Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Commits on Aug 9, 2013
  1. @t-matsuo
  2. @t-matsuo
Commits on Jun 27, 2013
  1. @t-matsuo
Commits on Jun 26, 2013
  1. @t-matsuo

    Low: pgsql: control slave status once again when sync mode is changed

    t-matsuo authored
    execute control_slave_status() once again to improve performance
    of catching up slave status.
  2. @t-matsuo
  3. @t-matsuo

    Medium: pgsql: set only one node into sync mode when using 3 nodes or…

    t-matsuo authored
    … higher
    
    Only one node can be sync mode to avoid data loss when using 3 nodes or higher.
    See http://www.gossamer-threads.com/lists/linuxha/pacemaker/85077?do=post_view_threaded#85077
  4. @t-matsuo
  5. @t-matsuo

    Low: pgsql: report failed query and return code

    t-matsuo authored
    - report_psql_error() reports given message to distinguish failed query
      when using replication mode.
    - change some loglevels from warn to err when calling report_psql_error(),
      because warn should be used for starting up PostgreSQL.
  6. @t-matsuo

    Low: pgsql: improve error handling of crm_xxxx commands and reload_co…

    t-matsuo authored
    …nf()
    
    - if crm_xxx command fails, retry executing it
      because the number of IPC connections reaches maximum,
      crm_xxxx commands fail.
    - if reload_conf() fails, retry it, because the failure is critical
  7. @t-matsuo
  8. @t-matsuo

    Low: pgsql: change name and specification of exec_func_with_timeout

    t-matsuo authored
    - change function name from exec_func_with_timeout to exec_with_timeout
    - change specification of exec_with_timeout to conform it to exec_with_retry
    - exec_with_timeout outputs more logs
    - increase in performance
  9. @t-matsuo
Commits on May 23, 2013
  1. @t-matsuo
Commits on Apr 2, 2013
  1. @t-matsuo
Commits on Feb 21, 2013
  1. @t-matsuo
  2. @t-matsuo

    Low: pgsql: fix not working when using capital hostname with corosync

    t-matsuo authored
    decapitalize hostname to maintain compatibility with Heartbeat stack
Commits on Jan 24, 2013
  1. @t-matsuo

    Low: pgsql: add check_wal_receiver parameter

    t-matsuo authored
    add check_wal_receiver parameter to check wal_receiver process on monitor
    and notify its status using "<resource name>-receiver-status" attribute
    to check  whether PostgreSQL(Hot Standby) connects to primary.
    The attribute shows status as "normal" or "ERROR".
    Default is false.
  2. @t-matsuo

    Low: pgsql: support starting as Hot Standby

    t-matsuo authored
    rep_mode=slave starts PostgreSQL as Hot Standby to connect to Primary.
    This mode does not support Master/Slave configuration.
Commits on Nov 15, 2012
  1. @t-matsuo
Commits on Nov 8, 2012
  1. @t-matsuo

    Low: pgsql: add restart_on_promote parameter

    t-matsuo authored
    - Default is false.
    - True means that RA deletes recovery.conf and restarts PostgreSQL
      on promote to keep Timeline ID.
Commits on Nov 5, 2012
  1. @t-matsuo

    Low: pgsql: delete include directive if rep_mode is async

    t-matsuo authored
    - delete include directive which is add by pgsql RA from
      postgresql.conf if rep_mode is async.
    - improve meta-data
Commits on Oct 29, 2012
  1. @t-matsuo

    Low: pgsql: check existence of instance number in replication mode

    t-matsuo authored
    check existence of instance number
    because Pacemaker 1.1.8 or higher do not append instance numbers
    if globally-unique=false.
Commits on Oct 22, 2012
  1. @t-matsuo

    Low: pgsql: fix string processing of crm_mon's output on Pacemaker 1.1.8

    t-matsuo authored
    Replication mode dosen't work on Pacemaker 1.1.8
    because the output of "crm_mon -n" is different from one
    on Pacemaker 1.0.x.
    
    So I fix string processing to work on 1.1.8 and 1.0.12.
Commits on Sep 28, 2012
  1. @t-matsuo

    Low: pgsql: change pgsql-status into UNKNOWN when validation is faile…

    t-matsuo authored
    …d on stop
    
    Change pgsql-status into UNKNOWN when validation is failed on stop.
    Because pgsql-status remains in HS even though validation failure
    means stop succeeded in pgsql RA.
Commits on Aug 30, 2012
  1. @t-matsuo
Commits on Aug 23, 2012
  1. @t-matsuo

    Low: pgsql: add option recovery_end_command

    t-matsuo authored
    - add $OCF_RESKEY_recovery_end_command parameter for recovery.conf
    - add a new fuction to make addtional parameters of recovery.conf
    - cosmetic change
Commits on Jul 31, 2012
  1. @t-matsuo
Commits on Jul 12, 2012
  1. @t-matsuo

    Low: pgsql: always set synchronous_standby_names in rep_mode.conf

    t-matsuo authored
    - always set synchronous_standby_names in rep_mode.conf
      to avoid synchronous_standby_names in postgresql.conf
    - remove unnecessary code
Commits on Jun 28, 2012
  1. @t-matsuo

    Low: pgsql: fix grep failure when using some master/slave resources

    t-matsuo authored
    - don't match other master/slave resources
    - remove some unnecessary variables
    - simplify finding a master on monitor
Commits on Jun 27, 2012
  1. @t-matsuo
Commits on May 15, 2012
  1. @t-matsuo
  2. @t-matsuo

    Low: pgsql: fix a bug ignoring config parameter in replication mode

    t-matsuo authored
     - use $OCF_RESKEY_config parameter when it's specified.
     - change meta-data of config parameter
Commits on May 1, 2012
  1. @t-matsuo

    Low: pgsql: simplify a code switching between sync and async.

    t-matsuo authored
    - simplify the check of replication mode
    - remove an unnecessary code
  2. @t-matsuo
  3. @t-matsuo
Something went wrong with that request. Please try again.