Appendix G. MySQL Enterprise Monitor Change History

Table of Contents

G.1. Changes in MySQL Enterprise Monitor 2.2.3 (8 September 2010)
G.2. Changes in MySQL Enterprise Monitor 2.2.2 (15 July 2010)
G.3. Changes in MySQL Enterprise Monitor 2.2.1 (26 May 2010)
G.4. Changes in MySQL Enterprise Monitor 2.2.0 (5 May 2010)

This appendix lists the changes to the MySQL Enterprise Monitor, beginning with the most recent release. Each release section covers added or changed functionality, bug fixes, and known issues, if applicable. All bug fixes are referenced by bug number and include a link to the bug database. Bugs are listed in order of resolution. To find a bug quickly, search by bug number.

G.1. Changes in MySQL Enterprise Monitor 2.2.3 (8 September 2010)

Bugs fixed:

  • When performing an unattended installation, the MySQL Enterprise Service Manager would not be restarted automatically if the configuration file contained changes. You can now force the restart to occur by using the --forceRestart option. When set to 1, the server will be restarted using the old (unmodified) configuration file. (Bug#56192)

  • The content of the user field on the Edit User screen was not protected so it was possible to insert scripting instructions into the field contents. (Bug#56189)

  • It was possible to fake the request and promote a user to a different role without having the privileges to do so. (Bug#56187)

  • The --skip-locking option for the embedded MySQL server was being used in place of the deprecated --skip-external-locking option. (Bug#56004)

  • The embedded OpenSSL has been updated to version 0.9.8o. (Bug#55949)

  • It was possible for a user without the ability to view the Manage Users page by visiting the URL, even if the user was not authorized to view the page. Users without the right credentials are now provided with a warning that the page is unavailable. (Bug#55585)

  • When the MySQL server supporting MySQL Enterprise Service Manager is unavailable, the catalina.out file is populated with error messages for each MySQL Enterprise Agent communicating with MySQL Enterprise Service Manager. The quantity and frequency of the messages can cause the error log to grow at a very fast rate. The logging configuration has been altered so that Tomcat-specific errors go to tomcat.log, rotating every 10MB for up to 10 logs. Application errors are reported in mysql-monitor.log. (Bug#55516)

  • The same example query could be consistently resent to MySQL Enterprise Service Manager, even though the example query had already been reported. (Bug#55453)

  • Using LDAP authentication for mapping roles using user attributes would not authenticate correctly. (Bug#55450)

  • The supplied setenv.sh script sets the wrong temporary directory value for HeapDumpPath. The path has been updated to the monitor/apache-tomcat/temp directory within the MySQL Enterprise Service Manager installation. (Bug#54533)

  • The MySQL Enterprise Agent installer would ask for agent credentials, even if you had indicated that the correct user credentials had already been created. (Bug#53943)

  • The Tomcat config.properties file could be installed with unsafe permissions. (Bug#52985)

  • The file permissions on the mysql directory within the MySQL Enterprise Service Manager installation would not be set correctly on Mac OS X. (Bug#35203)

G.2. Changes in MySQL Enterprise Monitor 2.2.2 (15 July 2010)

Bugs fixed:

  • LDAP roles using the search method did not work when searching for authentication credentials. (Bug#54806)

  • The output from MySQL Enterprise Agent using --version would show incorrect versions for some components. (Bug#54701)

  • The data reported for the binary log file size (in mysql::masterlogs::filesizesum) was erroneously reporting the count of the files, not the sum of the size of the binary logs. (Bug#54618)

  • Monitoring a database with a large number of tables would lead to long queries on INFORMATION_SCHEMA which could lead to operating system memory graphs not being updated correctly. (Bug#54591)

  • The help output from MySQL Enterprise Agent could show missing or corrupt characters. (Bug#54559)

  • The MySQL Enterprise Agent could fail to start on 64-bit platforms. (Bug#54433)

  • Using the diagnostic zip functionality would fail if you had configured a server group with a forward slash / in the name. (Bug#54136)

  • Using the Error option for the Referral method for LDAP settings generates an error trace. (Bug#54047)

  • A MySQL Enterprise Agent could report the a MySQL instance as being down, even though the instance was available. This was related to the way local and remote filesystems were identified and reported by MySQL Enterprise Agent. A new option, --agent-report-network-fs has been added to indicate whether remote filesystems should be reported in the diskspace reports. By default, the MySQL Enterprise Agent no longer reports network filesystem information, only data for local filesystems. (Bug#54002)

  • The ssh fingerprint used to identify agents would fail on Ubuntu Linux. (Bug#53977)

  • On Windows platforms, the MySQL Enterprise Agent could show increasing memory usage over time. (Bug#53880)

  • When mysqld is in a hung state and unresponsive, MySQL Enterprise Agent and MySQL Enterprise Service Manager may not report the server as unavailable. (Bug#53090)

  • If a MySQL Enterprise Agent got a timeout during the initial checkin with MySQL Enterprise Service Manager (for instance, if MySQL Enterprise Service Manager was busy), it would fail to resynchronize properly and show the monitored MySQL instances as down. (Bug#52952)

  • The GRANT for the agent user could include permissions for a wide range ofd hosts, and different grant sould be created by different platform installers. (Bug#52933)

  • During and upgrade installation from 2.1 to 2.2, some files in the MySQL Enterprise Agent installation directory would remain. During an upgrade, the installer now performs the following operations:

    • The file custom.xml will be copied from share/mysql-proxy/items to share/mysql-monitor-agent/items.

    • The contents of share/mysql-proxy and the directory are deleted.

    • The README files in share/doc are removed. The 2.2 files are located in share/doc/mysql-monitor-agent.

    (Bug#52770)

  • The installer dialog for the text installation for MySQL Enterprise Agent would show a blank password prompt when asking for confirmation of the root password. (Bug#52129)

  • When performing an upgrade, the value of agent-host-id would not be preserved. (Bug#47676)

  • A TRUNCATE TABLE statement could be mis-identified in the tokenizer. (Bug#44772)

  • When reporting errors related to multiple instances monitored by the same MySQL Enterprise Agent, the error message would not indicate which instance reported the error. (Bug#44748, Bug#47374)

  • Errors determining InnoDB data within MySQL Enterprise Agent would be reported, even when InnoDB had been disabled on the instance. (Bug#41168)

G.3. Changes in MySQL Enterprise Monitor 2.2.1 (26 May 2010)

Functionality added or changed:

  • The Query Analyzer now includes a Transactional filter to cover all transaction related statements, in addition to the individual statement filters for BEGIN, START, COMMIT, ROLLBACK, SAVEPOINT, and RELEASE. (Bug#53564)

Bugs fixed:

  • The required version for running a separate MySQL server has been updated to 5.1.43 in the MySQL Enterprise Service Manager installer. (Bug#53953)

  • Query Analyzer could fail to operate correctly when upgrading from MySQL Enterprise Service Manager 2.1 to MySQL Enterprise Service Manager 2.2. Accessing the details of a query could lead to a NullPointerException error. (Bug#53845)

  • It was possible to import a custom advisor bundle to replace the main advisor bundle. This would delete all existing rules, schedules and advisors. The system now checks the bundle before import. (Bug#53826)

  • The default settings for the LDAP interface for authenticating users has been updated. By default, lookups are made using Bind as user, and the default password digest mechanism is now MD5. (Bug#53679)

  • When generating a support report, the server provided no feedback on the production process. A page will now be generated to advise you of the current status before providing you with a completed message once the report generation has been completed. (Bug#53584)

  • If the backend MySQL database being used for MySQL Enterprise Service Manager is unavailable during an update installation, the installer may give a misleading message indicating the problem. (Bug#53557)

  • The filters for LOAD DATA INFILE, START TRANSACTION, and HANDLER CLOSE statements in the Query Analyzer interface did not filter statements correctly. (Bug#53425)

  • A LOAD DATA INFILE statement would not get normalized correctly within the Query Analyzer display. (Bug#53419)

  • If there are directories located within the configuration instances directory that are empty (i.e. contain no instance configuration file), MySQL Enterprise Agent would fail to start. The behavior has been updated to provide a message level error. (Bug#53317)

  • A warning is now provided before generating the support diagnostics information to indicate the potential time delay. (Bug#53193)

  • When upgrading an existing MySQL Enterprise Service Manager installation, where the MySQL database is identified as not being on the current host (either because it is on the same machine with a different IP, or a different machine), the upgrade installer would fail to provide a warning that such a MySQL database installation cannot be upgraded automatically by the installer. A warning providing more information on how to upgrade is now provided. (Bug#53158)

  • Adding a new instance to the list of monitored instances while using the MySQL Enterprise Dashboard could lead to a scheduling exception error. (Bug#52190)

  • Deleting a user while the user is currently logged in and using MySQL Enterprise Dashboard would not prevent the user from continuing to use the MySQL Enterprise Dashboard, even though the user no longer existed in the system. (Bug#52142)

  • The methods for replication discovery by MySQL Enterprise Agent have been modified. A new option for MySQL Enterprise Agent has been added that prevents the master UUID being used in the replication topology determination. This also prevents MySQL Enterprise Agent from obtaining replication information from the master.info file. (Bug#50604)

  • If the backend MySQL database being used for MySQL Enterprise Service Manager is unavailable during an update installation, the installer may give a misleading message indicating the problem. (Bug#38897)

  • If a Heat Chart rule had de-escalated and the corresponding higher event had not been closed, the Heat Chart dot link did not show the event on the Events tab. This was because the de-escalated state was shown in the Heat Chart, but an Event was not generated for it until the more serious Event was closed. (Bug#28529)

G.4. Changes in MySQL Enterprise Monitor 2.2.0 (5 May 2010)

Functionality added or changed:

  • The included Java runtime environment has been updated to JDK 6.0 Update 18. (Bug#50347)

  • You can delete multiple servers from the Manage Servers tab by selecting each server and using the delete servers button. (Bug#50259)

  • The MySQL Enterprise Dashboard has been confirmed as compatible with Google Chrome 4.0. (Bug#49407)

  • The Manage Servers window has been updated to allow you to view and filter the list of servers using a variety of criteria, making it easier to find servers, agents, or both that require upgrading. (Bug#48329)

  • The version of MySQL Server installed with MySQL Enterprise Monitor was upgraded to 5.1.40. (Bug#48218)

  • The smallest purge log time interval that could be set in the Dashboard was one week. MySQL Enterprise Monitor was changed to allow setting the smallest purge log time interval to one day. (Bug#46822)

  • Certain heat chart rules cannot be unscheduled or disabled, this is because they are required for correct operation of MySQL Enterprise Monitor. Should an attempt be made to unschedule or disable one of these heat chart rules, a suitable message is now displayed, explaining this requirement. (Bug#46535)

  • Custom graphs and advisors can now be imported, exported and deleted from the system. See Section 2.8.8, “Import/Export”. (Bug#45436)

  • The Manage Servers page was changed so that it was possible to view the Up Time for a server by hovering the mouse cursor over the value in the Up Since column. Additionally, the ability to sort servers based on the Up Since column was also added. (Bug#44309)

  • The Manage Servers page was changed to include the Agent version number, and additionally the ability to sort servers based on the Agent version column. (Bug#44308)

  • Support for LDAP authentication has been added. (Bug#43031)

  • The Query Analyzer interface has been improved to show more information about query states and for the filtering system to enable you to view and filter the contents to more easily identify problem queries. For more information, see Section 3.2.4, “Filtering Query Analyzer Data”. (Bug#41643)

  • When running an update installation on a machine where you have multiple instances of MySQL Enterprise Agent already installed, the installer could identify and upgrade the wrong version. You can now specify the installed version (as identified by the startup script in /etc/init.d) by using the --servicename=mysql-monitor-agent option to the command line installer.

    For new installations where a instance of the MySQL Enterprise Agent is already installed, the installer could identify an existing version and refuse to perform the installation process. During the installation, you are now prompted with the installation name to use for the startup script if, and only if, an existing instance has been identified. (Bug#41005)

  • The dashboard could be used to change the agent password to one containing the @ character, or other special characters, which subsequently caused errors. To fix this problem, special characters in passwords are now prevented by the dashboard. The list of disallowed special characters can be found at the following location: http://en.wikipedia.org/wiki/Percent-encoding#Types_of_URI_characters (Bug#37172)

Bugs fixed:

  • The full agent URL would be output in the logs in the event of an error, including the agent username and password used to connect to MySQL Enterprise Service Manager. The password is now stripped from the reported URL. (Bug#53199)

  • During installation of MySQL Enterprise Service Manager on Windows, a batch file was created to obtain the MySQL version number. This file could contain security credentials for accessing the server. The batch file is no longer created. (Bug#52847)

  • Uninstalling MySQL Enterprise Service Manager using the Uninstall Services tool, or later removing MySQL Enterprise Service Manager after servers have already been successfully removed, could fail. (Bug#52846)

  • The filter selection in the drop down menus in the Manage Servers tab have been changed from blank entries (indicating all items), to (all). (Bug#52800)

  • EXPLAIN output within the Query Analyzer data would not be rendered correctly if some of the returned EXPLAIN plan data contained NULL values. (Bug#52532)

  • When installing MySQL Enterprise Service Manager, the incorrect permissions could be applied to the embedded MySQL server. (Bug#52213)

  • When performing an installation of MySQL Enterprise Service Manager or MySQL Enterprise Agent and specifying an alternative service name during installation, the installer would recommend using the generic, not custom, service management script to initiate the service. (Bug#52198)

  • Creating a new custom rule without a configured variable would not raise an error. The system will now return U0068 One or more variables is required to make a functioning rule.. (Bug#51218)

  • During an upgrade installation of MySQL Enterprise Service Manager, custom SSL certificates that had been installed could be replaced by the standard versions. Existing custom certificates are now preserved during an upgrade. (Bug#51118)

  • Scheduling a rule against an instance, and then deleting that instance, would retain the instance within MySQL Enterprise Dashboard, identified as Unknown. (Bug#51095)

  • If the MySQL server holding the repository information for MySQL Enterprise Service Manager crashes and you are using MySQL Proxy to monitor the backend server, the problem may be difficult to identify from the logs. (Bug#50881)

  • When using the MySQL Enterprise Agent Proxy Service, if the backend MySQL server went down, and then the clock on the MySQL Enterprise Agent Proxy Service host went back in time (for example, during daylight savings time adjustments), the MySQL Enterprise Agent Proxy Service would stop sending queries to the configured backend. (Bug#50806)

  • The SSL certificates supplied with MySQL Enterprise Service Manager have been updated.

    The certificate shipped with MySQL Enterprise Service Manager is an example certificate that expires after 1 year and that folks should create their own and back it up between MEM service manager updates. For more information on updating the certificate, see Section C.1.2, “Creating a new SSL KeyStore”. (Bug#50694)

  • Clicking a link to a support issue within the dashboard is now configured to open the issue within a new window. (Bug#50651)

  • To aid in the replication discovery performed by the agent, you can now provide alternative credentials for connecting to the replication master to obtain the master UUID created by MySQL Enterprise Monitor. The master UUID enables MySQL Enterprise Agent to more easily identify and reconcile the master. You set these options on an instance basis, and setting the options prevents the agent using the master.info to determine the master UUID. (Bug#50604)

  • The MyISAM Key Buffer Usage Graph could report misleading information about the key buffer block usage allocation. (Bug#50365)

  • When the agent is shut down, the message used to be reported at the critical log level. This has been changed so that shutdown message are now reported at the message log level. You can enforce the shutdown message to be reported in the log regardless of the log level by using the verbose-shutdown option. (Bug#50340)

  • The installer for MySQL Enterprise Agent on Linux 64-bit using glibc-2.3 would fail before the installation had completed properly. (Bug#50289)

  • Deleting users within MySQL Enterprise Service Manager could lead to errors in the repository database that would affect further operations involving the deleted user. (Bug#49896)

  • When creating a new instance by copying an existing agent configuration, it is possible to create an orphaned agent. The recommended advice is to TRUNCATE the mysql.inventory table. However, doing this could lead to additional errors and an exception when the scheduled data updates on the now orphaned agent are executed. (Bug#49882)

  • The configuration_report.txt file would cont6ain information about the SSL port to use, even if SSL support had not been enabled during installation. (Bug#49769)

  • The memory footprint of MySQL Enterprise Agent would slowly increase over time, particularly on Mac OS X and Solaris/OpenSolaris. (Bug#49739)

  • If the agent-host-id value contained a period, information would fail to be reported to MySQL Enterprise Service Manager. (Bug#49699)

  • When starting up MySQL Enterprise Monitor, the relevant environment variables are now inserted into the catalina.out log duing startup.

    The environment variables reported include:

    $CATALINA_BASE
    $CATALINA_HOME
    $CATALINA_TMPDIR
    $JAVA_HOME
    $JRE_HOME
    $_RUNJAVA
    $JAVA_OPTS
    $CATALINA_OPTS
    $JAVA_ENDORSED_DIRS
    $CLASSPATH
    

    In addition, the version information for the Java environment used will also be reported in the log. (Bug#49626)

  • The information in the table on the Query Analyzer page was displayed in either ascending sort order, descending sort order, or an unsorted order. You could change this behavior by clicking the column you wanted to sort.

    The behavior has now been modified so that the table data is sorted in either ascending or descending order, clicking the columns toggles between these two states. (Bug#49471)

  • The method for generating a new UUID for the MySQL Enterprise Agent has been changed. You can create a new UUID using:

    shell> bin/mysql-proxy --plugins=agent --agent-generate-uuid

    Alternatively, you can use the bin/agent-generate-uuid command in the agent installation directory. (Bug#49306, Bug#49903)

  • When disabling or unscheduling Heat Chart rules, the following issues occurred:

    1. For the disable operation, the resulting page showed all rules as disabled, even though three rules were not disabled. However, if the page was refreshed, those three rules were correctly displayed as enabled.

    2. For the unschedule operation, there was no fading white cloud with rule counts.

    (Bug#48870)

  • The backend Lua scripts used with the MySQL Proxy and MySQL Enterprise Agent were not updated for the the MySQL Proxy 0.8.0 core. This would lead to some Lua scripts triggering an error. (Bug#48718)

  • The version information output for the MySQL Enterprise Agent did not include the version of the all the individual components loaded into the MySQL Enterprise Agent chassis. (Bug#48714)

  • When using the proxy (for query analyzer) with the MySQL Enterprise Agent, the proxy plugin must be listed first in the plugins configuration parameter. (Bug#48713)

  • The name of the section in mysql-monitor-agent.ini required for configuration is mysql-proxy, the default template had been mistakenly modified to mysql-monitor-agent. (Bug#48712)

  • When upgrading an agent, the existing agent is shutdown and then restarted after the installation process has completed. To reduce downtime during installation, you can now opt to have the agent restarted immediately after installation, but before the rest of the installer completes. (Bug#48581)

  • RAM statistics for FreeBSD machines would not be reported correctly by MySQL Enterprise Agent. (Bug#48493)

  • Cutting and pasting sample queries from Query Analyzer page in could lead to bad concatenation of the query components. (Bug#48479, Bug#50121)

  • When viewing pages of queries on the Query Analyzer tab, changing the filter options could lead to an empty list if you modified the filters while on a page that is higher than the number of the pages in the filtered display. The page numbering is now reset when you change the filter options to ensure that you are viewing the filtered queries. (Bug#48409)

  • Changing the filter parameters on the Query Analyzer tab would lose your current sorting preferences. (Bug#48408)

  • When viewing the Query Analyzer tab and selecting a server, the server name would still shown in the table of queries. When a server has been explicitly selected, the server column is no longer displayed. (Bug#48390)

  • Deleting a server or group from the Manage Servers tab could result in a stacktrace dump if the server or group ID selected was no longer valid. A confirmation is now produced if MySQL Enterprise Service Manager identifies an issue when trying delete the selected server or group. (Bug#48255)

  • When a time range was selected on a dashboard graph, the user was taken to BrowseQueries, and the correct graph and graph time range was selected. However, the following problems occurred:

    • The query time range did not agree with the graph time range.

    • Changes to the query range interval did not correctly update the graph or selection.

    • Changes to the graph type were not updated correctly.

    (Bug#48234)

  • The URL for viewing the contents of an event alerted by email were not populated correctly, making the supplied URL invalid. (Bug#48193)

  • The full-graph pop-up on the dashboard did not reflect the group/server selection of the containing dashboard, it always displayed the default group. (Bug#48167)

  • The Add to Group context menu item, on the Manage Servers page, listed all servers when clicked, rather than the subset pre-selected from the main server list. (Bug#48141)

  • When adding a new rule without providing a version a stacktrace would be produced. A default version (1.0) will be added to new rules if the version has not been specified. (Bug#47877)

  • The indentation on alert emails could be incorrect if the list included a bullet list of items. (Bug#47802)

  • When upgrading an installation of MySQL Enterprise Service Manager where the MySQL being used is not the bundled MySQL server, the upgrade would fail while creating a backup of the MySQL installation. (Bug#47775)

  • Support has been added for SNMPv2 traps in addition to the existing SNMPv1 traps. You can configure the version to use for SNMP traps, see SNMP Traps.

    In addition, support has also been added to send SNMP notifications to two hosts in place of just one host. Notifications are sent to both hosts simultaneously. (Bug#47686, Bug#48955)

  • Support issues can have a Severity from S1 through S4, but can also have a value of “NS” meaning “No Severity Set”. MySQL Enterprise Monitor was not able to parse this, which resulted in errors such as the following:

    2009-09-23 13:53:51,812 ERROR [em-worker-pool-thread-6:monitor.support.DevSpPoller] error
    consuming successful response
    java.lang.IllegalArgumentException: No enum const class
    com.mysql.etools.monitor.support.SpIssue$Severity.NS
    	at java.lang.Enum.valueOf(Enum.java:196)
    ...
    

    (Bug#47562)

  • MySQL Enterprise Monitor would fail to install on Miracle Linux4. (Bug#47209)

  • Occasional Null Pointer Exception errors could be raised after an initial login to the dashboard due to problems determining the correct name to use for displaying the list of servers. (Bug#47181)

  • The diagnostic report file has been updated so that the filename of the report includes the timestamp when the report was generated. (Bug#47164)

  • The REPLACE and CALL statements did not show a rows graph within the statement pop-up graph tab. (Bug#46796)

  • When trying to change the password for the agent user in the Dashboard to a password that contained a “@” or other special character, the following error message was displayed:

    An Error Occurred
    
        * U0160 The following characters are invalid in an agent password:
    
    
    !*'();:@&=+$,/?%#[]
    
    

    However, this password worked correctly when first connecting the agent user to the Monitor. (Bug#46757)

  • The subject line in alert emails was not configurable. (Bug#46689)

  • The schedule for the MySQL Agent Not Reachable rule cannot be modified. This is to ensure that problems communicating with the agent are identified immediately. If you attempt to modify the scheduling an error will be reported. (Bug#46667)

  • Pressing cancel on a pop-up would cause a page reload, instead of just closing the pop-up window. (Bug#46604)

  • Pressing cancel on a pop-up would cause a page reload, instead of just closing the pop-up window. (Bug#46604)

  • The cry for help email included a stack trace, however it was displayed on a single line without any line breaks. (Bug#46458)

  • Some queries would show multiple tables in the Example Query displayed within the Query Analyzer. (Bug#46438)

  • On very large deployments with a large number of servers, the Events tab could take some time to load. (Bug#46436)

  • When using MySQL Enterprise Agent, if the current configured time went backwards (for example during time correction), then the MySQL Enterprise Agent would stop reporting data and induce a high load on the machine running MySQL Enterprise Agent (Bug#46052)

  • When viewing queries within the Query Analyzer, the example query shown could be taken from outside of the time range configured within the query analyzer view. (Bug#45683)

  • The Windows installers for MySQL Enterprise Service Manager did not include 64-bit versions of the various binary tools. (Bug#45682)

  • MySQL Enterprise Monitor alert emails used incorrect formatting.

    The following problems occurred:

    • Text in the subject and body was not formatted properly. Variables appeared to be displayed such as '.name', '.advice', and '.description', rather than values.

    • Line formatting (CR/LF) appeared to be absent.

    • No local time was shown (GMT time only).

    (Bug#44769)

  • When a MySQL Enterprise Agent is disconnected from MySQL Enterprise Service Manager, the agent would keep 10 minutes of information by default, that would then be sent to the server when the agent reconnected. Any data older than 10 minutes would be lost. You can now configure the backlog timeout by using the agent-backlog-timeout to specify the amount of time that data will be kept. For more information, see agent-backlog-timeout (Bug#44262)

  • When the license for MySQL Enterprise Monitor was downgraded, for example from Platimun to Gold, items that should have no longer been available were present in the Dashboard. (Bug#43823)

  • The IP address of the URL used to communicate with MySQL Enterprise Service Manager would be resolved only during initial startup by an individual MySQL Enterprise Agent. This causes problems where the MySQL Enterprise Service Manager is on a virtual IP address and the DNS is used to provide failure to a new IP address. IP address caching has now been disabled, which should cause the IP address to be re-looked up and updated during a failover of the MySQL Enterprise Service Manager. (Bug#43537)

  • Starting the MySQL Enterprise Agent would prevent the current session for logging out while the agent was still running. (Bug#43051)

  • Queries containing DATE_SUB functions could show incorrect queries when normalized within Query Analyzer. (Bug#42688)

  • After an upgrade from 2.0.3 to 2.0.4.7141 there was a missing label for the agent_cpu_usage graph.

    MySQL Enterprise Monitor was changed so that any graphs or advisors that are of a higher subscription level than currently licensed for are unscheduled and removed from the product. (Bug#42687)

  • The Agent would not reconnect to a monitored database if it was started when the monitored server was down. The agent log contained the following error:

    Can't connect to MySQL server on '127.0.0.1' (0) (mysql-errno = 2003)

    The agent only sent OS data to the Dashboard. Further, when the monitored server was later started, no attempts to reconnect were logged.

    The problem could be worked around by restarting the agent when the monitored server was running again. (Bug#42581)

  • The installer for MySQL Enterprise Agent has been updated to include support for creating the username and password required by the MySQL Enterprise Agent during operation. This process creates the user with the necessary privileges within your monitored MySQL server. For more information, see Section 5.3.1, “Creating a MySQL User Account for the Monitor Agent”. (Bug#41630)

  • When viewing detailed query information within the query analyzer, the incorrect 'worst' query could be shown in the detail view. (Bug#41076)

  • When hovering over the heat chart, you now get a count of the number of servers affected by the corresponding rule within the tooltips, even when the list of servers, or specific server groups, are collapsed. (Bug#40823)

  • In the event of a failure by MySQL Enterprise Agent to communicate with the MySQL Enterprise Service Manager, for example when the server has been restarted, it was possible for there to be gaps in the reported information. This may have shown up as gaps in the graph output. (Bug#40729)

  • Queries containing an INNER JOIN clause could show incorrect queries when normalized within Query Analyzer. (Bug#40676)

  • It was not possible to order the rows in the table within the Manage Servers tab. You can now sort by clicking the required column. (Bug#39457)

  • Queries containing an LIMIT clause with value placeholders could show incorrect queries when normalized within Query Analyzer. (Bug#38892)

  • The formatted query view in the pop-up within the Query Analyzer tab could show an empty query if the query was short and could not be formatted properly. (Bug#36885)

  • When events have been raised by MySQL Enterprise Monitor, new events of the same type would not be notified until the original event had been acknowledged. This could lead to critical events being missed. In addition, when viewing existing events that have been triggered, it is impossible to know if the notified event is still causing a problem.

    The fix has been rectified through the implementation of the auto-close feature on events. With this feature enabled, events will automatically be closed once the event is not longer identified as causing a problem. (Bug#34240, Bug#23640, Bug#27132)

  • If a Heat Chart rule had de-escalated and the corresponding higher event had not been closed, the Heat Chart dot link did not show the event on the Events tab. This was because the de-escalated state was shown in the Heat Chart, but an Event was not generated for it until the more serious Event was closed. (Bug#28529)

  • When performing an installation of MySQL Enterprise Agent, installing multiple agents could lead to service names (Windows) and startup scripts (Unix) being overwritten, corrupted, or creating multiple entries. The installer will now ask for a different service name, or you can explicitly specify one to the installer using the servicename option. See installer_servicename. (Bug#27332)

  • It was possible to import a Platinum level bundle into a MySQL Enterprise Monitor server that only had a Gold product key. (Bug#26237)

  • Different Advisor jar and subscription levels could be installed at the same time, which could lead to incompatibilities and scheduling issues. In addition, you could install different Advisor jar files and corrupt the localization descriptions for the advisors. (Bug#26167, Bug#35278)

  • It was not possible to assign a user a Role that had read-only permissions. (Bug#23508)