+ /usr/bin/clustercheck clustercheck password 0 is fine. But mention this part: # " # Recommended: server_args = user pass 1 /var/log/log-file 0 /etc/my.cnf.local" # Compatibility: server_args = user pass 1 /var/log/log-file 1 /etc/my.cnf.local" +In CentOS ``mysql-libs`` conflicts with ``Percona-XtraDB-Cluster-server-56.x86_64`` package. To avoid this you can replace the ``mysql-libs`` with ``Percona-Server-shared-51.x86_64`` package. To replace you need to remove mysql-libs (there is no 'replace' in yum) before installing PXC packages. Dependency resolution ensures -shared-51 is installed. + - Currently replication works only with |InnoDB| storage engine. Any writes to tables of other types, including system (mysql.*) tables, are not replicated. However, ``DDL`` statements are replicated in statement level, and changes to mysql.* tables will get replicated that way. So, you can safely issue: ``CREATE USER...``, but issuing: ``INSERT INTO mysql.user...``, will not be replicated. Mention about wsrep_replicate_myisam here. +Based on `Percona Server 5.6.15-63.0 `_ including all the bug fixes in it, `Galera Replicator 3.3 `_ and on `Codership wsrep API 5.6.15-25.2 `_ is now the first **General Availability** release. All of |Percona|'s software is open-source and free, all the details of the release can be found in the `5.6.15-25.5 milestone `_ at Launchpad. It is Galera 3.4 here. https://launchpad.net/galera/+milestone/25.3.4 Codership API 25.5 - https://launchpad.net/codership-mysql/+milestone/5.6.16-25.5 + Bug fixed :bug:`1219605`. Replication of partition tables without binlogging enabled failed, partition truncation didn't work because of lack of TO isolation there. + Bug fixed :bug:`1272982`. Ignore this. :) + Bug fixed :bug:`1281682`. New options (under [sst]) are added: inno-backup-opts, inno-apply-opts, inno-move-opts which pass options to backup, apply and move stages of innobackupex. + Using ``LOAD DATA INFILE`` in with :variable:`autocommit` set to ``0`` and :variable:`wsrep_load_data_splitting` set to ``ON`` could lead to data loss. Bug fixed :bug:`1281810`. Let us not mention 'data loss' here. It was not that explicit of a loss. Instead, 'incomplete loading of records while chunking' would be better here. + Bug fixed :bug:`1284670`. Not user visible, safe to skip (may cause unneeed alarm among users :)). + If a particular node was setup to be the donor via setting :variable:`wsrep_sst_donor` and the donor node was in desynced state for some reason, then the joiner would wait for the donor to become synced again which could take a long time. Bug fixed :bug:`1285380`. This is a bit more complex. Best way to describe would be to say "The joiner would wait and not fall back to choosing other potential donor nodes (not listed in wsrep_sst_donor) by their state. This happened even when comma was added at the end. This fixes it for that particular case". + Initial timeout, of 100 seconds, to receive packets from donor has been added to SST so that if donor dies somewhere in between, joiner doesn't hang. Bug fixed :bug:`1292991`. first packet via. SST would be better here. Also, better to say, 'Initial configurable timeout'. The SST configuration variable here is 'sst-initial-timeout'. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% To add: https://bugs.launchpad.net/codership-mysql/+bug/1280557 Better diagnostic error message has been when max write set had been succeeded. Bug# 1281696 Incorrect warnings have been fixed and also better handling of repeated usage with same value. Bug# #1290612 Issue with wsrep_slave_threads wherein if the number of slave threads was changed before closing threads from an earlier change, it could increase the total number of threads beyond value in wsrep_slave_threads. Also, add here that, wsrep_slave_threads is a dynamic variable. Bug #1293624 A regression in mutex handling caused dynamic update of wsrep_log_conflicts to hang the server. Bug##1294760 Presence of /tmp/test directory and an empty test database caused Xtrabackup to fail, causing SST to fail. This is an xtrabackup issue. But, this has been fixed in PXC's xtrabackup SST separately by using unique temporary directories with Xtrabackup. Finally, The security issue with auth_socket has been fixed here as well: https://bugs.launchpad.net/percona-server/+bug/1289599 (Mention this only if it was mentioned in PS's release notes/BP).