Permalink
Commits on May 25, 2012
Commits on May 24, 2012
  1. Medium: IPaddr2: fix recent regression, nic=lo always "stopped"

    lge committed May 24, 2012
    Intended logic:
       If lvs_support is enabled, and IP is only found active on loopback,
       pretend IP was not served,
    
    Regression introduced in
      d93b5fd IPaddr2: add support for same ip/netmask on different interfaces
    caused any IP found on lo* to be ignored and considered not served.
  2. Medium: oracle: quote regex

    lge committed May 23, 2012
    To avoid it getting glob-expanded by the shell by accident.
    demo:
     set -x
     mkdir -p ^/.blafoo$
     echo /foo | grep ^/.*foo$ || echo BROKEN
     echo /foo | grep "^/.*foo$" || echo BROKEN
    
    Admittedly unlikely, but let's be paranoid.
  3. Medium: oracle, oralsnr: get rid of eval

    lge committed May 23, 2012
    I tend to avoid eval if possible.
    We better do this as shell function instead.
  4. rgmanager: fix nfsrestart metadata in fs.sh and clusterfs.sh

    fabbione committed May 24, 2012
    Resolves: rhbz#822053
    
    Signed-off-by: Fabio M. Di Nitto <fdinitto@redhat.com>
Commits on May 23, 2012
Commits on May 22, 2012
  1. High: dhcpd: new RA to manage ISC DHCP servers

    Chris Bowlby committed with dmuhamedagic May 22, 2012
    Signed-off-by: Dejan Muhamedagic <dejan@suse.de>
  2. Medium: mysql: improve handling of reset slave

    y-trudeau committed with dmuhamedagic May 22, 2012
    Signed-off-by: Dejan Muhamedagic <dejan@suse.de>
  3. High: iSCSILogicalUnit: correctly match for target IQN and backing de…

    fghaas committed May 22, 2012
    …vice name (iet)
    
    Fix is practically identical to the one in fb2a5d4, modified to match
    the iet implementation.
  4. Merge pull request #96 from nozawat/myFeatureSpike-jboss

    dmuhamedagic committed May 22, 2012
    Addition to rotate of console log.
  5. High: iSCSILogicalUnit: correctly match for target IQN and backing de…

    vchepkov committed with fghaas May 22, 2012
    …vice name (tgt)
    
    When we're dealing with several targets, or several backing devices,
    where the name of one is a substring of the other, the grep/sed magic
    to figure out target IDs and LUNs would fail because of multiple matches.
    
    Fix the expressions so we match to the end of the line ("$" anchor).
  6. Merge pull request #94 from raoulbhatia/mysql

    dmuhamedagic committed May 22, 2012
    minor MySQL documentation and logging bugfixes
  7. Merge pull request #95 from nozawat/myFeatureSpike-slapd

    dmuhamedagic committed May 22, 2012
    The correction that bind_dn is not usable
Commits on May 21, 2012
  1. rgmanager: add nfsdrestart option as last resource to umount clusterfs

    fabbione committed May 21, 2012
    Resolves: rhbz#822053
    
    Signed-off-by: Fabio M. Di Nitto <fdinitto@redhat.com>
Commits on May 18, 2012
Commits on May 16, 2012
  1. Merge pull request #92 from raoulbhatia/postfix

    dmuhamedagic committed May 16, 2012
    Low: postfix: issue an error if multiple data directories are set
  2. rgmanager: add nfsdrestart option as last resource to umount fs

    fabbione committed May 16, 2012
    Resolves: rhbz#822053
    
    Signed-off-by: Fabio M. Di Nitto <fdinitto@redhat.com>
  3. Low: nfsserver: check if we are running before trying to start again

    ikedaj committed with dmuhamedagic May 16, 2012
    Signed-off-by: Dejan Muhamedagic <dejan@suse.de>
  4. Low: nfsserver: don't validate twice

    ikedaj committed with dmuhamedagic May 16, 2012
    Signed-off-by: Dejan Muhamedagic <dejan@suse.de>
  5. High: mysql: improve replication support

    y-trudeau committed with dmuhamedagic May 16, 2012
    This patch modifies the behavior of the mysql RA in order to better
    support replication.  Upon promotion, the newly selected master stores
    in the cib the information needed by the slaves to reconnect to the new
    master, the binlog file and binlog position.  The new mysql RA also
    maintains, through the monitor call, a transient attribute that can be
    used by the cluster configuration to control the behavior of reader
    VIPs, basically moving them away if the slave is not in a sane state.
    It is also possible to use specific IPs for replication by adding node
    attributes.
    
    Signed-off-by: Dejan Muhamedagic <dejan@suse.de>
Commits on May 15, 2012
  1. Merge pull request #90 from ikedaj/issue-filesystem

    dmuhamedagic committed May 15, 2012
    remove a status file only when OCF_CHECK_LEVEL is set as 20 to prevent the unexpected timeout
  2. Merge pull request #89 from AlexanderKrauth/master

    dmuhamedagic committed May 15, 2012
    SAPDatabase: changed default of monitoring for one database
  3. Merge pull request #88 from t-matsuo/Filesystem-patch

    dmuhamedagic committed May 15, 2012
    Low: Filesystem: support loopback
  4. Merge pull request #87 from t-matsuo/pgsql-patch

    dmuhamedagic committed May 15, 2012
    Low: pgsql: fix a bug ignoring config parameter in replication mode