Performance and Stability Related Fixes in Post-SQL Server 2012 SP1 Builds

Updated to include SP1 CU16: There have already been sixteen Cumulative Updates (CU) for the Service Pack 1 branch of SQL Server 2012. There have been a fairly high number of hotfixes in nearly every one of these Cumulative Updates, as more people are using SQL Server 2012 over the past two years. SQL Server 2012 SP1 has all of the fixes through SQL Server 2012 RTM CU2. If you are running SQL Server 2012, I really think you should be running the latest SQL Server 2012 Service Pack and Cumulative Update. Right now, that means Service Pack 2, CU6 (Build 11.0.5592), which was released on May 18, 2015.

If you are still on the SQL Server 2012 SP1 branch, then you want to be on SP1 CU16.

Table 1 shows the SP1 CU builds that have been released so far.

Build	Description	Release Date
11.0.3000	SP1 RTM	November 7, 2012
11.0.3321	SP1 CU1	November 20, 2012
11.0.3339	SP1 CU2	January 21, 2013
11.0.3349	SP1 CU3	March 18, 2013
11.0.3368	SP1 CU4	May 30, 2013
11.0.3373	SP1 CU5	July 15, 2013
11.0.3381	SP1 CU6	September 16, 2013
11.0.3393	SP1 CU7	November 18, 2013
11.0.3401	SP1 CU8	January 20, 2014
11.0.3412	SP1 CU9	March 17, 2014
11.0.3431	SP1 CU10	May 19, 2014
11.0.3449	SP1 CU11	July 21, 2014
11.0.3470	SP1 CU12	September 15, 2014
11.0.3482	SP1 CU13	November 17, 2014
11.0.3486	SP1 CU14	January 19, 2015
11.0.3487	SP1 CU15	March 16, 2015
11.0.3492	SP1 CU16	May 18, 2015

Table 1: SQL Server 2012 SP1 CU Builds

 

You can follow the KB article link below to see all of the CU builds for the SQL Server 2012 Service Pack 1 branch.

The SQL Server 2012 builds that were released after SQL Server 2012 Service Pack 1 was released

Like I did in my previous post, I decided to scan the hotfix list for all of the Cumulative Updates in the SP1 branch, looking for performance and general reliability-related fixes for the SQL Server Database Engine. I came up with the list below, but this listing is completely arbitrary on my part. You may come up with a completely different list, based on what SQL Server 2012 features you are using.

Here are the fixes in the RTM branch, before Service Pack 1 was released (you will get these when you install SP1):

SQL Server 2012 RTM Cumulative Update 1  (Build 11.0.2316) , 65 total public hot fixes

FIX: Low CPU spikes at a set interval even without user activity in SQL Server 2008 R2 or in SQL Server 2012

FIX: Slow performance when an AFTER trigger runs on a partitioned table in SQL Server 2008 R2 or in SQL Server 2012

FIX: It takes a long time to restore a database in SQL Server 2008 R2 or in SQL Server 2008 or in SQL 2012

FIX: Poor performance when you run a query that contains correlated AND predicates in SQL Server 2008 or in SQL Server 2008 R2 or in SQL Server 2012

FIX: Slow performance occurs in SQL Server 2008 R2 or in SQL Server 2012 if high CPU usage is observed with contention over the QUERY_EXEC_STATS spinlock

FIX: Out-of-memory error when you run SQL Server 2012 on a computer that uses NUMA

SQL Server 2012 RTM Cumulative Update 2  (Build 11.0.2325) , 37 total public hot fixes

FIX: An access violation occurs intermittently when you run a query against a table that has a columnstore index in SQL Server 2012

Here are the fixes in the Service Pack 1 branch:

SQL Server 2012 SP1 Cumulative Update 1 (Build 11.0.3321), 44 total public hot fixes

FIX: “out of memory” error when you use ODBC with SQL Server 2012, SQL Server 2008 R2 or SQL Server 2008

FIX: Performance of a SELECT statement that contains a LIKE operator and an ESCAPE clause is low in SQL Server 2008 R2 or in SQL Server 2012

FIX: SQL Server 2012, SQL Server 2008 R2 or SQL Server 2008 stops responding and a “Non-yielding Scheduler” error is logged

FIX: Slow performance or deadlock when you restore a database and execute statements at the same time in SQL Server 2012

FIX: Worker threads do not wake up immediately when multiple I/O-intensive tasks are running at the same in SQL Server 2012

FIX: “Process appears to be non-yielding on Scheduler ” error message when you run a query in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 2 (Build 11.0.3339), 50 total public hot fixes

A database transaction log continues to grow after you upgrade to SQL Server 2012, SQL Server 2008 R2 SP1, SQL Server 2008 SP2 or SQL Server 2008 SP3

FIX: The size of a database file is not reduced when you use the DBCC SHRINKFILE command in SQL Server 2008 R2 or in SQL Server 2012

FIX: High CPU usage when you query a binary large object column by using the NOLOCK hint in SQL Server 2008 R2 or in SQL Server 2012

FIX: Large queries that modify data run slower than expected when many locks accumulate in a SQL Server 2008 R2 or in SQL Server 2012

FIX: A transaction log restore operation takes longer than expected and a 3402 error occurs if a database file is removed in SQL Server 2008 R2 or in SQL Server 2012

SQL Server 2012 experiences out-of-memory errors

FIX: Deadlocks occur when you execute a stored procedure to alter a temporary table if lock partitioning is enabled in SQL Server 2008 R2 or in SQL Server 2012

FIX: Memory leak if you enable the AUTO_UPDATE_STATISTICS_ASYNC statistics option in SQL Server 2008, in SQL Server 2012 or in SQL Server 2008 R2

FIX: A “non-yielding” error occurs on a server for Service Broker Message Forwarding in SQL Server 2008 R2 or in SQL Server 2012

FIX: Error 17883 when you run a query on a server that has many CPUs and a large amount of memory in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 3 (Build 11.0.3349), 38 total public hot fixes

FIX: Access violation when you run a query that contains many constant values in an IN clause in SQL Server 2008 or in SQL Server 2012

FIX: Memory leak when you run queries against a temporary table in a nested stored procedure in SQL Server 2012

FIX: Poor performance in SQL Server 2012 when you run a SQL Server trace

FIX: You experience poor performance when you run a query against an RCSI-enabled table in SQL Server 2012

FIX: CPU spike when there is no load on a server after you install SQL Server 2012 on the server

SQL Server 2012 experiences performance issues in NUMA environments

SQL Server 2012 SP1 Cumulative Update 4 (Build 11.0.3368), 38 total public hot fixes

FIX: Out-of-memory errors related to a memory clerk in SQL Server 2012

FIX: Out of memory error when you build a columnstore index on partitioned tables in SQL Server 2012

Update that improves the Service Broker when you send messages to remote servers in SQL Server 2012 is available

An update is available for SQL Server 2012 Memory Management

FIX: “Non-yielding Scheduler” error occurs when you insert a row in SQL Server 2012

FIX: You may experience performance issues in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 5 (Build 11.0.3373), 27 total public hot fixes

FIX: Poor performance when you run a query that uses the LIKE operator in the WHERE clause in SQL Server 2008 R2

FIX: Non-yielding scheduler error when you run a stored procedure that uses a TVP in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 6 (Build 11.0.3381), 24 total public hot fixes

FIX: Slow performance on SQL Server 2012 linked server when you update the data together with different collation on the remote server

FIX: A memory leak occurs when you enable AlwaysOn Availability Groups or SQL Server failover cluster in Microsoft SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 7 (Build 11.0.3393), 47 total public hotfixes

FIX: Access violation when you trace an RPC event class by using SQL profiler or XEvents in SQL Server 2012

FIX: Nonclustered index corruption may occur when you run a complex UPDATE statement together with a NOLOCK hint against a table in SQL Server 2008, SQL Server 2008 R2 or SQL Server 2012

FIX: The threads are not scheduled evenly in SQL Server 2012 Standard Edition

A memory leak occurs when a SQL Server Native Client OLE DB provider application calls the Prepare method in SQL Server 2012

FIX: “Non-yielding Resource Monitor” when you run a workload that executes concurrent queries in SQL Server 2012

FIX: Function sys.fn_hadr_backup_is_preferred_replica costs almost all the CPU usage in Log Shipping in SQL Server 2012

FIX: Slow performance in SQL Server when you build an index on a spatial data type of a large table in a SQL Server 2012 instance

FIX: Slow performance in SQL Server 2012 when you build an index on a spatial data type of a large table

An access violation occurs when “sys.dm_db_index_physical_stats” is used in an IF EXISTS statement in an SQL query in SQL Server 2012

FIX: Suboptimal execution plan is generated when you run a query in SQL Server 2012

Slow SQL Server performance and a memory leak occurs after you apply Cumulative Update 3 for SQL Server 2012 Service Pack 1

FIX: Slow performance in SQL Server 2012 when you build an index on a spatial data type of a large table

Slow performance or error messages are logged when you have a transactional replication publisher server in SQL Server 2012

FIX: Insufficient system memory error occurs when you try to create an index in a char, varchar, or nvarchar type column in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 8 (Build 11.0.3401), 32 total public hotfixes

FIX: Data purity corruption in sys.sysbinobjs table in master database when you log on to SQL Server 2008 R2 or SQL Server 2012 by using the SA account and then run DBCC CHECKDB

FIX: Interrupted distributed query may return partial result set without any error in SQL Server 2008 R2 or in SQL Server 2012

FIX: Replication Log Reader Agent fails when you enable CDC and transactional replication for some columns in SQL Server 2008 R2 or SQL Server 2012

FIX: Query that you run against a partitioned table returns incorrect results in SQL Server 2008, SQL Server 2008 R2 or SQL Server 2012

FIX: A small number of rows are sampled in SQL Server 2008 R2 or in SQL Server 2012 when table statistics are updated automatically or together with specified sampling rate

FIX: Access violation may occur when you query sys.dm_server_memory_dumps in SQL Server 2008 R2 or SQL Server 2012

FIX: Replication Queue Reader Agent crashes when you run transactional replication in SQL Server 2008 R2 or SQL Server 2012

FIX: Validation of service accounts takes a long time when you set up SQL Server 2012

FIX: The system function sys.fn_hadr_backup_is_preferred_replica does not work correctly after you have CU7 for SQL Server 2012 SP1 installed

FIX: The query deadlocks when the ALLOW_SNAPSHOT_ISOLATION and READ_COMMITTED_SNAPSHOT are enabled in SQL Server 2012

FIX: The clustered index table may take longer than you expect to be rebuilt when you use the ALTER INDEX REBUILD statement in SQL Server 2008 or SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 9 (Build 11.0.3412), 30 total public hotfixes

FIX: Access violation in SQL Server internal deadlock monitor when you run multiple applications together with Multiple Active Result Sets enabled in SQL Server 2008 R2 or in SQL Server 2012

FIX: Access violation in replication Distribution Agent in SQL Server 2008 R2 or SQL Server 2012 Transactional Replication

FIX: Poor performance when many table-valued parameters are used in SQL Server 2008 R2 or SQL Server 2012

FIX: Merge Agent fails or you experience non-convergence when you use a custom stored procedure conflict resolver in SQL Server 2008 or SQL Server 2008 R2 or SQL Server 2012

FIX: Performance problems occur when database lock activity increases in SQL Server 2012

FIX: Performance problems occur in NUMA environments during foreign page processing in SQL Server 2012

FIX: Operating system version information that is returned from SQL Server 2008 or SQL Server 2008 R2 or SQL Server 2012 is incorrect in Windows 8.1

FIX: Performance decreases after an ALTER INDEX…ONLINE operation is aborted in SQL Server 2012

FIX: Access violation may occur when trigger query joins large dataset in deleted/inserted table and runs in parallel in SQL server 2012

FIX: Database shows “recovery pending” state when you use TDE with EKM provider in SQL Server 2012

FIX: Incorrect usage of built-in FORMAT function brings down SQL Server 2012 Instance

FIX: SQL Server 2012 instance shuts down when you join database as secondary replica during AlwaysOn Availability Groups configuration

FIX: Merge agent fails when you apply snapshot in SQL Server 2012 instance if cross database dependencies are present

FIX: Distribution Agent skips applying sp_MSins_, sp_MSupd_, sp_MSdel_ stored procedures on a Subscriber that is initialized with Backup in SQL Server 2012

FIX: Access violation occurs when you update a table that has an XML calculated column in SQL Server 2012

FIX: SQL Server takes long time to open the databases after the recovery phase when the number of databases or database files or both is large in SQL Server 2012

FIX: Message is not removed from transmission queue even though the ACK is received successfully in an AlwaysOn availability group in SQL Server 2012

FIX: Non-yielding scheduler tries to close the listener for availability group in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 10 (Build 11.0.3431), 36 total public hotfixes

FIX: Intense query compilation workload does not scale with growing number of cores on NUMA hardware and results in CPU saturation in SQL Server 2012

FIX: Assertion failure when you execute a query specifying TOP N and ORDER BY in SQL Server 2008 R2 or SQL Server 2012

FIX: Access violation occurs when you run CHECKTABLE or CHECKDB against a table that has persisted computed columns in SQL Server 2008 R2 or SQL Server 2012

FIX: Incorrect results when you run queries that contain UNION operator by using parallel query plan in SQL Server 2012

FIX: An access violation occurs when you execute update query on a table that has a DML trigger in SQL Server 2008 R2 or SQL Server 2012

FIX: Access violation when the start offset is larger than the string length in function Substring in SQL Server 2012

FIX: Poor cardinality estimation when the ascending key column is branded as stationary in SQL Server 2012

FIX: Poor performance on I/O when you execute select into temporary table operation in SQL Server 2012

FIX: Parallel deadlock or self-deadlock occurs when you run a query that results in parallelism in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 11 (Build 11.0.3449), 32 total public hotfixes

FIX: “Cannot insert duplicate key” error occurs in Spool operation when you update a table in SQL Server 2012

FIX: Incorrect result when you execute a query that uses WITH RECOMPILE option in SQL Server 2012

FIX: Memory leak occurs when you start and stop an XEvent session repeatedly in SQL Server 2012

FIX: Cannot reclaim unused space by using shrink operation in the table that contains a LOB column in SQL Server

FIX: Fatal exception when you run a query that contains CLR functions against an indexed view in SQL Server 2012

FIX: Undetected deadlock occurs when you use a sequence object in SQL Server 2012

FIX: Data corruption occurs in clustered index when you run online index rebuild in SQL Server 2012 or SQL Server 2014

FIX: The database sticks in role configuration phase when you perform a manual failover of a mirrored database

FIX: Error when you execute a Transact-SQL query that contains a long case statement in SQL Server 2012

FIX: Error when you insert a record into a FileTable or FILESTREAM in SQL Server 2012

FIX: Log Reader Agent crashes during initialization when you use transactional replication in SQL Server 2012

FIX: Poor performance when you create spatial index on Point data in SQL Server 2012

FIX: Log Reader Agent fails when you upgrade from SQL Server 2008 to SQL Server 2012

FIX: A severe error occurs when you run a query that uses a hash join parallel execution plan in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 12 (Build 11.0.3470), 14 total public hotfixes

FIX: Error 7105 when you execute UPDATE on sparse column set while ONLINE INDEX REBUILD is running on the same table in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 13 (Build 11.0.3482), 10 total public hotfixes

FIX: High CPU consumption when you use spatial data type and associated methods in SQL Server 2012

FIX: SQL Server is in script upgrade mode for a long time after you apply a SQL Server hotfix or security update 

SQL Server 2012 SP1 Cumulative Update 14 (Build 11.0.3486), 8 total public hotfixes

FIX: SOS_CACHESTORE spinlock contention on system table rowset cache causes high CPU usage in SQL Server 2012

FIX: SOS_CACHESTORE spinlock contention on ad hoc SQL Server plan cache causes high CPU usage in SQL Server 2012

SQL Server 2012 SP1 Cumulative Update 15 (Build 11.0.3487), 1 total public hotfix

The one hotfix in this CU is not particularly interesting, unless you are using Master Data Services (MDS)

SQL Server 2012 SP1 Cumulative Update 16 (Build 11.0.3492), 4 total public hotfixes

FIX: Corruption occurs on the page of secondary replica when you change the secondary replica to unreadable

FIX: Error 33415 occurs when you restart the SQL Server service or alter database in SQL Server 2012

 

Once again, the idea here is to give you a lot of concrete reasons to want to stay current with the latest SQL Server 2012 SP and CU, by pointing out some of the more valuable fixes in each CU in the Service Pack 1 branch.  If my opinion does not sway everyone, this relatively new Microsoft KB article might be more convincing:

Recommended updates and configuration options for SQL Server 2012 and SQL Server 2014 used with high-performance workloads

Happily, Microsoft has been updating this KB article with new information, so you might want to read it again.