Notes & articles

Release notes - version 2.3.0.xxx

Version 2.3 – 26 May, 2011

New Features:

• Top 10 expensive queries – the 10 queries that use the most resource based on Execution Count, Duration, CPU Time, Physical Reads, Logical Reads and Logical Writes are now displayed on the SQL Server Instance and Database overview pages. This data is also available in the Alert Details page for SQL Server alerts.
• Data purging – SQL Monitor now supports explicit purge options for high volume data channels. Note: Purge options that were customized in earlier versions will be translated accordingly. You should review your purge settings after upgrade.
• Analysis page – The Analysis page now supports high resolution graphs and can plot a maximum of 500 data points.
• Performance improvements have been made to the Base Monitor and UI.
• SQL Monitor now works with PagerDuty.
• Emails for Job duration unusual and Job failure alerts now contain job name.
• Monitoring error (host machine data collection) and Monitoring error (SQL Server data collection) alerts now have a two minute time tolerance before they are raised.
• Long running query alert now contains database name and query duration.
• Behavior change for Job duration unusual alert. It now validates the “Ignore less than“ value against the current run job duration instead of the base line.

Bug Fixes:

• SRP-3645: Monitoring error (host machine data collection) and Monitoring error (SQL Server data collection) alerts are too sensitive.
• SRP-3614: Cannot set a SQL connection port greater than 32767.
• SRP-3483: Processor utilization alert can be raised for disabled alert levels.
• SRP-2905: Include query duration for Long running query alert.
• SRP-3688: Analysis graph plots data but not the lines when the custom time range goes further back than the data available..
• SRP-3036: Trace statements in the SQL Processes / Profiler trace tab are not chronologically ordered.
• SRP-2969: Trace information in 2.2 less than 1.3.
• SRP-3090: Offline alerts: include seconds on alert timings.
• SRP-2472: Job failed and Job duration unusual alert email does not contain job name.
• SRP-2924: Allow more than 100 percent for job duration unusual configuration.
• SRP-3399: Blocked process and Long running query alert pages do not display database name.
• SRP-3586: Log backup overdue alert is being raised wrongly after maintenance window.
• SRP-1294:The list of SQL processes in an alert does not include the process which caused the alert.

Known Issues:

• SRP-495: If the Base Monitor loses its connection to the SQL Monitor Data Repository then you may see the following error, “System.Data.SqlClient.SqlException was thrown by method Query on service DataPresenterService: Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding”.
• SRP-1443: An internal error occurs if the database name contains a non-latin character.
• SRP-2167: A user will see a “script stack space quota is exhausted” JavaScript error on Firefox when there is a lot of performance data in an alert details page. Use an alternative browser to view SQL Monitor.
• SRP-2310: “Back in time” can sometime crash with an error when using IE 7. Use an alternative browser to view SQL Monitor.
• SRP-2702: Long running query process ID not appearing in the SQL Processes/Trace tab.

Version 2.2 – 16 February, 2011

New Features:

• Maintenance window - you can now set maintenance windows to temporarily suspend alerting during set periods.
• Four new alerts - you will now receive alerts when connection issues occur.
• Machine offline and SQL Server instance offline alerts have been renamed Machine unreachable and SQL Server instance unreachable respectively.

Bug Fixes:

• SRP-2134: The Fragmented index alert is broken.
• SRP-2927: Alert descriptions for Backup overdue, Log backup overdue and Integrity check overdue alerts need changing.
• SRP-3135/ SRP-3136: The time displayed in the Occurrences tab of the Alert details page is out by 12 hours or is in the wrong format.
• SRP-3018: Purge behavior does not work as expected; a backlog of data due to be purged builds up, and attempts to purge the backlog affects performance. To fix this, SQL Monitor has increased its usage of tempdb, so you may see tempdb grow by up to half the size of your Data Repository database. Once the backlog of data has been purged successfully and your SQL Server service has been restarted, tempdb size will revert back to normal.
• SRP-3140: The Disk used alert description is not entirely accurate.
• SRP-3090: Offline alerts: include seconds on alert timings.
• SRP-2924: Allow a threshold of more than 100% for the Job duration unusual alert configuration.
• SRP-2202/ SRP-1769: The availability of button in the Alert Inbox should not be determined by the visible page.
• SRP-1137: Enable sorting of the SQL Process by ID to make it easier to find a particular process.
• SRP-3243: On the Alert details page, trace data is retrieved only when users click a specific SQL process.

Known Issues:

• SRP-495: If the Base Monitor loses its connection to the SQL Monitor Data Repository then you may see the following error, “System.Data.SqlClient.SqlException was thrown by method Query on service DataPresenterService: Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding”.
• SRP-1443: An internal error occurs if the database name contains a non-latin character.
• SRP-2167: A user will see a “script stack space quota is exhausted” JavaScript error on Firefox when there is a lot of performance data in an alert details page. Use an alternative browser to view SQL Monitor.
• SRP-2310: “Back in time” can sometime crash with an error when using IE 7. Use an alternative browser to view SQL Monitor.
• SRP-2702: Long running query process ID not appearing in the SQL Processes/Trace tab.

Version 2.1 – 10 December, 2010

New Features:

• License deactivation is now supported.
• Long running query alert now supports excluding by process name.
• Improvements to email format.
• Improvement to Job duration unusual alert.
• Performance improvement when trace in running.
• Long running query: Regex on new lines are treated as distinct new regex rather than a single regex with a new line in it.

Bug Fixes:

• SRP-2126/ SRP-2522: Database creation fails during installation and displays the following error message, “invalid sequential schema version”. This can happen when there are user objects in the model database.
• SRP-1891: The breadcrumbs for any job-specific alert configuration, e.g., “Job failed” alert, contain the name of the instance twice.
• SRP-753/ SRP-425: Sometime XSP webserver crashes with socket exception.
• SRP-2547: Trace not working as expected for SQL Server 2000.
• SRP-2624: “Database **** cannot be opened. It is in the middle of a restore.” was wrongly categorized as an authentication error.
• SRP-2489: Analysis page not plotting Total Size, Data Size and Log Size graphs.
• SRP-2398: Invalid default SQL Server Log path causes connection failure.
• SRP-2475: TypeAccessException thrown when viewing alert inbox.
• SRP-2604: Installer fails to start website service on French Windows XP & 2003 when on a work group.
• SRP-2512: SQL fragment truncated for Long running query, Blocked process, Deadlock and SQL Server error log entry alert.

Known Issues:

• SRP-495: If the Base Monitor loses its connection to the SQL Monitor Data Repository then you may see the following error, “System.Data.SqlClient.SqlException was thrown by method Query on service DataPresenterService: Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding”.
• SRP-1443: An internal error occurs if the database name contains a non-latin character.
• SRP-2167: A user will see a “script stack space quota is exhausted” JavaScript error on Firefox when there is a lot of performance data in an alert details page. Use an alternative browser to view SQL Monitor.
• SRP-2310: “Back in time” can sometime crash with an error when using IE 7. Use an alternative browser to view SQL Monitor.
• SRP-2702: Long running query process ID not appearing in the SQL Processes/Trace tab.
• SRP-2735: The Fragmented index alert is broken in build 2.1.0.380. To obtain a patch to fix this problem, contact Red Gate support.

Version 2.0 – 4 November, 2010

Known Issues:

• If the Base Monitor loses its connection to the SQL Monitor Data Repository then you may see the following error, “System.Data.SqlClient.SqlException was thrown by method Query on service DataPresenterService: Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding”.
• A user will see a “script stack space quota is exhausted” JavaScript error on Firefox when there is a lot of performance data in an alert details page. Use an alternative browser to view SQL Monitor.
• Database creation fails during installation and displays the following error message, “invalid sequential schema version”. This can happen when there is a user object in the model database.
• The breadcrumbs for any job-specific alert configuration, e.g.  ”Job failed” alert, contain the name of the instance twice.
• “Back in time” can sometime crash with an error when using IE 7. Use an alternative browser to view SQL Monitor.

Previous versions of this product were released as SQL Response. See version 1.xx SQL Response release notes.