Permalink
Commits on Jun 13, 2017
  1. Revert "Make ocf_is_ver() handle dist. specific versions"

    oalbrigt committed Jun 13, 2017
  2. Merge pull request #974 from oalbrigt/systemd-deps-target

    oalbrigt committed Jun 13, 2017
    systemd: add resource-agents-deps target
Commits on Jun 8, 2017
  1. Merge pull request #989 from davebiffuk/master

    oalbrigt committed Jun 8, 2017
    rabbitmq-cluster: typo fix
  2. rabbitmq-cluster: typo fix

    davebiffuk committed Jun 8, 2017
    fix a small typo which causes errors in corosync.log e.g.
    Jun 08 09:00:14 [6504] overcloud-controller-1.localdomain       lrmd:   notice: operation_finished:     rabbitmq_start_0:7504:stderr [ Error: syntax error before: ')' ]
Commits on Jun 1, 2017
  1. Merge pull request #986 from jackiehjm/pr-master

    oalbrigt committed Jun 1, 2017
    fs.sh: fix builds when srcdir and builddir are seperated
  2. fs.sh: fix builds when srcdir and builddir are seperated

    jackiehjm committed Jun 1, 2017
    It fails to find fs.sh.in when srddir and builddir are not the same:
    
    make[5]: Entering directory '/path/to/builddir/rgmanager/src/resources'
    cat fs.sh.in | sed \
        -e 's#@''LOGDIR@#/var/log/cluster#g' \
        > fs.sh.out
        cat: fs.sh.in: No such file or directorychmod +x fs.sh.out
        mv fs.sh.out fs.sh
    
    Add abs_srcdir to fix this.
    
    Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Commits on May 31, 2017
  1. Merge pull request #984 from oalbrigt/LVM-warn-when-cache-mode-not-wr…

    krig committed May 31, 2017
    …itethrough
    
    LVM: warn when cache mode is not writethrough
Commits on May 30, 2017
  1. Merge pull request #983 from lemenkov/rabbitmq-typo-fix

    krig committed May 30, 2017
    [rabbitmq] Typo fix
  2. [rabbitmq] Typo fix

    lemenkov committed May 30, 2017
    Unfortunately we introduced a regression with commit
    1f57e26. This patch addresses it and
    fixes #982.
    
    Thanks @seabres for the heads up.
    
    Signed-off-by: Peter Lemenkov <lemenkov@redhat.com>
Commits on May 26, 2017
  1. Merge pull request #950 from oalbrigt/oraasm

    krig committed May 26, 2017
    oraasm: new resource agent for Oracle ASM Disk Groups
  2. Merge pull request #963 from lemenkov/rabbitmq_preserve_policies

    krig committed May 26, 2017
    [RFE] Backup and restore policies
  3. Merge pull request #887 from jnpkrn/exportfs-fsid

    krig committed May 26, 2017
    Low: exportfs: check fsid parameter for not containing a comma
  4. Merge pull request #952 from vvidic/redis

    krig committed May 26, 2017
    redis: fixes for compatibility with redis 3.2
  5. Merge pull request #957 from oalbrigt/CTDB-fix-logging

    krig committed May 26, 2017
    CTDB: fix issue with --logfile being replace with --logging
  6. Merge pull request #975 from greemo/greemo-patch-1

    krig committed May 26, 2017
    Make ocf_is_ver() handle dist. specific versions
  7. Merge pull request #980 from oalbrigt/LVM-vgscan-cache

    krig committed May 26, 2017
    LVM: use vgscan --cache to update metadata during start/relocate
  8. Merge pull request #973 from krig/raid1-spellfix

    krig committed May 26, 2017
    Low: Raid1: Clean up spelling and whitespace
Commits on May 22, 2017
Commits on May 17, 2017
  1. Merge pull request #972 from krig/bsc1036486

    krig committed May 17, 2017
    Medium: SAPInstance: Add IS_ERS parameter (bsc#1036486)
Commits on May 14, 2017
  1. Merge pull request #978 from antblant/master

    krig committed May 14, 2017
    fix nginx resource
  2. fix return code for configfile err

    antblant committed May 14, 2017
    OCF_ERR_CONFIGURED: cluster manager shuts down the resource (on all nodes)
    OCF_ERR_INSTALLED: cluster manager will try to recover resource on different node (which may have the correct configuration)
Commits on May 8, 2017
  1. Medium: SAPInstance: Add IS_ERS parameter (bsc#1036486)

    fmherschel authored and krig committed Apr 27, 2017
    If IS_ERS is true, mark a per cluster attribute for a
    specific ASCS/ERS pair describing which node is the
    best place to failover a failed ASCS.
  2. Make ocf_is_ver() handle dist. specific versions

    greemo committed May 8, 2017
    for example: 1.1.14-8.el6_8.1
Commits on May 5, 2017
Commits on May 4, 2017
  1. Merge pull request #970 from matelakat/refactor-shellfuncs-run

    dmuhamedagic committed May 4, 2017
    Refactor ocf-shellfuncs.in/ocf_run
Commits on May 3, 2017
  1. Merge pull request #964 from oalbrigt/DB2-HADR

    krig committed May 3, 2017
    DB2: fix HADR support for DB2 V98+
  2. Refactor ocf-shellfuncs.in/ocf_run

    matelakat committed May 3, 2017
    When calling `ocf_run` people expect that the function returns with the
    exit status of the subprocess. In case of 0 exit code we returned
    $OCF_SUCCESS which is 0 as well, but that doesn't read well. Refactoring
    the return statements, so it is evident that the function always returns
    with the exit code of the subprocess.
  3. Merge pull request #968 from dciabrin/ipaddr2-ipv6-preferred-lft

    oalbrigt committed May 3, 2017
    IPaddr2: add option for specifying IPv6's preferred_lft
  4. Merge pull request #969 from beekhof/master

    oalbrigt committed May 3, 2017
    docker: Allow callers to specify a set of directories that should be …
  5. docker: Allow callers to specify a set of directories that should be …

    beekhof committed May 2, 2017
    …created if the don't exist
Commits on Apr 26, 2017
  1. IPaddr2: add option for specifying IPv6's preferred_lft

    Damien Ciabrini
    Damien Ciabrini committed Apr 26, 2017
    This change allows setting the preferred_lft option when creating an
    IPv6 address. This can be used to ensure that the created IP address
    will not be used as a source address for routing.
Commits on Apr 19, 2017
  1. Merge pull request #965 from dciabrin/galera-hostmap-bundle

    oalbrigt committed Apr 19, 2017
    galera: fix master target during promotion with cluster_host_map
Commits on Apr 13, 2017
  1. galera: fix master target during promotion with cluster_host_map

    Damien Ciabrini
    Damien Ciabrini committed Apr 13, 2017
    When option cluster_host_map is in use, it is assumed that galera node
    names map to pacemaker node names _because_ those galera names are not
    part of the pacemaker cluster in the first place.
    
    This is not always the case (e.g. when using pacemaker bundles), so
    fix accordingly.