull request #773 from gao-yan/order-stonith_dc-stonith_complete
      Merge pull request #770 from kgaillot/stonith
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Test: tools: Update regression tests
      Merge pull request #775 from gao-yan/remote_proxy_cb-memory-leak
      Merge pull request #774 from HideoYamauchi/master
      Merge pull request #772 from gao-yan/suggested-ipc-buffer-size
      Merge pull request #776 from kgaillot/doc
      Merge pull request #777 from kgaillot/stonith
      Fix: crm_resource: Correctly check if a resource is unmanaged or has a target-role
      Fix: PE: Bug cl#5247 - Imply resources running on a container are stopped when the container is stopped
      Fix: Date: Correctly set time from seconds-since-epoch
      Test: PE: Bug cl#5247 - Imply resources running on a container are stopped when the container is stopped
      Merge pull request #778 from kgaillot/cleanup
      Merge pull request #783 from davidvossel/master
      Merge pull request #782 from HideoYamauchi/master
      Merge pull request #779 from jnpkrn/cluster_test+README
      Merge pull request #786 from gao-yan/services-log-trace
      Merge pull request #785 from gao-yan/xml-move-dirty
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Feature: crmd: Implement reliable event notifications
      Fix: cman: Suppress implied node names
      Fix: crmd: Choose more appropriate names for notification options
      Fix: crmd: Correctly enable/disable notifications
      Fix: crmd: Report the completion status and output of notifications
      Fix: cman: Print the nodeid of nodes with fake names
      Refactor: Tools: Isolate the paths which truely require corosync-2.x
      Fix: corosync: Display node state and quorum data if available
      Fix: pacemakerd: Do not forget about nodes that leave the cluster
      Fix: pacemakerd: Track node state in pacemakerd
      Merge pull request #784 from kgaillot/doc
      Merge pull request #780 from kgaillot/stonith
      Fix: PE: Resolve memory leak
      Fix: cman: Purge all node caches for crm_node -R
      Refactor: membership: Safely autoreap nodes without code duplication
      Merge branch 'master' of github.com:ClusterLabs/pacemaker
      Merge pull request #789 from kgaillot/premote
      Merge pull request #787 from feldsam/snmp-trap
      Merge pull request #781 from jnpkrn/cts-script
      Merge pull request #791 from jnpkrn/various-typos
      Merge pull request #793 from kgaillot/premote
      Log: The package version is more informative
      Build: xml: Prepare for modification to the fencing schema
      Fix: Fencing: Use dedicated fields for fencing level patterns and attribute matching
      Log: fencing: Some additional trace logs
      Fix: Fencing: Correctly parse 'by-attribute' fencing levels
      Refactor: cts: Simplify fencing choice logic
      Fix: crm_resource: Allow the resource configuration to be modified for --force-{check,start,..} calls
      Log: lrmd: Improved logging when no pacemaker remote authkey is available
      Merge pull request #790 from kgaillot/doc
      Merge pull request #796 from roidelapluie/kind-doc
      Merge pull request #795 from kgaillot/cts
      Merge pull request #794 from gao-yan/resource-agent-remote
      Build: Prefer built binaries when invoking help2man
      Fix: Fencing: Correctly remove topologies and detect invalid entries
      Fix: Fencing: Correctly construct simple topologies with stonith_admin
      Feature: PE: Allow resources to start and stop as soon as their state is known on all nodes
      Feature: PE: drop probe_complete actions
      Build: PE: Prototype of start A before probing B
      Merge pull request #799 from gao-yan/cib_config_changed-v1
      Merge pull request #798 from kgaillot/sh
      Merge pull request #797 from kgaillot/coverity
      Test: tools: Update regression tests for ordered probes
      Merge pull request #801 from wferi/typo-fixes
      Merge pull request #805 from kgaillot/stonith
      Merge pull request #802 from wferi/typo-fix
      Merge pull request #804 from kgaillot/coverity
      Merge pull request #792 from jnpkrn/crm_attribute
      Fix: PE: resolve memory leak
      Fix: Tools: Repair the logging of 'interesting' command-lines
      Feature: Tools: Do not send command lines to syslog
      Fix: Fencing: The device list is expected to be comma separated
      Log: cibadmin: Default once again to LOG_CRIT
      Fix: crm_resource: Correctly update existing meta attributes regardless of their position in the heirarchy
      Merge pull request #807 from kgaillot/stonith
      Merge pull request #806 from gao-yan/cibadmin-use-of-null
      Merge pull request #808 from kgaillot/premote
      Log: crm_resource --restart: Improved user feedback on failure
      Build: Do more when bootstrapping an fresh repo
      Merge pull request #800 from wferi/build-fixes
      Merge pull request #809 from gao-yan/crm_resource-delete-parameter
      Fix: crm_resource: Correctly observe --force when deleting and updating attributes
      Test: crm_resource: Additional regression tests
      Test: acls: Update whitespace in regression results
      Fix: cib: Increased paranoia when peer updates fail to apply in compatability mode
      Build: Switch to the LTS version of ubuntu for travis
      Merge pull request #814 from HideoYamauchi/master
      Merge pull request #812 from gao-yan/fencing-provider-internal
      Merge pull request #813 from gao-yan/cib-forward-request-message
      Merge pull request #815 from kgaillot/premote
      Fix: Correctly check for recycled PIDs in crm_pid_active
      Fix: cib: Do not terminate due to badly behaving clients
      Fix: systemd: The underlying systemd calls default to 90s for their timeouts
      Merge pull request #833 from kgaillot/premote
      Fix: systemd: Prevent segfaul when logging failed operations
      Merge pull request #816 from kgaillot/stonith
      Merge pull request #832 from gao-yan/multiple-active-block-group
      Log: Fencing: Update log messages for greater clarity
      Merge pull request #836 from kgaillot/stonith

David Vossel (46):
      Merge pull request #688 from digimer/master
      Merge pull request #703 from ClusterLabs/1.1
      Feature: remote: reconnect_delay option for remote nodes to delay reconnect after fence
      High: crmd: fence remote node after connection failure only once
      High: enable reloads for select remote connection options
      Low: pengine: regression test updates
      Low: pengine: make node_state key 'remote_was_fenced' generic for all nodes
      Merge pull request #708 from davidvossel/failure_reconnect_delay
      High: crmd: make actions incompletable after remote-node fails
      Fix: crmd: handle resources named the same as cluster nodes
      Merge pull request #726 from davidvossel/rsc_node_name
      High: services: prevent recurring monitors from colliding with other operations
      Low: lrmd: regression tests for serializing recurring monitor ops
      Low: extras: Use dummy resource to help test serialized action execution
      High: lrmd: set recv timeout upper bound for tls connections
      High: remote: handle merging duplicate recurring monitors for connection rsc
      High: remote: do not fail operations because of a migration
      High: pengine: fixes segfault in pengine when fencing remote node
      Low: pengine: regression test update for stonith pengine fix
      High: pengine: do not stop notify a fenced node that the rscs on the fenced node stopped
      High: remote: copy transition userdata to duplicate monitor during merging
      High: pengine: do not timeout remote node recurring monitor op failure until after fencing
      Low: pengine: regression tests for fence before clear remote monitor failure
      High: pengine: allow guest remote nodes using containers/vms to be nested in a group resource
      Low: pengine: regression test for guest remote node in group
      Merge pull request #736 from davidvossel/dup_recurring_conn_monitors
      Fix: lrmd: prevent double free after unregistering stonith device for monitoring
      Merge pull request #744 from davidvossel/lrmd-segfault-st
      Low: crmd: don't synthesize op failures during shutdown
      Low: cts: phd fencing agent
      Merge pull request #747 from davidvossel/cts-phd
      Merge pull request #742 from kgaillot/stonith
      Merge pull request #749 from kgaillot/cleanup
      Merge pull request #751 from gao-yan/dummy-stop-error
      High: pengine: clear failures after reconnect_interval expires
      Merge pull request #748 from kgaillot/stonith
      Merge pull request #754 from gao-yan/crmd-lrm-fixes
      Merge pull request #752 from davidvossel/pcmk-remote-failcount
      Low: cts: randomly use reconnect-interval during pcmk remote tests
      Merge pull request #755 from davidvossel/master
      Low: cts: properly cleanup RemoteStonithd test
      Merge pull request #756 from davidvossel/master
      Feature: pengine: Addtion of the clone resource clone-min metadata option
      Fix: pengine: properly handle blocked clone actions
      Low: pengine: clone-min regression tests
      Low: update contact info

Ferenc Wágner (12):
      Fix typos: dependan{t,cy,cies} -> dependen{t,cy,cies}
      Fix typo: maintainence -> maintenance
      Fix typo: initalize -> initialize
      Fix typo: Dont -> Don't
      Fix typo: OPSTIONS -> OPTIONS
      Fix typo: Shuting -> Shutting
      Link libcrmcommon with the math library
      Substitute additional configure macros used in pkg-config files
      Library options go into Libs, not Cflags
      Don't create doc/775, the intent was setting the mode
      hb2openais was removed by c2344c9, kill its documentation, too
      Imported Upstream version 1.1.14~rc4

Gao,Yan (48):
      Refactor: cib: Calculate and add digest for a patchset after accepting changes for the target xml
      Fix: crm_shadow: Fix "crm_shadow --diff"
      Fix: dbus: Use operation timeout for dbus reply timeout
      Fix: systemd: No need to trigger assert if cannot obtain an unit
      Fix: pengine: cl#5235 - Prevent graph loops that can be introduced by "load_stopped -> migrate_to" ordering
      Test: pengine: cl#5235 - Prevent graph loops that can be introduced by "load_stopped -> migrate_to" ordering (update regression test)
      Test: pengine: cl#5235 - Prevent graph loops that can be introduced by "load_stopped -> migrate_to" ordering (regression test)
      Fix: cib: Prevent use-after-free when invoking "cibadmin --delete-all --xpath"
      Fix: cib: Prevent use-after-free and return -EINVAL when attempting to delete the whole "/cib"
      Fix: pengine: Resolve memory leak in expand_notification_data()
      Fix: RA: Prevent an error message on stopping "Dummy" resource
      Fix: crmd: Do not enforce restart for changes of non-unique parameters
      Fix: crmd: Do not enforce restart for resources that support reload and don't support any unique parameters
      Fix: crmd: Remove a duplicate invocation of snprintf()
      Fix: crmd: Make resource_supports_action() really general
      Fix: fencing: Broadcast uname as soon as a peer's state has changed
      Build: Update version numbers to 1.1.13 in version.m4 and sample spec file
      Fix: ipc: Correctly compare values for the size of ipc buffer and prevent suggesting a negative value when it's insufficient
      Fix: crm_simulate: Prevent segfault on arches with 64bit time_t
      Refactor: fencing: Remove unnecessary casts
      Fix: ipc: Fix output formats
      Fix: pengine: Ensure fencing of the DC precedes the STONITH_DONE operation
      Test: pengine: Ensure fencing of the DC precedes the STONITH_DONE operation (update regression tests)
      Fix: ipc: Do not constantly increase suggested size for PCMK_ipc_buffer every time we find it's insufficient
      Fix: crmd: Resolve memory leak in remote_proxy_cb()
      Fix: crmd: Initialize an integer
      Fix: xml: Mark xml nodes as dirty if any children move
      Log: services: Reduce severity of noisy log messages
      Fix: remote: Correctly display the usage of the ocf:pacemaker:remote resource agent
      Fix: remote: Revise a misleading message in the ocf:pacemaker:remote resource agent
      Fix: cib: Check if the configuration changes with cib_config_changed() only for v1 diffs
      Fix: cibadmin: Prevent potential use-of-NULL in print_xml_output()
      Fix: crm_resource: Correctly delete existing meta attributes regardless of their position in the heirarchy
      Fix: fencing: Return a provider for the internal fencing agent "#watchdog" instead of logging an error
      Log: cib: Downgrade the log message on forwarding CRM_OP_NOOP requests from INFO to DEBUG
      Fix: lrmd: Finalize all pending and recurring operations when cleaning up a resource
      Fix: systemd: Reconnect to System DBus if the connection is closed
      Fix: services: Correctly determine if operations are in-flight
      Fix: systemd: Directly return an error if the connection to System DBus is closed
      Fix: services: Prevent potential use-after-free when trying to add an operation into in-flight list
      Fix: dbus: Correctly return the result for pcmk_dbus_find_error() no matter if there's a DBusError passed in
      Fix: systemd: Return FALSE for systemd_mask_error() if it only maps an error instead of masking one
      Fix: systemd: Return PCMK_OCF_UNKNOWN_ERROR instead of PCMK_OCF_NOT_INSTALLED for uncertain errors on LoadUnit
      Refactor: services: Directly add operations into in-flight list in the execution functions
      Fix: crm_resource: Correctly clean up all children for anonymous cloned groups
      Feature: pengine: Support of multiple-active=block for resource groups
      Test: pengine: Support of multiple-active=block for resource groups (regression test)
      Doc: ping: Clarify the description of host_list parameter in ping metadata

Hector Martin (1):
      Fix: lrmd: cancel currently pending STONITH op if stonithd connection is lost

Hideo Yamauchi (8):
      Low: mcp: Add 'start on started local and runlevel' to Upstart job.
      Fix: Set the attribute from remote node.
      Fix: Set the attribute from remote node.
      Low: Change the log of the noise to the trace log.
      Fix: pengine: The failed action of the resource that occurred in shutdown is not displayed.
      Low: libcommon: Ignore CDATA of metadata of the resource.
      Low : Change of the mention example.
      High : Mistake of the parameter order.

Jan Pokorný (54):
      Low: doc: update mailing list references
      Build: prefer more effective Docbook template discovery
      Med: doc: fix unability to build CFS on Fedora 22
      Low: doc: unify Preface.xml between CFS and PE
      Low: doc+build: remove bogus Makefile target
      Med: doc+build: make shared chapters a proper dependency
      Doc: asciidoc.reference: cleanup (tabs, typos)
      Doc: asciidoc.reference: extend, use more abstract styles
      Doc: asciidoc.reference: frame a note as admonition + headlines
      Doc: shared: gently rewrap to width of 72, fix a footnote
      Doc: shared/pcmk-intro: fix extraneous/missing punctuation
      Doc: shared/pcmk-intro: add more mark-up
      Doc: shared/pcmk-intro: re{spell,word} "aka." and "eg."
      Doc: shared/pcmk-intro: add a "distributed != clustered" note
      Doc: shared/pcmk-intro: more elaborated pcmk explanation
      Doc: shared/pcmk-intro: cluster stack is a little bit bigger
      Doc: shared/pcmk-intro: note about HA cluster vs. HPC
      Doc: shared/pcmk-intro: don't assume reader knows STONITH
      Doc: asciidoc.reference: add an example of footnote
      extra/rgmanager: finally obsolete in favor of clufter
      Low: drop legacy SysVStartPriority from service units
      Test: cts: cluster_test: plain printf -> echo
      Test: cts: cluster_test: printf to use positional args
      Test: cts: cluster_test: better formatting on SSH failure
      Test: cts: cluster_test: simplify storing of test profile
      Test: cts: cluster_test: fix tests =/!= -> -eq/-ne
      Test: cts: cluster_test: fix enquoting of the variables
      Test: cts: cluster_test: simplify comparisons
      Test: cts: cluster_test: prefer parameter expansion over sed
      Test: cts: cluster_test: prevent overwriting run-time info
      Test: cts: cluster_test: only override numtests from cmdline if suitable
      Test: cts: cluster_test: avoid using obsolete CTS_master
      Test: cts: README: same indentation for code blocks
      Test: cts: README: reference cluster_test script
      Test: cts: cluster_test: dedicated die function
      Test: cts: cluster_test: automated identity installation
      Test: cts: cluster_test: ensure trailing space at prompts
      Test: cts: simplify ssh-agent invocation
      Test: cts: cts: add a help screen
      Test: cts: cts: simplify+disambiguify distro selection
      Test: cts: cts: more general "case" match pattern goes last
      Test: cts: cluster-helper: fix adding extraneous '-w' node(s)
      Test: cts: cts: save cluster/pdsh group members for the next time
      Test: cts: cts: make --wget download cluster-init as well
      Test: cts: cts: fix quoting/expansion of "run" pass-through args
      Test: cts: cts: remove extraneous comment
      Test: cts: cluster-init: fix redundant assignment
      Test: cts: cluster-init: fix digits-only + range of $dsh_group
      Low: fix various typos
      Tools: crm_attribute: fix typos
      Tools: crm_attribute: make help screen more consistent
      Codedoc: libcib: cib_process_xpath
      Med: libcib: find_nvpair_attr_delegate: check alloc failure
      Med: libcib: find_nvpair_attr_delegate: check buffer overflow

John Ruemker (1):
      fencing: Allow semi-colon delimiter for pcmk_host_list

Julien Pivotto (1):
      Fix letter case in ordering 'kind' parameter documentation

Ken Gaillot (246):
      tools: prettify crm_mon --simple-status output
      tools: prettify crm_mon HTML output
      Refactor: tools: remove dead code and add more comments in crm_mon
      tools: prettify plaintext output
      Refactor: tools: use bitmask for selecting display items in crm_mon
      tools: improve guest node output in crm_mon
      Refactor: tools: use enum for selecting output format in crm_mon
      tools: improve cluster summary display in crm_mon
      tools: implement printing node attributes in crm_mon HTML/XML output
      tools: implement printing operations and failcounts in crm_mon HTML/XML output
      tools: implement printing negative location constraints in crm_mon HTML/XML output
      tools: implement printing tickets in crm_mon HTML/XML output
      tools: improve failed actions display in crm_mon
      Fix: cts: get rid of useless startup messages
      Fix: cts: improve dummy systemd service file used in testing
      Refactor: cts: remove dead code including unused imports
      Fix: cts: replace call to undefined usage() function
      Test: cts: make log message patterns independent of function name
      xml: update RelaxNG schema for crm_mon XML output
      Fix: tools: ignore stderr from crm_mon when running via systemd
      Fix: fencing: missing argument to stonith_device_register()
      Fix: fencing: correct choice of fencing topology level
      Refactor: attrd,tools: use defined constant for attrd peer-remove op
      Refactor: crmd: improve log messages for node state
      Fix: libcluster: do not re-add peer entries for exiting nodes
      Fix: libcluster,pacemakerd: improve trace messages and comments
      Fix: Ensure peer cache management supports all stacks
      Fix: libcluster: avoid possibility of invalidating a hash table iterator
      Fix: attrd: handle removals of cluster peers and remote peers consistently
      Fix: libcluster: overhaul peer cache management
      Low: libpengine,libcluster,crmd: update obsolete comments
      Low: cts,mcp,pengine,libcluster,libpengine: misc. spelling corrections
      Test: cts: make more log message patterns independent of function name
      Test: cts: add more error checking and debugging
      Feature: Log: make syslog less noisy
      Feature: Log: improve IPC connection error messages
      Fix: fencing: properly sort peers by number of fencing devices found
      Fix: fencing: properly decide whether a topology fencing device has been found
      Refactor: fencing: use defined constant for "st-available-devices" XML
      Refactor: fencing: remove dead code
      Refactor: fencing: simplify and comment get_action_timeout() for readability
      Refactor: fencing: simplify and comment cib_device_update() for readability and code reuse
      Refactor: fencing: code reuse, improve comments/messages in process_remote_stonith_query()
      Refactor: fencing: improve comments and log messages
      Fix: cts: don't rely on "Updating failcount" pattern, which is no longer syslogged
      Fix: cts: use proper kill syntax in dummy systemd service
      Fix: libcommon: crm_ipc_get_fd() must set errno on error
      doc: fencing: add README describing stonith request handling
      Fix: fencing: do not set call options twice in query
      Refactor: fencing: functionize testing whether localhost is eligible to fence
      Refactor: fencing: functionize adding action-specific device properties to XML
      Refactor: fencing: functionize parsing action-specific device properties from XML
      Refactor: fencing: isolate access to peers' device lists
      Refactor: fencing: make structure member name for device count clearer
      Fix: fencing: use "off" timeout when remapping "reboot" to "off"
      Refactor: fencing: improve comments and log messages
      Fix: cts: handle clones when checking for resource failures
      Fix: testing: suppress valgrind warning about memory leak in bash
      Fix: test: add fence delays to avoid regression test timing issues
      Refactor: fencing: isolate access to the required devices list
      Refactor: fencing: functionize advancing to next device in topology
      Refactor: fencing: simplify required device handling
      Refactor: fencing: functionize adding one device's properties to peer result
      Fix: fencing: log all peers' results consistently
      Refactor: fencing: abstract query results' device properties into own structure
      Refactor: fencing: mark devices as done rather than remove them
      Refactor: fencing: pass operation rather than default timeout to timeout functions
      Feature: fencing: add support for operation phases
      Feature: fencing: add support for operation phases to device properties
      Fix: fencing: improve log messages
      Fix: libcluster: work around peer cache destruction in client status callbacks
      Fix: testing: escape dollar sign when creating DummySD service file
      Fix: lrmd: memory leak when freeing command structure
      Fix: fencing: memory leak when searching for a target's fencing topology
      Fix: attrd: use-after-free in attrd_client_update()
      Fix: libcib,libfencing,tools: memory leaks from xmlGetNodePath()
      Fix: crmd: check for errors when searching for CIB daemon's load file
      Feature: fencing: support remapping reboots in fencing query results
      Feature: fencing: treat remapped reboots as reboots when appropriate
      Feature: fencing: support remapping reboots when processing fencing query results
      Feature: fencing: remap sequential topology reboots to off
      Feature: fencing: remap sequential topology reboots to all-off-then-all-on
      doc: document reboot remapping in Pacemaker Explained
      Refactor: test: reformat fencing regression test to current standards
      Feature: test: add remapped topology reboots to fencing regression tests
      Fix: cib,cts: more sensible messages under valgrind
      Fix: testing: don't print "Filtering" messages twice when running under valgrind
      Feature: testing: add lrmd to CTS's default valgrind processes and allow user to override via option
      Fix: crmd: memory leaks in recurring operation history
      Fix: libservices: memory leak in DBus-related log messages
      Fix: libservices: potential memory leaks when executing systemd actions
      Fix: tools,libcluster: zero memory before calling CMAN node functions
      Fix: libcluster: memory leak in CMAN cluster setup
      Fix: testing: add known CMAN non-issue to valgrind suppressions
      Feature: libcib: allow free function for user data supplied to CIB callbacks
      Fix: attrd: memory leak in CIB callback
      Fix: attrd: memory leak in legacy attrd's CIB callback
      Fix: crmd: memory leak in CIB callbacks
      Test cts: tweak pattern for failed ops
      Fix: cib: don't destroy peer cache in peer status callback
      Fix: stonithd: potential device list corruption
      Fix: libservices: add DBus call reference when getting properties
      Refactor: fencing: use enum value for max phase for array sizes
      Fix: fencing: ensure "required"/"automatic" only apply to "on" actions
      Refactor: fencing: simplify required action handling for fencing devices
      Refactor: fencing: simplify required action handling for fencing operations
      Refactor: fencing: remove dead code
      Feature: fencing: update comments/variables/messages for ability to target by attribute
      Feature: fencing: add ability to target by attribute
      Feature: fencing: update stonith_admin help text for targeting by attribute
      Test: fencing: update stonith-test for targeting by attribute
      Testing: cts: simplify node counting
      Testing: cts: get CIB.py unit test working again
      Testing: cts: print tag in CIB log message when there is no id
      Testing: cts: add target-by-attribute to the random fencing configurations
      Fix: add generated XML files to .gitignore
      doc: Clusters From Scratch: make OS references more generic
      doc: Clusters From Scratch: minor corrections and improvements to text
      doc: Clusters From Scratch: use xfs instead of ext4 in examples
      doc: Clusters From Scratch: leave firewalld and SELinux enabled
      doc: Clusters From Scratch: remove Fedora 13 images and add CentOS 7.1 images
      doc: Clusters From Scratch: redo for CentOS 7.1
      Fix: tools: memory leak in crm_resource
      Feature: fencing: allow colon as name-value separator when targeting by node attribute
      doc: Pacemaker Remote: update feature support in Pacemaker versions
      doc: Pacemaker Remote: replace LXC references
      doc: Pacemaker Remote: format consistently
      doc: Pacemaker Remote: target CentOS 7
      doc: Pacemaker Remote: update for current node terminology
      doc: Pacemaker Remote: bump documentation version
      Fix: cts: properly add node attributes when targeting fencing by attribute
      doc: Pacemaker Remote: document command line support better
      doc: Pacemaker Remote: add detail to configuration chapter
      doc: Pacemaker Remote: add "Alternative Configurations" chapter
      Fix: crmd: don't add node ID to proxied remote node requests for attrd
      Log: remote: make logs less noisy
      Fix: cts: properly clean up after remote node testing
      Fix: cts: update detection of recovery of remote node dummy resource
      Testing: cts: improve remote node fencing configuration
      Testing: cts: record errors consistently in remote node tests
      Doc: cts: document testing with valgrind and remote nodes
      Doc: Pacemaker Explained: spelling/grammar corrections
      Doc: Pacemaker Explained: add detail about infinity scores
      Doc: Pacemaker Explained: add detail about fencing parameters
      Doc: CTS: move README to README.md
      Doc: CTS: convert README.md contents to markdown format
      Doc: CTS: document how to run pacemaker_remote under valgrind
      Fix: pacemaker_remote: memory leak in ipc_proxy_dispatch()
      Fix: crmd: test script existence properly before sending fencing notification
      Refactor: CTS: functionize code for checking whether file exists on all nodes
      Test: CTS: add command line options for notification agent and recipient
      Test: CTS: more validation on user input of fencing parameters
      Test: CTS: use secure temporary files for bandwidth and remote tests
      Doc: Pacemaker Explained: add detail about resource operations
      Fix: controld: replace bashism
      Fix: crmd,libservices,tools: cast unwanted return values to void
      Fix: libcib: potential user input overflow
      Refactor: libcrmcommon: convert crm_element_name() from macro to inline function
      Refactor: fencing,libcrmcommon,libpengine: remove dead code
      Fix: crmd: avoid potential null dereference
      Fix: libcrmcommon: better validation of environment variable value
      Fix: libcib: properly handle temporary file
      Refactor: libcrmcommon: tweak crm_perror() macro to keep coverity quiet
      Fix: liblrmd: don't print error if remote key environment variable unset
      Fix: crmd,fencing: avoid potential null dereference in string searches
      Fix: crmd,libcrmcommon,libservices,tools: potential memory leaks
      Refactor: libcrmcommon: add procfs handling functions
      Refactor: crmd,libcrmcommon,mcp: use new procfs functions to reduce duplicated code
      Fix: tools: use floating-point division when converting ms to seconds
      Fix: liblrmd: make LSB header parsing more efficient
      Fix: tools: improve error handling when modifying configuration
      Refactor: libcrmcommon: ensure crm_md5sum() never passes NULL to md5_buffer()
      Fix: fencing: parse from correct XML location when registering/unregistering topology
      Fix: libfencing: make topology level removal compatible with latest stonithd API
      Test: fencing: remove target-by-attribute test from stonith-test
      Fix: crmd: properly detect CIB update failures for remote nodes
      Fix: fencing,libfencing: use level tag consistently in internal XML
      Test: lrmd: fix description of ocf_stress test
      Test: CTS: get Reattach test working again and up-to-date
      Refactor: CTS: simplify pacemaker_remote test classes
      Test: CTS: fix issues with multiple runs of a pacemaker_remote test
      Doc: Pacemaker Remote: update KVM walk-through
      Refactor: crmd,libcluster: deprecate the crm_remote_container and crm_remote_baremetal node flags
      Refactor: crmd: add enumeration for all node update fields
      Refactor: crmd: improve generation of node state
      Refactor: libcrmcommon: separate xpath-related functions into their own source code file
      Feature: libcrmcommon: new function for processing xpath search results
      Fix: libcluster: never autoreap from remote peer cache
      Refactor: crmd: abstract handling of stonith cleanup list
      Fix: crmd: remote nodes can't be DC so never add them to stonith cleanup list
      Fix: crmd: do not accept remote peer data as enough for startup
      Fix: libcluster,crmd: never update join info or processes for remote nodes
      Fix: attrd: don't try to communicate with attrd on a remote node
      Fix: fencing: don't poke peers when remote node status changes
      Refactor: libcib,libcluster: use defined constants for XML attributes
      Refactor: fencing: functionize stonith notifications
      Feature: fencing,libfencing: add API support for targeting topology by pattern or attribute
      Feature: stonith_admin: add ability to target topology by pattern or attribute
      Test: fencing: add regression test for targeting topology by node name pattern
      Test: fencing: add regression test for using commas and semicolons as pcmk_host_list delimiters
      Fix: crmd: use 300s timeout with notification scripts, not 300ms
      Merge pull request #817 from krig/quickref-update
      Merge pull request #819 from wenningerk/doc_proposal
      Merge pull request #820 from gao-yan/finalize-all-ops
      Merge pull request #821 from kgaillot/doc
      Merge pull request #822 from gao-yan/systemd-dbus-reconnect
      Merge pull request #823 from gao-yan/services-op-return
      Merge pull request #825 from gao-yan/services-op-inflight
      Merge pull request #824 from gao-yan/systemd-loadunit-error
      Merge pull request #826 from gao-yan/cleanup-anonymous-cloned-group
      Fix: libservices: restore build without dbus
      Merge pull request #828 from kgaillot/build
      Merge pull request #829 from liu4480/master
      Merge pull request #830 from liu4480/master
      Refactor: attrd,crmd,libcluster: add enum CRM_GET_PEER_ANY
      Refactor: libcluster: add new internal function for setting uname for peer cache entry
      Refactor: crmd: functionize choosing CIB call options
      Refactor: crmd: remove three inaccurate comments
      Fix: crmd: Consider only fence and shutdown transition events to be "down" events
      Feature: libcrmcommon: add new API call for adding a boolean XML attribute
      Merge pull request #834 from HideoYamauchi/master
      Fix: crmd: Don't update cluster/peer CIB fields when processing stonith results
      Merge pull request #837 from gao-yan/ping-metadata-host_list
      Merge pull request #840 from liu4480/master
      Merge pull request #839 from oalbrigt/dlm-wait-fencing-fix
      Fix: resources: allow for top output with or without percent sign in HealthCPU
      Merge pull request #842 from kgaillot/health
      Merge pull request #838 from wenningerk/upstream_pull
      Fix: Build: expand git alias in GNUmakefile
      Doc: Pacemaker Explained, Pacemaker Remote: update for 1.1.14
      Feature: crmd: make built-in notifications conditional at compile-time
      Build: update version information for 1.1.14
      Doc: update change log for 1.1.14
      Merge pull request #844 from kgaillot/1.1
      Fix: crmd: ensure compilation works with built-in notifications disabled
      Doc: update change log for 1.1.14-rc2
      Merge pull request #845 from kgaillot/1.1
      Merge pull request #846 from wenningerk/fix_stonith_attribute_delete
      Merge pull request #848 from HideoYamauchi/1.1
      doc: Pacemaker Explained: clarify some options
      Build: GNUmakefile: get correct line count when doing "make summary"
      Build: travis: update documentation URL
      Doc: update change log for 1.1.14-rc3
      Merge pull request #850 from kgaillot/1.1
      Low: crmd: re-enable built-in notifications
      Merge pull request #851 from kgaillot/master

Klaus Wenninger (5):
      DC expands to Designated Controller in most places - make consistent
      Fix CLBZ#5253: stonithd: Do not intermingle stdout & stderr coming from
      don't try to loop through non-existand node-list and return proper error
      cts brings an RA that needs systemd-python
      Fix RHBZ#1287315: stonithd: Trigger cib_devices_update in case of deletion of just an attribute

Kristián Feldsam (1):
      Added snmp trap of type "NOTIFICATION-TYPE" to MIB and snmp helper script

Kristoffer Grönlund (3):
      Doc: pcs-crmsh-quick-ref: crmsh also has the ban command
      Doc: pcs-crmsh-quick-ref: Add information about features of crmsh
      Doc: pcs-crmsh-quick-ref: Update TOC

Madison Kelly (digimer) (1):
      Added a missing word (as reported by 'The_Ball' in #clusterlabs on freenode.net).

Oyvind Albrigtsen (1):
      Return error when DLM status is wait fencing

Richard B Winters (11):
      Fix spelling error in binary libtransitioner
      Fix spelling error in binary pacemakerd
      Fix spelling error in binary libcrmservice
      Fix spelling error in binary libpe-status
      Fix spelling error in binary libpengine
      Fix spelling errors in binary libpengine
      Fix spelling error in binary crmd
      Fix spelling error in binary crmd
      Fix spelling error in binary stonithd
      Fix spelling errors in binary crm_resource
      manpage-has-bad-whatis-entry ocf_pacemaker_remote 7

Tomas Jelinek (1):
      Doc: pcs-crmsh-quick-ref: Add information about pcs commands

bin.liu (3):
      add "try except" to deal with errors that raw_input gets EOFError and add "--yes" to skip interaction
      crm_resource --show-metadata drops documentation strings for fencing agents
      remove a dup line from function fail_pending_op

liu4480 (1):
      change the stack from openais (white-tank) to corosync (plugin v0) in set_stack of environment.py

-----------------------------------------------------------------------

No new revisions were added by this update.

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-ha/pacemaker.git



More information about the Debian-HA-Commits mailing list