File resource-agents.changes of Package resource-agents

Tue Apr 02 10:50:22 UTC 2019 -

- LVM-activate returns the correct value on initial probe (bsc#1114855)
  * 0007-LVM-activate-return-OCF_NOT_RUNNING-on-initial-probe.patch 

Mon Feb 18 13:06:33 UTC 2019 -

- Update to version 4.2.0+git63.0f7987fb:
  * clvm: support exclusive mode
  * Revert " Added support for multiple slaves to pgsql."
  * azure-lb: remove reference to status from usage
  * configure: add Python library detection
  *  Added support for multiple slaves to pgsql.
  * LVM-activate: make vgname not uniqe
  * LVM-activate: fix dmsetup check
  * LVM-activate: fix bashism [[
  * Route: make family parameter optional
  * LVM-activate: dont fail initial probe
  * docker: Fix issues with stop operation
  * rabbitmq-cluster: always use quiet flag for eval calls
  * aws-vpc-move-ip: multiple VPC routing tables in routing_tables parameter - included comment in metadata section (bsc#1125138)
  * aws-vpc-move-ip: multiple VPC routing tables in routing_tables parameter - adjusting sleep time (bsc#1125138)
  * aws-vpc-move-ip: new feature: include support for multiple routing tables (bsc#1125138)

Thu Dec 06 08:49:05 UTC 2018 -

- Update to version 4.2.0+git33.69a622fd:
  * rabbitmq-cluster: better ensure node attributes are removed
  * Do not log at debug log level when HA_debug is unset
  * Enable --query flag in DescribeRouteTable API  (#1269)
  * Moved more Code to the Legacy section and cleaned the Legacy code as we already know that we are on pre-1.0 Version
  * Add lxc-stop Timeout
  * Remove Parameter and replace with Version Check
  * lxc: Add Support for lxc-stop
  * rabbitmq-cluster: fix regression in rmq_stop
  * rabbitmq-cluster: debug log detailed output when mnesia query fails
  * SAPDatabase: Improved the documentation about HANA usage

- Remove 0001-Revert-Add-additional-info-to-SAPDatabase-RA.patch

Thu Nov 01 12:45:58 UTC 2018 -

- Update to version 4.2.0+git8.7c750bab:
  * heartbeat/{docker,podman,rkt}: nicer treatment of "reuse" parameter
  * Maint: fix several typos discovered with "make spellcheck"
  * Fix demote example log line.

Mon Oct 22 08:13:35 UTC 2018 -

- Included the missing azure-events RA needed for maintenance on the
  Azure Public Cloud. (bsc#1112334)
- Included the library used by the azure-events RA (bsc#1112334)

- Update to version 4.1.1+git24.9b664917:
  * apache: fix return value for silent_status
  * ZFS: cleanup different indentations
  * Fix spelling error in gcp-vpc-move-ip/route
  * Initial version of the azure-events RA
  * rabbitmq-cluster: fail monitor when node is in minority partition
  * ZFS agent: check status without locks when possible
  * Mid: pgsql: Fix to ignore Master's re-promote.
  * Low: oracle: print more appropriate log when dbopen was failed
  * Low: oracle: accept any boolean value for clear_backupmode option
  * nfsserver: run prepare_directory after bind_tree and stop var-lib-nfs-rpc_pipefs.mount to fix issues when changing shared_infodir
  * CTDB: fix --logging/--logfile version string comparison (bsc#1102935)
  * Do not use the absolute path in redis' pidof calls
  * awseip: Documentation updates to prerequisites
  * nfsserver: only mount rpc_pipefs if it's not mounted
  * nfsserver: mount based on rpcpipefs_dir variable
  * LVM-activate: fixes
  * IPv6Addr/nagios/sybaseASE: add missing "s"-suffix in metadata
  * Add gcp-pd-move python script
  * Medium: Squid: Use ss if netstat is not available
  * Medium: portblock: Use ss or netstat (partial)
  * Medium: garbd: Use netstat or ss
  * Mid: apache: Retry pid check.
  * timeout/interval: add "s" suffix where it's missing
  * Low: Xen: Set utilization for cpu and hv_memory
  * pgsql: create replication slots after promoting master
  * VirtualDomain: add stateless support

- Add 0001-Revert-Add-additional-info-to-SAPDatabase-RA.patch

Tue Sep 07 15:57:39 UTC 2018 -

- Created the symlink for the suse:aws-vpc-move-ip RA for backward 
  compatibility reasons. The package aws-vpc-move-ip will be dropped
  and the Upstream resource agent will be the only one maintained.
Tue Sep 04 15:57:39 UTC 2018 -

- Update to version 4.1.1+git165.1ddbf85d:
  * Add Python library
  * gcp-vpc-move-vip: use Python library
  * gcp-vpc-move-vip: moved alias-parameters to top of metadata
  * gcp-vpc-move-route: use Python library
  * Python: add logging.basicConfig() to support background logging
  * CTDB: fix incorrect db corruption reports (bsc#1101668)
  * CTDB: explicitly use bash shell
  * CTDB: fix OCF_RESKEY_ctdb_recovery_lock validation (bsc#1088692)
  * aws-vpc-move-ip: Use ip utility to check address
  * rabbitmq-cluster: retry start when cluster join fails
  * ipsec: Add tunnel fallback option
  * LVM: fix missing dash (bsc#1104900)
  * Remove output redirection to file in /tmp
  * Medium: mysql: Remove obsolete DEBUG_LOG functionality (bsc#1021689)
  * lvmlockd: add cmirrord support
  * Initial podman support
  * LVM-activate: return OCF_ERR_CONFIGURED for incorrect vg_access_mode
  * spec: fix some overkil path-based Requires, trailing whitespace, thinko
  * rabbitmq-cluster: get cluster status from mnesia during monitor
  * aws-vpc-move-ip: Fix broken shell quoting
  * Filesystem: Support symlink as mountpoint directory
  * aliyun-vpc-move-ip: improve metadata and manpage
  * aws-vpc-move-ip: Check the rounting table also during monitor probe action ((bsc#1106707)
  * aws-vpc-move-ip: Force text output during awscli call ((bsc#1106707)
  * aws-vpc-move-ip: Included the address param for backward compat. Evaluate if the address param is set and if the ip para is null, and set the ip param with the address value. ((bsc#1106707)
  * pgsql: Avoid the change of /dev/null to postgres owner/group The check_log_file performs a -f test on the logfile which is set to /dev/null by default, and it returns 1 for non-regular files. (bsc#1090882)
  * Filesystem: Canonicalize mountpoint symlinks

Thu Aug 09 10:05:28 UTC 2018 -

- Update to version 4.1.1+git118.9a4fbd24:
  * SAPInstance: Improved SAP instance profile detection (bsc#1096744)
  * findif: only match lines containing netmasks
  * SAPInstance: Improved indents (bsc#1096744)
  * SAPInstance: implemeted reload method (bsc#1096744) 
  * build: add missing manpages
  * aliyun-vpc-move-ip: fix manpage (fate#326557)
  * aliyun-vpc-move-ip: fixes. (fate#326557)
  * Add Aliyun vpc-move-ip agent. (fate#326557)
  * minor fixes
  * move stackdriver parameter
  * filter call to aggregatedList
  * add deprecation message
  * python implementation of
  * Do not call ocf_attribute_target in the meta-data function
  * sfex: fixes to avoid errors with latest gcc
  * minor fixes
  * implement validate-all
  * minor fixes
  * manage ip alias
  * CI: skip Python agents in shellcheck
  * configure: add Python path detection
  * Volume groups and logical volumes "-" in their name get mangled with double dashes in dmsetup.  Switching to wc and just counting lines while depending on the vgname + lvname filter in the dmsetup call gets around the issue with dmsetup outputting correctly but grep failing due to the name mangle.
  * Mid: IPaddr2: Change return code.
  * ra-dev-guide: update instructions for GitHub
  * sybaseASE: bash path-detection
  * sybaseASE: new resource agent
  * gcp-vpc-move-ip: check specificed gcloud binary in validate

Wed Jun 27 07:35:01 UTC 2018 -

- Unify source package with SLE 12 / Leap 42:
  * High: oracle: Fix alter user syntax for set_mon_user_profile (bsc#1089279)
  * Medium: Raid1: Ignore transient devices after stopping a device (bsc#1077416)
  * Low: Raid1: remove unnecessary wait flags (bsc#1077416)
  * Medium: aws-vpc-route53: Fix tempfile race (bsc#1059312)
  * High: aws-vpc-route53: Add agent for AWS Route 53 (fate#322781)
  * VirtualDomain: Properly migrate VMs on node shutdown (bsc#1074014)
  * Low: sg_persist: Read empty value when when no attribute exists (bsc#1048288)
  * High: aws-vpc-route53: Add agent for AWS Route 53 (fate#322781)
  * Medium: Raid1: Handle case when mddev is a symlink(bsc#1047991)
  * Low: ocf-shellfuncs: Avoid printing empty INFO messages (bsc#1053621)
  * Low: SAPInstance: Mention monitor support in documentation (bsc#1051913)

- Obsoletes the following patches (SLE 12 / Leap 42):
  * 0001-High-oracle-Fix-alter-user-syntax-for-set_mon_user_p.patch
  * 0005-Medium-CTDB-Use-logging-syntax-for-ctdbd-bsc-981863.patch
  * 0007-Low-sg_persist-Read-empty-value-when-when-no-attribu.patch
  * 0008-High-aws-vpc-route53-Add-agent-for-AWS-Route-53-fate.patch
  * 0009-Medium-Raid1-Handle-case-when-mddev-is-a-symlink.patch
  * 0010-High-CTDB-Don-t-fail-on-empty-directory-bsc-1052577.patch
  * 0011-Low-ocf-shellfuncs-Avoid-printing-empty-INFO-message.patch
  * 0012-Low-SAPInstance-Mention-monitor-support-in-documenta.patch
  * 0013-High-galera-Backport-galera-fixes-from-upstream-bsc-.patch
  * 0014-VirtualDomain-Properly-migrate-VMs-on-node-shutdown-.patch
  * 0015-Low-aws-vpc-route53-Fix-missing-ocf_log-in-log-messa.patch
  * 0016-High-aws-vpc-route53-Add-agent-for-AWS-Route-53-fate.patch
  * 0017-Low-aws-vpc-route53-Fix-missing-ocf_log-in-log-messa.patch
  * 0018-Medium-aws-vpc-route53-Fix-tempfile-race-bsc-1059312.patch
  * 0019-Dev-aws-vpc-route53-Eliminate-duplicated-code.patch
  * 0020-Dev-aws-vpc-route53-Clean-up-logging.patch
  * 0021-Low-Raid1-remove-unnecessary-wait-flags-bsc-1077416.patch
  * 0022-Medium-Raid1-Ignore-transient-devices-after-stopping.patch

Wed Jun 20 08:14:25 UTC 2018 -

- Update to version 4.1.1+git70.00ae6a17:
  * Medium: pgsql: Fix incorrect SQL is selected with PostgreSQL 11 or later.
  * log the error message in abnormal_end() function even when we are in 'stop' operation so it is clear from logs on how did we end up here
  * Medium: SAPInstance: Add monitored services for ENSA2 (bsc#1092384)
  * Squid: cleanup description for squid_port parameter
  * ldirectord: add manpage to service documentation
  * Medium: Add openstack resource agents
  * findif: improve IPv6 NIC detection
  * awseip: fix allocation_id not found error
  * awsvip: get secondary-private-ip more precisely
  * awsvip: get network-id from meta-data directly
  * syslog-ng: added note that qdisk_dir is only supported in PE6
  * syslog-ng: merge Premium Edition 6 and 7 code into main agent
  * send_arp: dont use "-Wcast-align" due to false-positive fail on ARM.
  * gcp-vpc-move-ip: fix bash path-detection
  * enable chap authentication for lio-t iscsi target implementation. if attribute 'incoming_username' is set: enable attribute 'authentication', disable attribute 'generate_node_acls' and set chap 'userid' and 'password' for each 'allowed_initiator'
  * Low: exportfs: Fix inconsistent whitespace in exportfs_monitor
  * Medium: exportfs: Fix square bracket stripping in clientspec
  * Low: VirtualDomain: fix warning messages in log
  * New RA that manage IP moving in GCP

Mon Apr 16 09:02:40 UTC 2018 -

- Require /usr/bin/logger (provided by util-linux-systemd) 

Tue Mar 06 13:02:41 UTC 2018 -

- Update to version 4.1.1+git0.5a1edf2b:
  * High: mpathpersist: Fixed problem that character string could not be correctly parsed (bsc#1083896)
  * High: VirtualDomain: Properly migrate VMs on node shutdown (bsc#1074014)
  * Medium: awseip/awsvip: improvements (incl multi NIC support)
  * Medium: awseip/awsvip: increase default "api_delay" to 3s to avoid failures
  * Low: pgsql: try to create stats_temp_directory
  * Low: Filesystem: Add support for cvfs
  * Low: pgsql: improve validations for replication mode

Wed Feb  7 18:44:49 UTC 2018 -

- Edits to description, and compact build commands.

Wed Feb 07 13:31:54 UTC 2018 -

- Update to version 4.1.0+git83.6c77ce7e:
  * Medium: Raid1: Ignore transient devices after stopping a device (bsc#1077416)
  * redis: add support for tunneling replication traffic
  * Low: ovsmonitor: Generate man page for ovsmonitor (fate#323276)
  * Low: Raid1: remove unnecessary wait flags (bsc#1077416)
  * Pacemaker 2.0 compatibility (fate#324508):
    * Low: Xen: Use "crm_attribute --name/-n" instead of deprecated "--attr-name"
    * Low: Xen,redis: Use "crm_attribute --query/-G" instead of deprecated "--get-value"
    * Low: db2,galera,NodeUtilization: Use "crm_attribute --quiet/-q" instead of deprecated "-Q"
    * Low: sfex_daemon: Use "crm_resource --node/-N" instead of deprecated "--host-uname/-H"
    * Low: Xen,NodeUtilization: Use "crm_attribute --node/-N" instead of deprecated "--node-uname/-U"
    * Low: vmware: Do not reference the dropped cluster property "default_action_timeout"

- Remove patch merged upstream:
  * 0001-Drop-python-from-build-system-since-nothing-uses-it.patch 

Wed Jan 17 15:51:22 UTC 2018 -

- Drop python requirement (bsc#1076413)
  * 0001-Drop-python-from-build-system-since-nothing-uses-it.patch 

Wed Jan 17 15:45:47 UTC 2018 -

- Use python3 (bsc#1076413) 

Fri Jan 12 14:33:12 UTC 2018 -

- Update to version 4.1.0+git59.a0452207:
  * lvmlockd: auto change lvm config to use lvmlockd (fate#323822)
  * LVM-activate: put all validations into lvm_validate (fate#323822)
  * mariadb: GTID/semi-sync based MariaDB Master/Slave resource-agent

Tue Jan 09 10:13:13 UTC 2018 -

- Update to version 4.1.0+git54.b264e401:
  * mpathpersist: New agent (fate#324044)
  * jira: New resource agent
  * azure-lb: New resource agent
  * lxc-info/machine-info: New agents for monitoring containers/machines
  * Low: redis: Check read bit on redis config
  * Low: IPsrcaddr: Only check for ifconfig on BSD/solaris
  * Medium: CTDB: Cope with deprecated "idmap backend" smb.conf option
  * Medium: oracle: Fix alter user syntax for set_mon_user_profile
  * Medium: db2: improve monitor and simplify STANDBY/.../DISCONNECTED
  * Medium: CTDB: fix initial probe
  * Medium: ethmonitor: add intel omnipath support
  * Medium: LVM-activate: avoid unnecessary validation in stop action
  * Medium: IPsrcaddr: match exact route to avoid failing

Thu Nov 23 11:42:29 UTC 2017 -

- Update to version v4.1.0rc1~git8.2900c751:
  * Drop libnet dependency (bsc#1069596)
  * db2: fix HADR promote when master failed
  * Add resource agent for IPSEC tunnels
  * Mid: apache: Move the stop judgment last.
  * Mid: apache: Eliminate wasteful TERM transmission, correct the waiting time correctly.
  * galera: recover from empty gvwstate.dat
  * Revert "LVM: use vgscan --cache to update metadata during start/relocate"
  * Medium: IPaddr2: do not send Gratuitous ARPs in the Cluster IP configuration
  * Medium: IPaddr2: add send_arp_opts parameter
  * Medium: IPaddr2: allow to choose arping instead of send_arp binary
  * Low: send_arp.linux: print proper usage
  * Low: IPaddr2: document arp_interval as deprecated
  * Low: IPaddr2: remove metadata of an unused parameter arp_mac
  * Low: IPaddr2: always remove orphan pidfile
  * Low: IPaddr2: use ocf_log for an output from send_arp

Fri Nov 03 10:37:33 UTC 2017 -

- Update to version 4.0.1+git.1509705432.020b5b34:
  * LVM-activate: add new RA for LVM activation (fate#323822)
  * nfsserver: allow stop to timeout
  * adding docker-native healthcheck query to docker

Wed Nov 01 08:56:02 UTC 2017 -

- Update to version 4.0.1+git.1508859013.2b3cb4a6:
  * Medium: aws-vpc-route53: Fix tempfile race (bsc#1059312)
  * High: pgsql: Support PG10: Following the change of the PostgreSQL function name.
  * High: pgsql: Support PG10: Following the change of the default action of "pg_ctl start" and "pg_ctl promote".

Mon Sep 04 12:25:39 UTC 2017 -

- Update to version 4.0.1+git.1504527822.722ffb5b:
  * High: ovsmonitor: Add OpenvSwitch monitor agent (fate#323276)

Thu Aug 31 19:31:11 UTC 2017 -

- Update to version 4.0.1+git.1504207605.22ac330c: (bsc#1053207)
  * galera: Honor "safe_to_bootstrap" flag in grastate.dat (bsc#1055017)
  * galera: Fix instance name in master_exists() (bsc#1056635)

Thu Aug 31 09:06:09 UTC 2017 -

- Update to version 4.0.1+git.1504170332.70fbcd8a:
  * Medium: ocf-shellfuncs: improve locking (ocf_take_lock)
  * Low: ocf-shellfuncs: Avoid printing empty INFO messages (bsc#1053621)
  * Low: SAPInstance: Mention monitor support in documentation (bsc#1051913)
  * High: lvmlockd: add resource agent for lvmlockd daemon (fate#323822)
  * High: rkt: Add agent for managing rkt containers
  * Medium: varnish: Compatibility for version 4.0

Mon Aug 28 08:07:15 UTC 2017 -

- Update to version 4.0.1+git.1503904762.b4582bdb:
  * Low: mysql: Fixed bug where crm_admin is never called, leaving master scores to -1 in certain conditions.
  * Low: exportfs: check fsid parameter for not containing a comma
  * Low: mysql: don't run ocf_is_ms check in a subshell
  * Low: redis: add check for redis-check-rdb
  * Low: redis: wait for pid file to appear after start
  * High: pgsql: Allow dynamic membership
  * Medium: CTDB: fix issue with --logfile being replace with --logging
  * Medium: oraasm: new resource agent for Oracle ASM Disk Groups
  * Medium: rabbitmq-cluster: Backup and restore policies
  * Low: Raid1: Clean up spelling and whitespace
  * Medium: LVM: use vgscan --cache to update metadata during start/relocate
  * Medium: LVM: warn when cache mode is not writethrough
  * Medium: iSCSILogicalUnit: add emulate parameters
  * Medium: mysql: Properly Detect read_only State
  * Medium: anything: Allow multiple instances of binfiles to be run (pidfile will be unique)
  * Medium: anything: Start should return OCF_RUNNING if monitor succeeds
  * Medium: anything: Attempt to fix missing PID directory before starting
  * Medium: kamailio fixes and enhancements for v5.0
  * Low: portblock: dont log dd "0+0 records in/out"
  * Medium: Raid1: Handle case when mddev is a symlink
  * Low: sg_persist: Read empty value when when no attribute exists (bsc#1048288)
  * Medium: VirtualDomain: new attributes migrateuri, remoteuri, migration_user
  * Low: sapdb: Add hdbnameserver to monitor services
  * High: CTDB: Don't fail on empty directory (bsc#1052577)
  * High: minio: Resource agent for the minio storage server
  * High: ZFS: Add ZFS Resource Agent
  * High: awseip/awsvip: fixes and improvements
  * Medium: VirtualDomain: new configuration option shutdown_mode
  * Medium: Route: Resource key for the address family (ip4, ip6 or detect)
  * Low: aws-vpc-route53: Fix missing ocf_log in log message

- Remove patches merged upstream:
  * Remove 0005-Medium-CTDB-Use-logging-syntax-for-ctdbd-bsc-981863.patch
  * Remove 0008-Low-sg_persist-Read-empty-value-when-when-no-attribu.patch
  * Remove 0010-High-CTDB-Don-t-fail-on-empty-directory-bsc-1052577.patch
  * Remove 0009-High-aws-vpc-route53-Add-agent-for-AWS-Route-53-fate.patch
  * Remove 0007-Medium-Raid1-Handle-case-when-mddev-is-a-symlink.patch

Mon Aug  7 12:52:17 UTC 2017 -

- High: CTDB: Don't fail on empty directory (bsc#1052577)
  * Add 0010-High-CTDB-Don-t-fail-on-empty-directory-bsc-1052577.patch 

Sun Jul 16 07:26:04 UTC 2017 -

- High: aws-vpc-route53: Add agent for AWS Route 53 (fate#322781)
  * Add 0009-High-aws-vpc-route53-Add-agent-for-AWS-Route-53-fate.patch
- Low: sg_persist: Read empty value when when no attribute exists (bsc#1048288)
  * Add 0008-Low-sg_persist-Read-empty-value-when-when-no-attribu.patch

Fri Jul 14 06:58:47 UTC 2017 -

- Medium: Raid1: Handle case when mddev is a symlink(bsc#1047991)

Wed May 17 22:04:17 UTC 2017 -

- Update to version 4.0.1+git.1495055229.643177f1:
  * High: SAPInstance: Add IS_ERS parameter (bsc#1036486)
  * High: DB2: fix HADR support for DB2 V98+ (bsc#1035470)
  * Medium: iSCSILogicalUnit/iSCSITarget: protect targetcli invocations with locks
  * Medium: iSCSILogicalUnit: lio-t IPv6-support
  * Medium: LVM: add check_writethrough parameter
  * Medium: galera: fix permission of temporary log file for mariadb 10.1.21+
  * Medium: galera: fix the first bootstrap when cluster has no data
  * Medium: docker: fix to deal with the image name correctly.
  * Medium: named: Add support for rndc options in named resource agent.
  * Medium: NodeUtilization: Add NodeUtilization agent
  * Medium: rabbitmq-cluster: Allow the rabbitmq cluster to work on pacemaker remote nodes
  * Medium: galera: fix master target during promotion with cluster_host_map
  * Medium: IPaddr2: add option for specifying IPv6's preferred_lft
  * Medium: docker: Allow callers to specify a set of directories that should be created if the don't exist
  * Low: pgsql: fix regex in set async mode
  * Low: added kamctl parameter as it was missing, causing issues if kamctl is not in PATH
  * Low: clvm: remove reload action from metadata
  * Low: iSCSILogicalUnit: replace openssl call with md5sum
  * Low: nginx: fix return code for configfile err
  * Add 0006-Revert-ocf_log-use-same-log-format-as-pacemaker.patch

Wed Feb 01 18:43:15 UTC 2017 -

- Update to version 4.0.0+git.1485962683.3b5faae:
  * Also backup users permissions. Skip guest user permission for vhost /
  * Low: slapd: Change of the stop processing.
  * LVM: add error messages for partial_activation
  * Low: apache: Reduce log spam for success
  * Low: Correctly redirect output to /dev/null
  * Low: eDir88: Fix incorrect shell usage
  * Low: redis: Fix error in log output
  * Low: tomcat: Fix incorrect shell usage
  * Low: rabbitmq-cluster: Fix incorrect shell usage
  * Low: Filesystem: Fix incorrect shell usage
  * Low: nginx: Fix incorrect shell usage
  * Low: ServeRAID: Fix incorrect shell usage
  * Low: postfix: data_dir loop could only ever run once
  * Low: SysInfo: Convert test series to legal syntax
  * tomcat: fix systemd tomcat start script check
  * galera: make last commit parsing compatible with mariadb 10.1.18+
  * galera: prevent promote right after demote
  * ifcheck must not suppress stdout of "ip" command so that it can be printed out in case of an error (e.g., usage in findif_check_params).
  * Medium: IPaddr2: add option to enable sending refresh arp packets in monitor
  * Low: IPaddr2: Log refresh arp packets at debug level instead of info
  * pgsql: fix problem with creating replication slot names, when node count greather then 2
  * Low: Update outdated mailing list/wiki references
  * Enable dump/restore users from RabbitMQ ver. 3.6.x
  * Don't run scriptlets if Mnesia isn't available
  * Don't call erlang:halt/0 inside of the working node
  * Remove file only if successfully imported users
  * Dump users only if non-empty
  * Improve dump/restore ofusers' permissions
  * Restore users' permissions only if Mnesia is running
  * RabbitMQ: Reset Mnesia before joining existing cluster
  * portblock: Use -w (wait) to avoid "insufficient privileges" error
  * portblock: version check for -w
  * Double-quote value of synchronous_standby_names in rep_mode.conf
  * specfile: make requirements general and update comments
  * Filesystem: check if there are any mounts mounted under the mountpoint
  * IPaddr2: return success on stop with invalid IP
  * Fix typo in IPaddr2
  * Fix grammar and spelling in some resource agents
  * galera: allow names in wsrep_cluster_address to differ from pacemaker nodes' names
  * IPaddr2 detect duplicate IP
  * redis: use "CLIENT KILL type normal" to notify clients of master being demoted
  * specfile: Add metadata.rng to the file to be collected
  * Adds support for fstype=zfs on Filesystem resource agent
  * LVM: fix partial activation
  * galera: revert "long SST monitoring" PR #684 and #762
  * Medium: exportfs: Only strip brackets from edges of clientspec (bsc#1022223)

- Rebase patches:
  * Add 0001-LVM-Default-partial_activation-to-true-in-SLE-HAE.patch
  * Add 0002-nfsserver-fix-path-to-sm-notify.patch
  * Add 0003-ldirectord-don-t-create-subsys-lock.patch
  * Add 0004-Revert-Low-build-Move-binaries-in-usr-lib-heartbeat-.patch
  * Add 0005-Medium-CTDB-Use-logging-syntax-for-ctdbd-bsc-981863.patch
  * Remove lvm-vg-partial-active.diff
  * Remove resource-agents-nfsserver-sle11.patch
  * Remove no-var-lock-subsys.patch
  * Remove 0006-Revert-Low-build-Move-binaries-in-usr-lib-heartbeat-.patch
  * Remove 0001-Medium-CTDB-Use-logging-syntax-for-ctdbd-bsc-981863.patch

Wed Nov 02 17:29:36 UTC 2016 -

- Update to version 3.9.7+git.1478106349.5ff08d4:
  * tomcat: Add force_systemd parameter to rely on systemd service
  * nfsserver: dont stop rpcbind, as other services might use it
  * ldirectord: a further fix for spurious reminder e-mails
  * Medium: docker: Reduce monitor log spam
  * Medium: mysql-proxy: Fix incorrect conditional reported by shellcheck
  * Medium: garbd: Fix broken probe check
  * Doc: docker: Clarify limitations of monitor_cmd in parameter description

Tue Nov 01 17:56:45 UTC 2016 -

- Update to version 3.9.7+git.1477655995.de4be97:
  * adjusting to work with HANA Multi-Tenant Databases (bsc#1007142)
  * Medium: docker: Use docker exec for monitor_cmd if supported (bsc#1007867)
  * sg_persist: fix matching of hex node ids
  * sg_persist: do not spill info output to stderr
  * fix status clustered to check the specified logical volume
  * Delay: set default startdelay lower than start timeout to avoid it timing out before starting with default values
  * Low: Filesystem: Clearer exit reason if mount fails

Tue Oct 18 09:29:39 UTC 2016 -

- Update to version 3.9.7+git.1475827855.a95ea74:
  * High: nfsserver: Delete the temp file correctly.
  * Low: nfsserver: Prevent unnecessary log output

Tue Oct 18 09:27:32 UTC 2016 -

- Revert send_arp to libnet version (bsc#985486) 
  * For infiniband support in IPaddr2, install ipoibtools

Thu Oct 06 07:46:38 UTC 2016 -

- Update to version 3.9.7+git.1475676827.4af1003:
  * High: rabbitmq-cluster: forget stopped rmq nodes with notify=true
  * [rabbitmq-cluster] Backup and restore rabbitmq users during resource restart
  * RabbitMQ: New error code - 69
  * RabbitMQ: Forget node before 2nd joining attempt
  * rabbitmq: Allow automatic cluster recovery before forcing it
  * oracle: fix MONPROFILE to use monprofile parameter when set
  * oracle: fix issue with C## in monprofile (bsc#1005424)
  * oracle: add quotes for monuser and monpassword
  * Medium: pgsql: change stop_escalate parameter as usable value
  * Low: pgsql: adjust stop_escalate time when it is longer than the timeout
  * [galera] Add /etc/default/clustercheck importation
  * Medium: ldirectord: Fix the issue in IPv4/IPv6 mixed environment
  * ldirectord: Do an initial check of all services in when in 'starting' status
  * Low NFS: add more default init scripts to check
  * exportfs: Add pseudo resource factor (bsc#978680)
  * exportfs: Add IPv6 support
  * Low: extend support for AWS elastic ip association
  * Low: Rename ra-dev-guide.txt to ra-dev-guide.asc
  * garbd: Introduces garbd resource-agent

- Drop merged patches:
  * Remove 0001-Medium-ldirectord-Remove-IPv6-workaround-bsc-977193.patch
  * Remove 0001-exportfs-Add-pseudo-resource-factor-bsc-978680.patch

Thu Aug 11 14:44:20 UTC 2016 -

- Use non-libnet version of send_arp for IB support (bsc#985486) 

Thu Jun 30 09:58:40 UTC 2016 -

- Remove unused conditions from the spec-file as we build only on
  new enough distributions
- Sort all variables with spec-cleaner for better readability
- Enable monitoring subpkg as we now have all the needed dependencies
- Convert to pkgconfig style dependencies where applicable

Tue Jun 28 15:54:26 UTC 2016 -

- exportfs: Add pseudo resource factor (bsc#978680)
- Add 0001-exportfs-Add-pseudo-resource-factor-bsc-978680.patch

Mon May 30 11:12:24 UTC 2016 -

- Medium: ldirectord: Remove IPv6 workaround (bsc#977193)
- Medium: CTDB: Use --logging syntax for ctdbd (bsc#981863)
- Add 0001-Medium-ldirectord-Remove-IPv6-workaround-bsc-977193.patch
- Add 0001-Medium-CTDB-Use-logging-syntax-for-ctdbd-bsc-981863.patch 

Fri May 20 09:16:27 UTC 2016 -

- Remove redundant %clean section. Ensure systemd-tmpfiles is run.

Tue May 03 11:28:40 UTC 2016 -

- Update to version 3.9.7+git.1461938976.cb7c36a:
  + High: managing "Elastic IP" for EC2 instances
  + High: managing "Secondary Private IP" for EC2 instances
  + Low: VirtualDomain: fix unnecessary error when probing nonexistent domain
  + Low: mysql: Handle non-standard mysql server port
  + Low: Fix protential error for RA:nagios
  + Medium: iSCSILogicalUnit: Fix when using rbd store.
  + Low: iSCSILogicalUnit: Add tgt_bsopts parameter.
  + Low: Don't send Inaccessible real server e-mails in fork=yes mode

Tue Apr 26 15:24:19 UTC 2016 -

- Update to version 3.9.7+git.1461679689.18a78f5:
  + Medium: symlink: Handle missing directories in target (bsc#973054)
  + oracle: "shutdown immediate;" is needed after cleanup to be able to recover from the ORA-01081 error
  + oracle: inform user that monprofile must start with C## for container databases
  + tickle_tcp: Fix "Failed to open raw socket (Invalid argument)" issue

Mon Mar 21 09:03:00 UTC 2016 -

- Update to version 3.9.7+git.1458512170.48350a9:
  + High: SAPDatabase: Add support for Oracle 12c (bsc#967380)
  + High: evaluate parameter DBOSUSER also in start and stop methods (bsc#971914)
  + Low: SAPInstance: Update version support statement (bsc#965872)
  + Low: apache: ensure readable $TESTCONFFILE
  + Added iSER enabled portal support
  + quote vgchange options
  + Added pgagent RA
  + galera: fix monitoring of joining node for long running SST
  + galera: don't bootstrap from a node with no grastate.dat when possible
  + oralsnr: fix status check fail when username is more than 8 characters long

Fri Feb 19 09:41:38 UTC 2016 -

- Update to version 3.9.7+git.1454950092.6f56933:
  + Dev: ocft: include runocft to the tarball

Wed Feb 03 14:48:30 UTC 2016 -

- Update to version 3.9.7+git.1454497075.e697f43:
  + Medium: nfsserver: fix monitor for systemd
  + galera: force crash recovery if needed during last commit detection
  + galera: prevent recovered nodes from bootstrapping cluster when possible
  + galera: remove bashism
  + Add portal check to open_iscsi_get_session_id()

Thu Jan 28 14:08:43 UTC 2016 -

- Update to version 3.9.7~rc1+git.1453889774.3446b99:
  + Low: ldirectord: Fix unset failcount error (bsc#962795)

Mon Jan 18 07:16:20 UTC 2016 -

- Update to version 3.9.6+git.1452867140.fc8ace0:
  + SAPDatabase: Add Oracle 12 to list of supported databases (bsc#953991)
  + High: clvm: fix issue with only first option of daemon_options being used
  + High: mysql: fix grep failure on MySQL5.6 or higher when checking read_only variable
  + Medium: sg_persist: remove uncalled for ocf_run calls
  + Medium: ocf_shellfuncs: suppress bash specific trace_ra log on dash
  + Medium: ocf-rarun: replace error log messages with calls to ocf_exit_reason
  + Medium: nfsserver: do redhat specific stuff on redhat
  + Medium: iscsi: replace error log messages with calls to ocf_exit_reason
  + Medium: oracle/oralsnr: replace error log messages with calls to ocf_exit_reason
  + Medium: pingd: replace error log messages with calls to ocf_exit_reason
  + Medium: Raid1: replace error log messages with calls to ocf_exit_reason
  + Medium: pgsql: replace error log messages with calls to ocf_exit_reason
  + Medium: exportfs: don't increment fsid for single directory
  + Medium: Filesystem: add overlay as supported filesystem
  + Medium: mysql: When mysql has been already stopped, mysql is finished normally.
  + Low: pgsql: fix to be used valid charcters when the invalid characters is used in replication_slot_name.
  + Low: mysql: fix missing variables
  + Low: mysql: tmpfile is leaking when set the OCF_CHECK_LEVEL to 10
  + Low: nfsserver: prevent error messages on platforms without rpcuser (again)
  + Low: pgsql : fix invalid log message
  + Low: apache-conf: handle parsing includes with values in quotes
  + Low: pgsql: properly show default values in meta-data
  + Low : Changed the check handling of pid to the common function.
  + Low : Changed it that the message which pid file deleted repeated and did not appear
  + Low: pgsql: fix grep failure when using pacemaker 1.1.12
  + LVM: Allow vgck failures if partial
  + LVM: Correct grammar in partial_activation descriptions
  + IPaddr2: Add IPv6 DAD collision detection
  + VirtualDomain: New resource attributes migration_speed and migration_downtime
  + Nagios Resource Agent
  + Fix 90s wait/killing of databases containing the name of the database being killed, and added cleanup code to kill remaining listener process.
  + wait up to startup_wait seconds before failing if mysqld startup is slow

- Remove patches that have been merged upstream
  - Remove 0001-nfsserver-Add-EXEC_MODE-for-systemd-without-nfs-lock.patch 
  - Remove 0002-nfsserver-Use-rpc-statd.service-for-NFS-locking-in-E.patch 

Tue Nov 24 22:33:25 UTC 2015 -

- nfsserver: Use rpc-statd.service for NFS locking in EXEC_MODE=3 (bsc#955114)
- Add 0002-nfsserver-Use-rpc-statd.service-for-NFS-locking-in-E.patch 

Fri Nov 20 10:14:30 UTC 2015 -

- nfsserver: Add EXEC_MODE for systemd without nfs-lock.service (bsc#955114)
- Add 0001-nfsserver-Add-EXEC_MODE-for-systemd-without-nfs-lock.patch 

Wed Oct 14 06:19:17 UTC 2015 -

- Update to version 3.9.6+git.1444655379.5f44ac6:
  + send_arp: fix for infiniband, re-merge from upstream iputils arping
  + Medium: docker: Correction of the mistake of the variable name.
  + Low: s/ist/is typo fix
  + Low: syslog-ng: set default binary to /sbin/syslog-ng
  + galera: start joining nodes during 'monitor' to allow long-running SST
  + galera: document the bootstrap flow

Wed Sep 16 07:21:58 UTC 2015 -

- Update to version 3.9.6+git.1442374860.7f3628a:
  + Low: pgsql: set_master_score use an undefined variable($target).
  + Fix exec_sql errors like "unknown variable select pg_ " on ubuntu
  + Low: Allow + and - as broadcast arguments (bsc#940931)
  + Low: IPaddr2: Describe + and - broadcast arguments (bsc#940931) (#656)
  + galera: add support for MYSQL_HOST and MYSQL_PORT from clustercheck

Tue Sep 08 08:44:21 UTC 2015 -

- Update to version 3.9.6+git.1441596198.a62fc5a:
  + High: clvm: activate_vgs option for enable/disable of automatic vg activation
  + High: IPsrcaddr: return correct error code during stop when misconfigured
  + Fixed redis start operation if replication sync takes > 20 seconds
  + Made 'slave of XXX' wait forever (demote)
  + Low: pgsql: eliminate duplicate "ocf_exit_reason".
  + Docker: remove trailing spaces
  + Docker: fix docker existance test report error
  + High: redis: avoid 0 byte dump.rdb start failures
  + High: dhcpd: use correct default chroot for rhel based systems
  + interpret listener stop results correctly
  + Support RozoFS
  + Low: pgsql: replace "output=su ..." to a new function, exec_sql()
  + Medium: pgsql: delete an old replication slot when creating the slot.
  + iSCSILogicalUnit: add check for leftover target/core entries for lio-t
  + iSCSILogicalUnit: return OCF_NOT_RUNNING on monitor if backing path does not exist
  + iSCSILogicalUnit: when deleting a LUN or initiator fails with lio-t, proceed with warning
  + iSCSITarget: properly create portals for lio-t implementation
  + more accurate search results when tag is not specified
  + Fix "docker images" difference between distro
  + more strictly image check for repository, name, and tag
  + Use runuser inteald of su command for SELinux enforcing mode
  + Use runuser command instead of su in order to run selinux enforcing mode.
  + Medium: CTDB: Preserve permissions on smb.conf (bsc#935253)
  + High: redis: use required client pass word when set
  + Added "maxfiles" parameter for increasing maximum number of open files (for ulimit -n)
  + Low: update contact info
  + pgsql: Fix return code override in pgsql_real_start()
  + [high] Fix redis password parser

Fri May 08 17:36:27 UTC 2015 -

- Update to version 3.9.6+git.1430920897.b7ea7f9:
  + High: galera: make sure to clear last known sequence number promote is even attempted
  + High: galera: properly redetect bootstrap after demote
  + High: ethmonitor: link_status_only option for skipping RX counter and arping tests

Tue Apr 21 06:11:26 UTC 2015 -

- Update to version 3.9.6+git.1429568091.f79322c:
  + Medium: multiple: make sure that the pidfile directory exist
  + Fix process name grep in exit_idle
  + Low: pgsql: fix check_wal_receiver to prevent incorrect "ERROR" status display and output WARN log in the master
  + Low: allow rgmanager lvm agent to handle hostname aliases
  + Fix rmq_join_list to only return online nodes
  + Medium: pgsql: Support replication slots
  + Low: pgsql: add validation check for replication slot.
  + Low: redis: smarter config argument default
  + Fix: redis: reliable shutdown.
  + Low: redis: loosen advertised default monitor timeout seconds
  + Fix: redis: do not attempt to demote if redis is dead
  + Low: redis: make sure to always delete master score on stop
  + Low: redis: reconnect to new master after promotion
  + High: redis: only connect to active master instances
  + High: redis: wait_last_known_master option for redis agent.
  + Low: redis: prevent bash syntax errors and lower priority some log messages
  + High: galera: retrieve last sequence number without using read-only mode
  + Fix return code in asterisk_monitor #2

Tue Apr 07 13:45:11 UTC 2015 -

- Update to version 3.9.6+git.1427133197.6897c9c:
  + High: introducing rabbitmq clustering agent
  + Filesystem: fix misspelled log messages
  + Filesystem: support overlayfs
  + Medium: sg_persist: do use the binary parameter
  + Medium: dnsupdate: use the advertised nsupdate_opts
  + build: add missing RAs to makefiles
  + Fix: nfsserver: avoid re-tagging nfs sysconfig with autogen header
  + nfsserver: merge options into existing sysconfig/nfs
  + portblock: portno param can be a string like 137,138
  + portblock: replace ancient heartbeat config with crm configure
  + portblock: clarify TCP RST vs ICMP port unreachable
  + portblock: capitalize "Linux"
  + portblock: ports can't be "turned off"
  + portblock: fix typo
  + Medium: VirtualDomain: enforce C locale in force_stop
  + Low: VirtualDomain: Fixed to correctly get the return code of the "virsh destroy" command
  + redis: retry on unknown error when starting
  + bash bug fix, no arbitrary stop timeout, placeholder master during election period
  + build: update ChangeLog for 3.9.6
  + Medium: VirtualDomain: enforce C locale in status
  + CTDB: Change default socket location to CTDB's expected default.
  + CTDB: Whitespace cleanup.
  + Medium: VirtualDomain: add migrate_options parameter
  + nginx: Allow different URLs for level 10 and 20 monitor
  + build: update ChangeLog for 3.9.6-rc1
  + Low: for rgmanager nfs agent, strip off trailing '/' from path option
  + Allow MySQL to run as user other than mysql
  + Fix missed detection of write failure
  + Low: fixes rgmanager pgsql agent's ability to detect user group
  + Modify manual for negotiatetimeout
  + Medium: iscsi: run iscsi discovery only when necessary
  + High: VirtualDomain: save the config before virsh undefine (bnc#891915)
  + RA and ocft for sg_persist
  + ldirectord: Add Install section for systemd unit file.
  + Use calculated default pidfile.
  + Removed 'pidof' because it's not cross-platform compatible
  + Using 'check_binary'
  + ldirectord: modify manual and sample for Radius check
  + ldirectord: Fix can not remove virtualservice when stopping ldirectord via systemctl.
  + Use the most specific matching route
  + Low: LVM: volgrpname is unique (meta-data update)
  + Medium: ldirectord: Get correct user for sending email (bnc#910497)
  + Low: remove POSIX-incompatible command
  + Low: add variables OCF_RA_VERSION_MAJOR, OCF_RA_VERSION_MINOR, OCF_EXIT_REASON_PREFIX OCF_RESOURCE_TYPE, and OCF_RESOURCE_PROVIDER in ra-dev-guide.txt based on pacemaker source
  + Low: add a few more references to notify in ra-dev-guide.txt
  + #526, do not set write_back when creating backstore/block
  + Medium: portblock: Feature: reset_tcp_on_unblock_stop
  + Low: nfsserver: prevent error messages on platforms without rpcuser
  + Medium: Xen: fix regression with xm and quoting (lf#2671)
  + re-add LXCpre1.0 logic
  + remove check for unused binary
  + fix for missing lxc-ps command in ocf:heartbeat:lxc
  + Low: Route: some unique attributes are not unique
  + Log domain status in debug mode for xl and xen-list
  + Medium: IPaddr2 findif: accept dotted quad netmask
  + Low: mysql: document use of notify metadata attr for master/slave
  + Medium: VirtualDomain: For Xen, prefer xl to xen-list (boo#901453)
  + Medium: Xen: Use xl list $domain return code in status check (boo#901453)
  + Low: pgsql: PostgreSQL 9.3 compatibility for unix_socket_directories

- Remove merged patches:
  + findif-multiple-routes-bnc892755.patch
  + fix-sg_persist-commented-byLars.patch
  + fix-sg_persist-demote-error.patch
  + fix-sg_persist-devs-required-and-not-defined.patch
  + fix-sg_persist-for-code-normalization.patch
  + resource-agents-new-sg_persist.patch
  + sg_persist-ocft-not-configure-iscsi.patch

Thu Dec  4 06:50:35 UTC 2014 -

- fix fix-sg_persist-devs-required-and-not-defined.patch         
  to eliminate build warning: script-without-shebang  

Fri Nov 21 11:30:30 UTC 2014 -

- Update to upstream cs v3.9.5-631-g992ca06
  - High: Introduces the docker agent to manage docker containers in an HA environment
  - High: mysql: report error when validation fails during monitor yet pid is still active
  - High: mysql-common: avoid use of check_binary in common validation function.
  - Medium: ocf-shellfuncs: set HA_LOGD depending on HA_use_logd
  - Low: VirtualDomain: downgrade error msg during probe
  - Modified replication_user description: RELOAD privilege is required for RESET SLAVE or RESET SLAVE ALL command.
  - Only strip tag on stop if we are owner
  - Low: pgsql: PostgreSQL 9.3 compatibility for unix_socket_directories
  - Medium: Xen: Use xl list $domain return code in status check (boo#901453)
  - Medium: VirtualDomain: For Xen, prefer xl to xen-list (boo#901453)
  - Low: mysql: document use of notify metadata attr for master/slave

Thu Oct 16 08:21:25 UTC 2014 -

- Update to upstream cs v3.9.5-589-gb644395
  - High: iSCSILogicalUnit: fixes syntax errors
  - High: galera: do not ignore specified check_password
  - Fix: shellfuncs: fix syntax error caused by exit_reason support for dash shell.
  - Fix: ocf_exit_reason: implicit format string "%s" for single argument version
  - Fix: ha_log: drop global __ha_log_ignore_stderr_once hack
  - $((var++)) breaks in the dash shell

Tue Sep 16 06:23:00 UTC 2014 -

- Update to upstream cs 5652c04 (bnc#889914)
  - fix output redirection of 'anything' resource
  - Medium: mysql: fix unexpected operation error that caused by MySQL client timeout.
  - Middle: anything: Prevent stop failure, even if the job takes time to stop.
  - High: doc: Add pcs to man page example section
  - Remove use of non standard command usleep
  - High: ethmonitor: exit reason support
  - High: Exit reason string support updates
  - Medium: Build: add --compat-habindir option for HA_BIN backward compatibility
  - Low: iSCSITarget: Support exit reason string
  - High: iSCSITarget: monitor operation must pass before start is considered complete
  - Low: iSCSILogicalUnit: support ocf exit reason string
  - High: iSCSILogicalUnit: monitor operation must pass before start is considered complete
  - Merge pull request #482 from davidvossel/iscsi-updates
  - High: iSCSILogicalUnit: Improves the resiliency of LIO iSCSILogicalUnit
  - Low: iSCSILogicalUnit: fix inconsistent formatting
  - High: Filesystem: when loading kernel modules wait for filesystem to initialize
  - High: ocf-shellfuncs: Remove action_filter from ocf_exit_reason parameters
  - High: ethmonitor: add infiniband status monitoring support
  - Medium: Xinetd: refine exit codes when xinetd is stopped
  - Low: ethmonitor: Add pcs example with crmsh example in metadata
  - Low: ethmonitor: Improve logging during monitor operation to help determine reason for failure
  - High: clvm: automatically set lvm.conf's locking_type=3
  - High: VirtualDomain: exit reason support
  - High: IPaddr2: support ocf_exit_reason
  - High: exportfs: support exit reason string
  - High: nfsserver: support exit string in nfsserver agent
  - High: nfsnotify: set exit reason strings in nfsnotify agent
  - High: ocf-binaries: have 'check_binary' set exit reason
  - High: ocf-shellfuncs: Log exit reason to log facilities as well as stderr
  - Revised IPv6 workaround code #379
  - Low: ethmonitor: Use consistent formatting
  - Low: ethmonitor: Increase recommended default timeout settings
  - Low: ethmonitor: Perform first monitor during start action
  - Low: nfsserver: only set rquotad options when port is set
  - separate lio and lio-t start
  - add targetcli support to iSCSI resource agents
- Remove merged patch: xen-replace-xm-with-xl.patch
- upstream cs: v3.9.5-578-g5652c04

Thu Aug 28 08:18:51 UTC 2014 -


Wed Aug 27 14:35:29 UTC 2014 -

- Disable monitoring-plugins-metadata

Tue Aug 26 12:34:00 UTC 2014 -

- Add patch findif-multiple-routes-bnc892755.patch
  - Use the most specific matching route (bnc#892755)

Thu Aug 14 15:19:21 UTC 2014 -

- drop dependency on monitoring-plugins-udp since it is folded into

Thu Aug  7 07:00:28 UTC 2014 -

- Rename nagios-plugins-metadata to monitoring-plugins-metadata (bnc#890337)

Tue Aug  5 16:33:20 UTC 2014 -

- Medium: Xen: Properly quote domain name in call to create
- Updated patch: xen-replace-xm-with-xl.patch 

Tue Aug  5 13:58:13 UTC 2014 -

- Low: remove process path from process search results
- Low: Dummy: add sample usage of 'ocf_exit_reason'
- High: shellfuncs: set exit error reason string
- Medium: oracle: fix setting monitor user profile for other languages (nls)
- Low: LVM: Warn users about the danger of lvmetad
- Low: nfsnotify: fixes error message output
- Low: nfsserver: fix usage of rquatad_port option
- Low: nfsserver: warn user when nfs port options can not be set
- Low: nfsserver: fixes issue with proc/fs/nfsd not being mounted during start on some distros
- Low: ocf-shellfunctions: handle local nodename when version has distro postfix
- Low: Update meta-data to reflect required arguments
- Low: mysql: Refactoring mysql logic into common library that can be shared with galera agent
- Medium: oracle: reset MONUSR password if it expired
- High: oracle: try as sysdba if monitor as MONUSR fails
- High: Do not fail during stop if listener has already stopped
- Low: Fix exit_idle loop to exit process wait loop correctly
- Low: the exit_idle function was a no-op because of a malformed grep search
- Low: oralsnr: ignore listener case during grep for pid
- dropped merged patches:
 - 0001-High-oracle-Require-monitoring-user-as-parameter-bnc.patch
 - 0001-Medium-VirtualDomain-Try-xen-list-if-no-emulator-is-.patch
 - 0001-Medium-mysql-Update-mysql-with-more-sensible-default.patch
 - apache-fix-bnc884674.patch
- upstream cs: v3.9.5-508-g4f369d8

Tue Aug  5 11:03:31 UTC 2014 -

- Update license to include GPL-3.0+ 

Tue Aug  5 07:15:07 UTC 2014 -

- Reintroduce nagios plugins metadata (bnc#890337)
  - Add nagios-plugins-metadata.tar.bz2

Mon Jul 21 15:41:32 UTC 2014 -

- Revised fix for apache RA on SUSE distributions (bnc#884674)
- Add apache-fix-bnc884674.patch

Tue Jul  8 09:49:22 UTC 2014 -

- Medium: VirtualDomain: Try xen-list if no emulator is set (bnc#885292)
- Add 0001-Medium-VirtualDomain-Try-xen-list-if-no-emulator-is-.patch 

Tue Jul  8 09:15:55 UTC 2014 -

- High: Xen: Enable use of xl instead of xm if available (bnc#882548)
- Add: xen-replace-xm-with-xl.patch

Tue Jul  8 09:02:47 UTC 2014 -

- Medium: exportfs: allow multiple exports
- High: nfsserver: Support new NFS Active Passive and Active Active use-cases.
- Low: nginx: Fix incorrect parameter name
- Reintroduce Xen RA (bnc#882548)
- dropped merged patches:
  - remove 0001-ctdb-add-ctdb_rundir-parameter-and-create-on-startup.patch
  - remove 0001-Medium-VirtualDomain-Add-support-for-qemu-dm-as-emul.patch
  - remove apache_conditional_initd_check-bnc884674.patch
- upstream cs: v3.9.5-453-g9eb8c17d467e 

Fri Jul  4 13:11:21 UTC 2014 -

- Create the CTDB runtime state directory on startup; (bnc#885049).
  + added 0001-ctdb-add-ctdb_rundir-parameter-and-create-on-startup.patch

Fri Jul  4 08:03:12 UTC 2014 -

- Medium: VirtualDomain: Add support for qemu-dm as emulator (bnc#885292)
- add 0001-Medium-VirtualDomain-Add-support-for-qemu-dm-as-emul.patch 

Tue Jul  1 11:25:25 UTC 2014 -

- Medium: apache: Check before calling init.d script (bnc#884674)
- add apache_conditional_initd_check-bnc884674.patch

Wed Jun 25 08:51:06 UTC 2014 -

- High: conntrackd: allow probe to return OCF_RUNNING_MASTER
- High: Filesystem: Add force_unmount option
- High: Force kill processes with access to shared libraries on mount point
- High: VirtualDomain: restore advertised start and stop timeout values to a sane value.
- Exclude Xen resource agent (bnc#882548)
- Merged upstream:
  - Removed 0001-Low-ldirectord-fix-the-list-output-to-include-ops-fl.patch
  - Removed 0001-ldirectord-Add-systemd-unit-file-bnc-863250.patch
- Dropped (bnc#882548)
  - Removed xen-replace-xm-with-xl.patch
- upstream cs: v3.9.5-432-g07aeed7c9035

Sun Jun 15 12:59:03 UTC 2014 -

- Update xm/xl patch with additional fixes (bnc#882548)
- Modified: xen-replace-xm-with-xl.patch 

Fri Jun 13 08:55:07 UTC 2014 -

- High: Xen: Replace use of xm with xl (bnc#882548)
- Add: xen-replace-xm-with-xl.patch

Mon Jun  9 07:43:27 UTC 2014 -

- rgmanager/fs: typo preventing passing some mount opts
- add an optional option for rpc.statd
- Fix usage of findmnt to work for filesystems containing bind mounts
- addition of bind-mount resource agent for rgmanager
- add startup_wait option to rgmanager postgres agent
- upstream cs: v3.9.5-410-g6f1dffee54d1

Fri May 30 13:25:09 UTC 2014 -

- sg_persist: change sg_persist ocft not configure isci and let user prepare it manually
  add sg_persist-ocft-not-configure-iscsi.patch

Fri May 30 12:07:13 UTC 2014 -

- bnc#880726: sg_persist devs required and not defined when define by configuration file
  add fix-sg_persist-devs-required-and-not-defined.patch

Tue May 27 09:28:08 UTC 2014 -

- bnc#879881: fix sg_persit demote error
  add fix-sg_persist-demote-error.patch

Fri May 23 12:29:27 UTC 2014 -

- Updated 0001-ldirectord-Add-systemd-unit-file-bnc-863250.patch
  - Use absolute paths for touch and rm

- ldirectord: Add missing dependency on perl-IO-Socket-INET6

Fri May 23 08:08:03 UTC 2014 -

- added necessary macros for systemd files

Mon May 19 11:05:11 UTC 2014 -

- rgmanager/fs: typo preventing passing some mount opts
- add an optional option for the rpc.statd listening port
- iscsi: iscsi status fails with open-iscsi with support for flash (bnc#878039)
- nfsserver: let systemd service files handle lock notifications
- Monitor kvm resources without requiring libvirtd
- Calculate stop loop according stop timeout
- Added kamailio resource agent
- IPaddr2: fix error msg from manual clone/incarnation # era
- clvm: Properly detect directory for vg tools
- Add IPv6 http(s) health checking to ldirectord
- On stop, kill sm-notify so /var/lib/nfs/statd can be unmounted
- VirtualDomain: Allow monitoring of lxc domains without libvirtd
- Fix issues running named daemon as a non-root user
- VirtualDomain: cache domain emulator after start
- CTDB: do not fail monitor operation when ctdb socket does not exist
- exportfs: Make unlock_on_stop_default=1 (bnc#864263)
- Stateful: Clean up implementation of Stateful (bnc#867372)
- dnsupdate: resource agent for IP take-over via dynamic DNS
- upstream cs: v3.9.5-404-g61a2ccc8b7b1 

- Updated resource-agents-nfsserver-sle11.patch
- Removed 0015-Medium-exportfs-Make-unlock_on_stop_default-1-bnc-86.patch

Thu May  8 11:28:32 CEST 2014 -

- use _rundir macro 

Wed Apr  2 06:51:58 UTC 2014 -

-Fate#312345-add fix-sg_persist-commented-byLars.patch

Fri Mar 28 14:59:47 UTC 2014 -

- mysql: Update mysql with more sensible defaults (bnc#863844)

- added patches:
  * 0001-Medium-mysql-Update-mysql-with-more-sensible-default.patch
- removed patches:
  * resource-agents-mysql-sle11.patch 

Mon Mar 10 11:19:06 UTC 2014 -

- exportfs: Make unlock_on_stop default to 1 (bnc#864263)

- added patches:
  * 0015-Medium-exportfs-Make-unlock_on_stop_default-1-bnc-86.patch 

Thu Feb 27 16:32:21 UTC 2014 -

- ldirectord: Add systemd unit file (bnc#863250)
- ldirectord: fix the list output to include ops flag (bnc#847005)
- VirtualDomain: Fixes parsing domain name from xml file
- Add 'sec=' as a valid exports option
- CTDB: Detect default persistent database directory
- CTDB: Ensure logfile directory exists so CTDB can start
- VirtualDomain: Attempt to determine vm status even when libvirt
  is unavailable
- Add ability to set custom httpd binary
- findif: Update default value for best_netmask
- VirtualDomain: avoid running "virsh uri" if hypervisor is set
- VirtualDomain: check process table for qemu-system-* too
- mysql: handle %secs_behind = NULL
- logging: Fix syntax errors with logging functions in LVM and
  asterisk agents
- upstream cs: 935b92fd

- added patches:
  * 0001-Low-ldirectord-fix-the-list-output-to-include-ops-fl.patch
  * 0001-ldirectord-Add-systemd-unit-file-bnc-863250.patch
Tue Feb 25 17:10:16 UTC 2014 -

- oracle: Require monitoring user as parameter (bnc#850589, bnc#859683)

- added patches:
  * 0001-High-oracle-Require-monitoring-user-as-parameter-bnc.patch
- removed patches:
  * 0003-Medium-oracle-Create-profile-for-OCFMON-user-bnc-850.patch
  * 0004-High-oracle-Make-monitoring-user-configurable-bnc-85.patch

Mon Feb 24 03:33:46 UTC 2014 -

-Fate#312345-add fix-sg_persist-for-code-normalization.patch

Thu Jan 23 06:23:41 UTC 2014 -

- Remove 0001-Medium-CTDB-Default-ctdb-manages-samba-and-ctdb-mana.patch:
  Accidentally reintroduced obsolete patch
- Remove 0002-Medium-CTDB-Add-smb_fileid_algorithm-parameter-defau.patch:
  Accidentally reintroduced obsolute patch

Wed Jan 22 09:53:09 UTC 2014 -

- Add 0001-Medium-CTDB-Default-ctdb-manages-samba-and-ctdb-mana.patch:
  CTDB: Default ctdb manages samba and ctdb manages winbind to "yes",
- Add 0002-Medium-CTDB-Add-smb_fileid_algorithm-parameter-defau.patch:
  CTDB: Add smb_fileid_algorithm parameter, defaults=fsid (bnc#696978),
- Add 0003-Medium-oracle-Create-profile-for-OCFMON-user-bnc-850.patch:
  oracle: Create profile for OCFMON user (bnc#850589),
- Add 0004-High-oracle-Make-monitoring-user-configurable-bnc-85.patch:
  oracle: Make monitoring user configurable (bnc#859683),
- Add 0006-Revert-Low-build-Move-binaries-in-usr-lib-heartbeat-.patch:
  Revert move of heartbeat binaries to /usr/libexec
- ldirector: fix using service name instead of port number (bnc#836759)
- pgsql: Support for non-standard port and library locations
- Xen: retry domain lookup in repeating monitor and stop
- Xen: report error if the monitor does not find the guest
- VirtualDomain: support more virsh domstate output formats
- LVM: only remove missing PVs on stop when they belong to mirrors
- Raid1: disallow md raid arrays as clone resources to avoid data corruption
- Raid1: make clone prevention overridable similar to the Filesystem RA
- Filesystem: switch to use ocf_is_clone helper function
- varnish: Added support for ulimit -l and ulimit -n
- jboss: Monitor rotatelogs process and restart when it is stopped
- Pure-FTPd: Create pid directory if needed
- lxc: support up-to-date lxc-ps versions
- Fix process name grep in exit_idle
- Remove quotes around listener name
- oralsnr: Add tns_admin option for setting custom TNS_ADMIN path
- VirtualDomain: Fix ability to use default libvirt hypervisor without explicitly setting agent attribute
- VirtualDomain: Ensure it is possible to manage a libvirt domain defined outside of VirtualDomain
- nfsserver: nfsserver not starting due to missing etab file
- Fixes failure to unmount local fs when process runs with cwd inside fs mount
- upstream cs: 468ed4f8 

Fri Jan 17 07:57:38 UTC 2014 -

- fix a few lines for monitor

Mon Jan 13 09:10:46 UTC 2014 -

- Fate#312345-Add resource-agents-new-sg_persist.patch: RA for persistent
  SCSI reservation

Tue Sep 10 14:35:52 UTC 2013 -

- Filesystem: add corosync as HA_cluster_type
- eDir88: multiple IP support (bnc#)
- apache: better handling of not installed apache
- jboss: add jboss_base_dir parameter to support multi-instances
  on JBoss 6
- jboss: stop the JBoss daemon by sending a signal on JBoss 6
- jboss: add run_command parameter to change the start up script
- jboss: add jboss_version parameter to support JBoss 6
- tomcat: use root as the default for tomcat_user instead of
  RUNASIS mode
- tomcat: multiple tomcat instances based on CATALINE_BASE
- VirtualDomain: no longer define a domain, but use "virsh
  create" to start it non-persistent
- pound: add parameter maxfiles to set ulimit
- apache: remove unnecessary and imperfect checks from
  validate_all (bnc#827927)
- pgsql: set only one node into sync mode when using 3 nodes or
- Raid1: stop arrays even with block device file missing
- oracle/oralsnr: explicitly require bash (bnc#825517)
- Raid1: set MDADM_NO_UDEV appropriately if udev is not used
- Raid1: wait for udevd to settle (bnc#821861)
- iSCSILogicalUnit: add tgt specific parameters bstype, bsoflags,
- apache: Attempt graceful stop before -TERM signal
- LVM: Warn user if initrd is older than lvm.conf when using
  exclusive activation with tags
- LVM: Exclusive activation without clvmd using filtering with
- named: Attempt to autogen /etc/rndc.key using rndc-confgen tool
- mysql: really use log setting (bnc#823095)
- mysql: test properly for failed process start (bnc#823095)
- LVM: Retry exclusive activation after deactivating vg
- LVM: Verify setup on start plus verify exclusive activation is
- Route: make Route RA IPv6 ready
- drbd: remove deprecated drbd agent
- tools: send_arp.libnet: reuse ARP packets (deb#701914)
- ethmonitor: correctly detect when the network is unplugged.
- Raid1: do not test for device existence in the stop operation
- VirtualDomain: Support saving and restoring virtual machine
  snapshot state files
- nfsserver: Cleanup shared nfs dir mount
- nfsserver: Enable file locking daemon when systemd is being
- nfsserver: Maintain SELinux permissions on failover for nfs v3
  lock state
- nfsserver: Improve support for v3 file lock recovery
- nfsserver: Add systemd unit-file support when init scripts are
  not present
- apache: add pidfile directive when not present in config file
- VirtualDomain: Properly detect defined lxc domains
- iSCSILogicalUnit: add lio_iblock parameter.
- upstream cs: 3cafd1

Thu Aug  1 09:05:05 UTC 2013 -

- Run over format_spec_file service.

Tue Mar 26 15:43:56 UTC 2013 -

- ocf-shellfuncs: use $HA_VARLIB as default destination for trace
- VirtualDomain: allow for custom migrateport
- IPaddr2: support nic:iflabel format in nic parameter
- upstream cs: ddad46 (v3.9.5)

Tue Feb  5 17:13:36 UTC 2013 -

- ocf-shellfuncs: RA tracing
- IPaddr2: make sure that some ARP send program runs
- pgsql: add check_wal_receiver parameter
- pgsql: support starting as Hot Standby
- nfsserver: improve rpc.statd support
- nfsserver: add option -n for rpc.statd (bnc#794479)
- nfsserver: make the retry time for sm-notify configurable
- nfsserver: make sm-notify running in the foreground configurable
- exportfs: handle '<world>' exportfs embelishment for '*' (bnc#791690)
- jboss: set JAVA_OPTS correctly
- pound: use correct default for control-binary
- pound: No error when stopping pound and no pound is running
- Squid: support systems using IPv6
- Xinetd: do not fail in stop if the daemon is not running
- Xinetd: improve finding Xinetd process (thanks to Vadym Chepkov)
- SendArp: add background parameter
- SendArp: fix monitor
- ocft: print the actual case names when testing
- ocft: make only the updated test-case file
- ocft: add "incremental" mode (ocft test -i)
- include a copy of LGPL license file
- upstream cs: e265f0 (3.9.5 rc1+)

Wed Nov 28 16:57:11 UTC 2012 -

- zabbixserver: RA for zabbix servers management
- IPaddr2: use better test for infiniband (bnc#783353)
- IPaddr2: replace the findif binary by
- IPaddr2: add IPv6 support
- IPaddr2: really send arps in background if requested
- IPaddr2: use send_arp instead of ipoibarping if not available
- Raid1: discover block size for dd (bnc#781137)
- Xen: repair node_ip_attribute use
- SAPDatabase: improvement of cleanup of saphostctrl process
- SAPInstance: monitor-master not advertised (bnc#782482)
- SAPInstance: don't wait for timeout if the stop attempt failed
- SAPInstance: failes on sapcontrol message Unauthorized (bnc#782486)
- tomcat: Correction of the process alive monitoring.
- ldirectord: Added READDQUIESCENT parameter 
- Squid: fix getting PIDs of squid processes (lf#2653)
- portblock: use end-of-word instead of space at the end of the line
- named: use pgrep instead of searching in ps(1) output
- named: fix monitor if named_rootdir is set to /
- tools: add
- tools: add send_ua binary for IPv6 support in IPaddr2
- ocft: new test for Raid1
- upstream cs: 4890c9 (3.9.4)

Thu Sep 27 14:10:24 UTC 2012 -

- dhcpd: new RA to manage ISC DHCP servers
- pound: new RA for Pound HTTP/HTTPS reverse-proxy and load-balancer
- ocf-rarun: add the RA driver
- apache: create /var/run/apache2 if it doesn't exist
- conntrackd: rename parameter "conntrackd" to "binary"
- conntrackd: resync from other hosts in start
- exportfs: allow expanding the fsid parameter to produce correct
  exportfs options
- exportfs: cleanup exportfs cache on stop (bnc#770210)
- exportfs: don't grow /var/lib/nfs/rmtab indefinitely (bnc#754067)
- Filesystem: add nfs4 to the list of well known types
- Filesystem: allow to force cloning for local mounts
- Filesystem: fix determining if the device is a block device
- Filesystem: improve read/write checks for CHECK_LEVEL 10, 20
- Filesystem: include ceph in the list of non-blockdev filesystems
- Filesystem: remove a status file only when OCF_CHECK_LEVEL is set to 20
- Filesystem: support ceph
- IPaddr2: fix regression introduce in d93b5fd, nic=lo always "stopped"
- IPaddr: remove colon at the end of the interface name
- IPv6addr: always use the provided nic and cidr_netmask when specified
- iscsi: add try_recovery parameter
- iscsi: check session status in monitor
- iscsi: do discovery in start only
- iscsi: do not rely on iscsid.startup being set correctly (bnc#751783)
- iscsi: don't fail on stop of the iscsi server fails
- iscsi: use iscsiadm -m session -r in stop
- iSCSILogicalUnit: correctly match for target IQN and backing
  device name (iet and tgt)
- iSCSITarget: treat an empty "implementation" parameter specially
- jboss: implement rotating of console log
- ldirectord: precedence error with perl v5.8.8 in IPv6 code
- lxc: fix LXC_status to work with lxc-0.7.5 or later
- mysql: fix validation return codes
- mysql: improve replication support
- mysql: support 5.5 slave status message format
- mysql-proxy: add in-depth monitoring
- mysql-proxy: add the 'plugins' parameter
- mysql-proxy: create pid/socket directories if needed
- nfsserver: add a new parameter for the rpc_pipefs mount point
- nfsserver: don't run sm-notify in foreground (bnc#759616)
- nfsserver: NSM state files correctly populate
  /var/lib/nfs/sm.ha (bnc#684139)
- nfsserver: Support of multiple IP addresses (bnc#684143)
- oracle: connect as a regular user in monitor
- oracle: ignore password expiry warning in monitor (bnc#770250)
- oracle: improve managing IPC objects
- oracle: improve matching instance specific files and processes
- oracle, oralsnr: get rid of eval
- oralsnr: match only one (our) listener process
- pgsql: add option archive_cleanup_command
- pgsql: add option recovery_end_command
- pgsql: support for replication
- pgsql: support multi-instance in replication mode
- Raid1: manage multiple arrays
- Raid1: stop processes using raiddev
- SAPDatabase: add support for Sybase ASE and SAP HANA database
- SAPDatabase: new method for checking responsiveness of saphostexec
- SAPDatabase: make use of saphostagent
- sfex: sfex_init fails on a 64 bit big endian platform (lf#2647)
- slapd: always set the exit code correctly in monitor
- slapd: Gracefully handle config check during probe
- slapd: pass bind_dn correctly to ldapsearch
- tomcat: Correction of the time-out level of the stop processing.
- VIPArip: fix exit codes
- VIPArip: make start idempotent
- VirtualDomain: honor virsh "in shutdown" state
- VirtualDomain: if the configuration file is missing on stop
  exit with success
- Xen: add support for HVM ACPI graceful shutdown
- Xinetd: reimplementation
- Tools: send_arp.libnet: fix for big endian platforms (bnc#721334)
- Tools: SendArp: exit with the right code when not properly configured
- ocft: new tests for mysql-proxy, named, IPv6addr, oracle, Raid1, Xinetd
- upstream cs: 810c9b

Fri Jun 29 19:11:51 UTC 2012 -

- added resource-agents require to ldirectord as it's init script
  requires /etc/ha.d/shellfuncs which that package provides

Tue Feb 14 03:19:59 UTC 2012 -

- ocf-shellfuncs: fix loglevel variable scope in ha_log
- exportfs: fix monitor action for special characters and common
- apache: add support for IPv6 in monitor
- iscsi: don't fail in stop or monitor if the portal is
  ambiguous (bnc#740718)
- ocf-tester: add -d and -q options
- Medium: Tools: findif: Use most specific matching route (bnc#740738)
- ldirectord: Allow IPv6 for NNTP, requires perl-IO-Socket-INET6 (bnc#669412)
- SAPInstance/SAPDatabase: fix unique attribute in metadata (bnc#737673)
- IPv6addr: handle a link-local address properly in send_ua
- iSCSILogicalUnit: fix default for scsi_sn
- named: make monitor work with IPv6 interfaces
- slapd: exit with the right code if validation fails (it used to
  exit always with the generic error code)
- upstream cs: de3074

Thu Dec 15 08:17:33 UTC 2011 -

- Add /usr/lib/tmpfiles.d/resource-agents.conf to ensure temp dir
  exists when using systemd (bnc#734760)

Mon Dec 12 13:37:37 UTC 2011 -

- varnish: new resource agent
- LVM: force dmevent monitoring for clones
- Filesystem: repair the fast_stop parameter use (its value has
  always been false)
- slapd: prevent possible timeout in stop (wrong logic in stop)
- upstream cs: 72fdc8

Fri Nov 25 12:25:57 UTC 2011 -

- asterisk: new resource agent
- IPaddr: add back the local_start/stop_script code
- iscsi: proceed if iscsid is not running if iscsid.startup is
  present in iscsid.conf
- mysql: check mysql status more thoroughly before stopping
- jboss: add the java_opts parameter for java options
- VirtualDomain: add a functionality that modifies utilization of
  resource automatically
- Tools: ocft: new option Agent, corresponding configs update
- doc: Add RA developer's guide
- move NodeUtilization to Pacemaker
- upstream cs: 1f8f3d

Fri Nov 18 10:54:10 UTC 2011 -

- Implemetation of Dynamic utilization detection and setting.
  Based on FATE#310117 and FATE#310115.

Tue Nov  8 08:35:28 UTC 2011 -

- cleanup sfex_uuid.patch 

Mon Nov  7 14:43:09 UTC 2011 -

- tomcat: remove pidfile before start, it may prevent some tomcat
  releases from starting
- ocft: new options variable and cleanup-agent, var/unvar renamed
  to env/unenv, corresponding configs update
- upstream cs: 5599aa

Mon Oct 31 09:41:38 UTC 2011 -

- fate#311000, sfex_daemon accept "-u $uuid". When assiging $uuid
  as a parameter, $lock_index is ignored.

Mon Oct 24 13:40:47 UTC 2011 -

- slapd: wait in start for the service to come up
- apache: fix sysconfig includes & enable status for default SUSE conf
- upstream cs: 80b640

Mon Oct 10 13:39:35 UTC 2011 -

- named: new resource agent
- Tools: send_arp.libnet: fix for big endian platforms (bnc#721334)
- rsyslog: new RA to manage rsyslog servers
- slapd: new RA to manage OpenLDAP servers
- Xen: wait in migrate_from for the migration to finish instead
  of bailing out immediately
- conntrackd: test for socket existence in monitor instead of
  process grep
- postfix: multiple fixes and improvements
- upstream cs: 6c24ae

Tue Sep 20 14:51:43 UTC 2011 -

- Filesystem: add tmpfs to the list of supported filesystems
- exportfs: fix adding the fsid parameter to options
- pgsql: fix meta-data the config param type in meta-data
- IPaddr: don't search for the IP address in the route table on
- CTDB: Improve monitor op (check output of ctdb status, bnc#712192)
- CTDB: Set ctdb_start_as_disabled=no by default (bnc#712410)
- Filesystem: add support for glusterfs (lf#2620)
- ocf-shellfuncs: fix wrong logic introduced in commit 8468b5
  (fixes wrong cluster prefix in log messages)
- Filesystem: don't use direct dd option in monitor depth 20 for
  non-blockdevice fs
- CTDB: Add smb_fileid_algorithm parameter
- Implemetation of Dynamic utilization detection and setting.
  Based on FATE#310117 and FATE#310115.
- ocf-shellfuncs: add RA instance name to ocf_log/debug
- Raid1: support for multiple MD arrays, as specified in raidconf
- ethmonitor: new RA to monitor network interfaces
- conntrackd: new RA
- lxc: new RA to manage lxc linux containers
- symlink: new RA to manage symbolic links
- VirtualDomain: if there's no config exit with success on stop
- Tools: send_arp: fix the PID file location (bnc#702099)
- iscsi: fix support for open-iscsi version 2.0-872 (lf#2562)
- pgsql: don't check directories on probes
- db2: new implementation with master/slave mode
- exportfs: backup and restore rmtab to ensure smooth client
  failover on node failures
- CTDB: Allow stop to succeed when using pkill on ctdbd (bnc#695829)
- CTDB: Use op timeout when invoking ctdb client (bnc#697428)
- CTDB: Use 'chmod' instead of 'ctdb enablescript' (bnc#697152)
- mysql: --skip-slave-start option is default now
- mysql: set connect timeout to 10 seconds rather than 1 second
- mysql: keep replication state (prevents data loss on master reset)
- mysql: don't rely on state information from pacemaker, but
  check if the instance is in the read-only mode
- mysql: if test parameters are all set, assume OCF_CHECK_LEVEL=10
- mysql: support for master/slave for more than two nodes
- mysql: don't wait for replication to finish, when not replicating
- mysql: store replication state in separate attributes for each master
- pgsql: improve configuration check and probe handling
- VirtualDomain: correctly create migration URI when target is an FQDN
- VirtualDomain: properly wait until domain_name is non-empty
- ldirectord: add a support of "netmask" directive for IPv6
- ldirectord: fix fwmark behavior for IPv6
- ldirectord: ignore children in Net::DNS
- iscsi: add support for open-iscsi version 2.0-872 (lf#2562)
- postfix: issue error if 'postfix abort' failed
- postfix: improve exit codes on installation problems
- postfix: use monitor to test if postfix works after the start action
- ocft: fix make command for compatibility with mawk/Debian (lf#2600)
- ocft: test case for pgsql
- ocft: test case for postfix
- ocft: test case for iscsi
- ocft: new testcases for db2, Filesystem, and LVM
- doc: add man page for sfex_init
- doc: improve man pages output
- doc: add examples for master/slave resource agents
- doc: improve legibility of generated man pages
- ocf-tester: tolerate OCF_ERR_INSTALLED on probes and missing binaries
- update spec file to match the new autoconf setup
- add OCF_ROOT/lib/heartbeat directory for ocf-shellfuncs et al
- remove compatibility scripts /usr/lib/heartbeat/ocf-*
- oracle: improve oracle process list test (bnc#673027)
- upstream cs: 957d32 (v3.9.2)

Fri Aug  5 06:18:49 UTC 2011 -

- Fix rpmlint non-ghost-in-var-run check 
- ldirectord: Remove usage of /var/lock/subsys in init script

Wed Feb 16 12:54:08 CET 2011 -

- change Conflicts/Obsoletes for now dropped heartbeat subpackages 
  to Provides/Obsoletes

Wed Feb  9 12:12:43 UTC 2011 -

- SAPDatabase: Avoid continuous output to syslog in monitor with SAP
  7.20 and J2EE_ONLY=1
- SAPInstance,SAPDatabase: remove unnecessary usage of eval to start processes
- SAPInstance,SAPDatabase: Allow blanks in path of userexit script
- Add GPLv3 license file (bnc#655700)
- Xen: implement stop of a migrating domain (bnc#656227)
- tomcat: Use here-documents to simplify start/stop operations
- tomcat: Ensure name of tomcat resource is only used on start operation
  and expose JAVA_OPTS variable for use
- tomcat: Fix to ensure default OCF_RESKEY_xx values are observed
- tomcat: Add CATALINA_BASE parameter, defaults to CATALINA_HOME,
  permits multiple tomcat instances
- tomcat: Use Tomcat stop TIMEOUT -force to improve stop
- Dummy: migrate_from/to: correct OCF_RESKEY_CRM_meta_migrate_xxx
  variable names
- Dummy: make method reload work
- IPaddr2: Fix reference to Infiniband arping binary (bnc#668447)
- Upstream version cs: 41cd73a9dedb

Wed Feb  9 10:15:11 UTC 2011 -

- Fix perl-Net-SSLeay prerequisite

Tue Feb  8 12:11:10 UTC 2011 -

- ocf-shellfuncs: allow ocf_run to return the actual exit code
- build: install jboss
- nginx: new RA
- LVM: add partial_activation parameter (lf#2490)
- Filesystem: allow cloning of some filesystems as read-only (lf#2440)
- Filesystem: add fast_stop parameter (lf#2402)
- sfex: output log messages also to stderr in sfex_init
- sfex: add the sfex_stat command
- Xen: check the allow_mem_management boolean properly (bnc#637525)
- MailTo: don't check if user exists for email address (might be
  an alias or remote)
- CTDB: Remove hard-coded timeout on start op
- CTDB: Don't manage Samba and Winbind by default
- IPaddr2: optionally flush kernel routing table on interface stop
- IPaddr2: exit with the right code when not properly configured
- IPsrcaddr: exit with the right code when not properly configured
- Tools: findif: differentiate between error conditions
- exportfs: add unlock_on_stop parameter
- exportfs: ensure graceful failover with NFSv4 clients
- exportfs: correctly test for lease time file
- ManageVE: add migration capability
- anything: add the workdir parameter
- pgsql: cd to pgdata before running commands (fixes permission error)
- pgsql: add optional username, password, and sqlcode parameters for monitor
- db2: Replace call to db2_local_ps with db2nps
- db2: guard against a hanging db2stop by spawning this into the
  background. Use db2_kill after grace period.
- db2: add multi partition support
- SAPDatabase,SAPInstance: improve LD_LIBRARY_PATH processing (bnc#640026)
- SAPInstance: prevent premature expansion of [:upper:] [:lower:] when
  producing sidadm uid
- SAPInstance: Moved testing of SAP profile directory and START profile
  to a later stage (only when needed), for more robustness
- SAPInstance: fix return codes in probes
- SAPInstance: New parameter: SHUTDOWN_METHOD
- SAPInstance: ensure enqueue failover in monitor_clone on process failure
- SAPDatabase: prevent premature expansion of [:upper:] and [:lower:]
  when producing sidadm/orasid/db2sid uids
- SAPdatabase: Changed Oracle recovery method from "recover automatic
  database" to "end backup"
- SAPDatabase: Adapt process search pattern for DB/2 9.5
- SAPDatabase: start listener only if database processes are found
- ldirectord: http: connect to server instead of protocol (Debian#594958)
- ldirectord: add implicit support for submission RFC4409
- ldirectord: example configuration for a submission virtual service
- ldirectord: Shutdown write-side of client connection after writing has
- ldirectord: Add RPM Requires perl-IO-Socket-INET6
- Filesystem: Clarify metadata and improve non-clone warning
- IPaddr2: exit early and with the right code if the ip parameter is not
- ocf-shellfuncs: add option -q to ocf_run to suppress verbose logging
- mysql: if appropriate, connect to master on start
- mysql: if appropriate, delete slave config on start
- pgsql: suppress irrelevant output in the monitor operation
- pgsql: postpone getting the socketdir parameter which depends on the
- IPaddr2: add missing 5 to the validation string for the multicast MAC
- Xen: Allow node configurable attribute to specify which IP to use for
  live migration (bnc#628735)
- Xen: use xen-list command for status check if available (bnc#628735)
- Xen: use xen-destroy for stop, if available.

Sat Nov 20 17:07:43 UTC 2010 -

- build against libnet-devel on 11.4

Mon Aug  9 19:52:45 UTC 2010 -

- mysql: replace the deprecated -O option (bnc#620275)
- mysql: use replication credentials to test the replication status
- ldirectord: Oracle compatibility
- ldirectord: allow underscore in service name
- ocf-shellfuncs: handle properly syslog facility set to none
- ocf-tester: show output from the agent in case of error
- IPv6addr: interface index in /proc/net/if_inet6 may be longer than 2
  chars (lf#2462)
- Patch for bnc#623788 is now in mainline.
- Upstream version cs d206bc8f1303

Mon Jul 26 11:48:33 UTC 2010 -

- CTDB: Deprecate (and make optional) smb_private_dir param (bnc#623788)

Tue Jul  6 08:48:43 UTC 2010 -

- ldirectord: use $1 instead of \1 in pattern replace (bnc#605086)
- VirtualDomain: Fix spurious stop failures
- RA: mysql: Numerous fixes.
- Raid1: Handle stop for failed arrays properly (bnc#618775)
- Raid1: Support attempting to re-add mirrors on deep monitor action
- sfex: wait in the start and stop actions until sfex_daemon
- db2: support for v9.x instances (bnc#608952)
- ocf-shellfuncs: correctly identify root by id only (bnc#602312)
- ocf-tester: meta-data also should never be affected by missing
- ocf-tester: Extend to cover initial probe (monitor_0) test.
- Upstream version cs 9475e2f5d8db

Mon May 10 14:12:00 UTC 2010 -

- RA: set the HA_RSCTMP directory to /var/run/resource-agents (lf#2378)
- pgsql: add new "config" parameter
- pgsql: properly implement pghost parameter
- High: SAPInstance RA: don't rely on op target rc when monitoring clones (lf#2371)
- upstream version cs a7c0f35916bf

Mon Apr 26 14:54:24 UTC 2010 -

- exportfs: new RA
- mysql: master-slave functionality
- oracle: reduce output from sqlplus to the last line for queries (bnc#567815)
- nfsserver: fix the default string for the notification parameter
- upstream version cs f92935082277

Tue Apr 20 06:23:21 UTC 2010 -

- VGs are activated with "--partial". This helps the mirrored LVs.

Mon Apr 12 11:49:40 UTC 2010 -

- VirtualDomain: bail out early if config file can't be read
  during probe (nbc#593988)
- VirtualDomain: spin on define until we definitely have a domain name
- Filesystem: prefer /proc/mounts to /etc/mtab for non-bind mounts (lf#2388)
- upstream version cs b3d16560e966

Mon Mar 29 09:14:47 UTC 2010 -

- build: add the postfix RA
- VirtualDomain: fix incorrect use of __OCF_ACTION
  (the stop operation may timeout otherwise)
- IPaddr2: don't bring the interface down on stop
  (otherwise IPv6 addresses may be removed)
- oracle/oralsnr: improve exit codes if the environment isn't valid
- Route: don't assume that OCF_RESKEY_CRM_meta_clone_node_max is
  set to a number (lf#2375)
- meta-data: improve timeouts in most resource agents
  (reduce the number of warnings by the shell)
- SFEX daemon: fix logging
- ldirectord: fix the configfile default (bnc#589457)
- drbd: fix metadata (bnc#588684)
- mysql: revert wrong patch erroneously included with ocft
- mysql: update defaults to match sle11 installations (bnc#590447)
- upstream version cs 000dd8956ad7

Mon Mar 15 15:11:34 UTC 2010 -

- ocft: fix remote shell
- upstream version cs 72f1ac55e70b

Tue Mar  9 18:37:51 UTC 2010 -

- sfex: don't use pid file (lf#2363,bnc#585416)
- IPsrcaddr: modify the interface route (lf#2367)

Thu Mar  4 13:51:05 UTC 2010 -

- ocft: new RA test suite

Wed Mar  3 15:36:24 UTC 2010 -

- log messages from RAs to stderr if attached to a terminal
- ldirectord: Allow multiple email addresses (lf#2168)
- ocf-shellfuncs: tests to check for clone/ms resources
- vmware: multiple improvements

Mon Feb 22 14:30:08 UTC 2010 -

- upstream version cs: da18e46a1f94
- ocf-shellfuncs: don't output to stderr if using syslog (prevents
  double logging from the RA and lrmd)

Mon Feb 22 12:04:54 UTC 2010 -

- upstream version cs: 44b1ba8c7804

Mon Feb 22 11:57:39 UTC 2010 -

- make sure that OCF_RESKEY_CRM_meta_interval is always defined (lf#2284)
- vmware: fix set_environment() invocation (lf#2342)
- oracle/oralsnr: improve logging
- apache: return the right exit code from monitor (bnc#578628)
- iSCSILogicalUnit: fix monitor for STGT
- Route: add route table parameter (lf#2335)
- build: don't include PACKAGE_NAME in HA_NOARCHBIN

Fri Feb  5 14:12:13 UTC 2010 -

- upstream version cs: f40d03a863f1

Wed Jan 27 13:24:32 UTC 2010 -

- Use same replacement syntax as for other packages.

Fri Jan 22 16:06:06 UTC 2010 -

- upstream version cs: 39b26d5af509
- portblock: fast reconnect/tickle ACK (new resource agent)
- ClusterMon: don't fail in stop if the process is missing (bnc#569957)
- CTDB: auto-generate cluster-specific part of smb.conf (LF 2308)
- doc: add a "see also" link to resource agent man pages

Tue Jan 19 17:02:38 UTC 2010 -

- Correct dependencies of ldirectord package (bnc#571825).

Mon Dec 28 16:00:32 UTC 2009 -

- add %dir /usr/share/resource-agents to .spec

Mon Dec 28 15:37:41 UTC 2009 -

- Upstream version cd: b10ad6afb1bd
- mark obsolete RAs as deprecated (LF 2244)
- VirtualDomain: fix forceful stop (LF 2283)
- IPaddr2: CLUSTERIP/iptables rule not always
  inserted on failed monitor (LF 2281)
- copy specfile from upstream

Mon Dec 14 21:38:26 UTC 2009 -

- Removed wrong conflicts.

Thu Dec 10 17:06:50 UTC 2009 - ""

- CTDB: New resource agent for clustered samba.
- proftpd: new resource agent.
- Upstream version cs: 520d53bd91dc

Tue Dec  8 08:23:21 UTC 2009 - ""

- Filesystem: allow configuring smbfs mounts as clones.
- Filesystem: implement deep monitor operation.
- IPaddr2: behave if the interface is down (lf#2147)
- IPaddr2: check binaries when it makes sense
- IPaddr2: fix nvalid default value for OCF_RESKEY_clusterip_hash
- IPv6addr: recognize network masks properly
- LVM: Make monitor operation quiet in logs (bnc#546353)
- MailTo: Provide a default for MAILCMD (bnc#534803, bnc#556366)
- MailTo: allow multiple word subject line
- Raid1: Improve monitor function (bnc#546551)
- VIParip: Pathname needed to be configurable (lf#1331)
- VirtualDomain: avoid needlessly invoking "virsh define"
- VirtualDomain: loop on status if libvirtd is unreachable.
- Xen: Remove instance_attribute "allow_migrate" (bnc#539968)
- apache: monitor operation of depth 10 for web applications (lf#2234).
- iSCSITarget, iSCSILogicalUnit: support LIO
- jboss: Added JBoss support.
- ldirectord: fix various bugs in OCF RA (lf#1949)
- mysql: escalate stop to KILL if regular shutdown doesn't work
- nfsserver: exit properly in nfsserver_validate (lf#2173)
- nfsserver: use check_binary properly in validate (lf#2211)
- nfsserver: validate should not check if nfs_shared_infodir exists
- oracle/oralsnr: export variables properly
- oracle: drop spurious output from sqlplus
- pgsql: remove the previous backup_label if it exists
- postfix: fix double stop
- vmware: make meta-data work and several cleanups (lf#2212)
- Upstream version cs: 002038be601b

Tue Sep 15 09:01:09 UTC 2009 - ""

- Revert to SLE HA 11 naming scheme.
- iSCSILogicalUnit: use a 16-byte default SCSI ID.
- iSCSITarget: be more persistent deleting targets on stop.
- portblock: add per-IP filtering capability.
- mysql-proxy: new instance parameters supported.
- Upstream version cs: cde070764e1f

Thu Aug 27 17:00:00 UTC 2009 -

- IPaddr2: include netmask in search for the right interface
- IPaddr2: remove all colons from the mac address before passing it to
  send_arp (LF#2165)
- IPsrcaddr: replace 0/0 with proper ip prefix.
- IPv6addr: new nic and cidr_netmask parameters
- IPv6addr: supply checksum for ICMPv6 packets
- SAPDatabase + SAPInstance: New versions from SAP.
- Squid: make the regexp match more precisely output of netstat
- VirtualDomain: destroy domain shortly before timeout expiry
- apache: make sure that proxies are not used for monitor
- iSCSILogicalUnit: add support for SCSI ID, SCSI SN, Vendor ID, and
  Product ID
- iSCSILogicalUnit: add support for per-LU parameters
- iSCSILogicalUnit: set default for SCSI SN, truncate SCSI ID default to
  24 bytes
- iSCSITarget, iSCSILogicalUnit: add support for tgt
- iSCSITarget, iSCSILogicalUnit: identify targets by IQN, not by tid
- iSCSITarget: add support for CHAP authentication
- iSCSITarget: add support for restricting target access
- iscsi: replace wrong variable reference (bnc#499291)
- portblock: fix invalid exit codes on monitor
- postfix: new agent.
- Upstream version cs: 9700f962ce0a

Thu Aug 13 22:13:35 UTC 2009 -

- Include ldirectord (formerly known as heartbeat-ldirectord).
- Upstream version cs: 1263ac948158

Thu Aug 13 09:55:27 UTC 2009 -

- Obsolete heartbeat-resources package.

Wed Jul 29 12:00:00 UTC 2009 -

- Initial import of the resource-agents project after being split off
  from Heartbeat (this package replaces heartbeat-resources).
- Upstream version cs: 3aa338887920