MySQL Lists are EOL. Please join:

List:Announcements« Previous MessageNext Message »
From:Sreedhar S Date:October 5 2017 5:05am
Subject:MySQL Cluster Manager 1.4.4 has been released
View as plain text  
Dear MySQL Users,

MySQL Cluster Manager 1.4.4 has been released and can be downloaded
from the My Oracle Support (MOS) website. It will also be available
on Oracle Software Delivery Cloud at http://edelivery.oracle.com with
the next monthly update

MySQL Cluster Manager is an optional component of the MySQL Cluster Carrier
Grade Edition, providing a command-line interface that automates common
management tasks, including the following online operations:
  - Configuring and starting MySQL Cluster
  - Upgrades
  - Adding and removing cluster nodes
  - Adding and removing site hosts
  - Configuration changes
  - Backup and restore

MySQL Cluster Manager is a commercial extension to the MySQL family of 
products.
More details can be found at http://www.mysql.com/products/cluster/mcm/

A brief summary of changes in MySQL Cluster Manager version 1.4.4 is 
listed below:

Changes in MySQL Cluster Manager 1.4.4      (2017-10-05)

    This section documents all changes and bug fixes that have
    been applied in MySQL Cluster Manager 1.4.4 since the release
    of MySQL Cluster Manager version 1.4.3.


    Functionality Added or Changed

      * Agent: On Windows platforms, when the collect logs
        command was run and the resulting file path for any
of
        the collected files was longer than 255 characters,
the
        command failed with ERROR 101 (00MGR) Access denied.
A
        new type of error, PATH_TOO_LONG (106) Path max
length
        exceeded for file filename, has been added and is
now
        returned at the situation. (Bug #26619163)

    Bugs Fixed

      * Agent: On Solaris platforms, the collect logs command
        created lots of zombie threads after it finished. It
was
        because a number of thread joins were missing in the
        code, which have been added by this fix. (Bug
#26665347)

      * Agent: For a large NDB Cluster site, the collect logs
        command frequently failed due to errors with the
network
        connections during file transfers. This was because
some
        error codes related to the network connections were
not
        properly handled, and this fix corrects that. (Bug
        #26642616)

      * Agent: The import config command sometimes failed, as the
        mcmd agent updated its configuration store before
all the
        values to be imported arrived. This fix makes mcmd
wait
        for all the values before starting the updates. (Bug
        #26610874)

      * Agent: The collect logs command reported success even
        when an agent failed and the file transfer from it
was
        incomplete. With this fix, when connection to an
agent
        fails, the command allows running file transfers to
        complete, and then throws an error. (Bug #26594948)

      * Agent: The collect logs command sometimes failed with
        ERROR 1003 Internal error: No clients connected,
even
        though clients were connecting and files were being
        transferred. That was due to a faulty mechanism for
        tracking client connections, which has been
corrected by
        this fix. (Bug #26575768)

      * Agent: The mcm client hung when running the collect logs
        command if the host names for the cluster nodes
could not
        be resolved, even if the cluster was actually
defined
        using IP addresses. (Bug #26564176)

      * Agent: A start cluster command timed out when its timeout
        value, being scaled for handling a large DataMemory
size
        for a cluster, was accidentally reset. This fix
makes
        sure the timeout for the execution step of a start
        cluster command is never lowered while the data
nodes are
        starting. (Bug #26543383)

      * Agent: After a node had failed and an update process
        command on it also failed, a stop cluster or stop
process
        command that followed ran into an assertion error.
It was
        because the node, automatically restarted after its
        failure, had its PID file modified by the update
process
        command that was still running. This fix adds a
check on
        the status of the target process for the update
process
        command to prevent the restarted node's PID file
from
        being modified. (Bug #26388603)

      * Agent: The collect logs command hung for 30 seconds after
        completing all file transfers before returning the
        success message. (Bug #26362910)

      * Agent: When a management node's status went from starting
        to stopped, it was not reported by its mcmd agent,
        causing commands like delete cluster to fail because
the
        management node was taken to be still running. (Bug
        #26288009)

On behalf of the Oracle MySQL RE Team
~Sreedhar S

Thread
MySQL Cluster Manager 1.4.4 has been releasedSreedhar S5 Oct