2079965 - Targeted CR List for ASE 15.7 SP63 - SAP ASE

CR number

Description

645898

In rare circumstances, after configuring a big data cache or a large stack size, SAP ASE might implicitly shut down following a JavaVM crash while executing a Java query. The message EXCEPTION_ACCESS_VIOLATION or 'ExceptionCode: c0000005' will be reported in the SAP ASE error log.

652328

TEXT/IMAGE page chains are not deleted properly from sysattributes when REORG REBUILD is called on a table with TEXT and IMAGE data. This can result in reduction of unreserved/available space in the database.

664848

In rare circumstances, the message "current process infected with 11 (SIGSEGV)" in the module 'ind__get_sysindrow()' together with a stack trace which includes the modules 'getindexname()' and 'CgGenLava::CGConsScanOp()' may be reported in the SAP ASE error log. This may happen when a statement, using temporary user tables or tables created in the same batch or stored procedure and with indexes defined, is being compiled before execution.

678527

In some long running ASE servers, an 8203 error - "Expected to find the descriptor for object < objid > in database < db_num > in < state > state." might be raised with crt_main in the stack trace while creating a temporary table. There will be an information message "INFO: Total Object Ids in temporary database generated: < total id count > , LWPs generated: < LWP id count > " printed for every 100 million object ids generated

684459

On AIX platform, the mapping of addresses to function names in SAP ASE stack traces has been improved. This will provide a superior mapping when the address is in a shared library.

689004

Setting the configuration parameter 'send doneinproc tokens' to zero, can cause the global variable @@rowcount to report an incorrect value inside a stored procedure or trigger.

715203

DBCC CHECKSTORAGE does not detect LOB compression flag inconsistencies in the TIPSA. With DBCC CHECKTABLE and DBCC CHECKDB these types of inconsistencies were already being reported with message 15960.

715312

Use of online utilities like 'reorg rebuild with online' might lead to incorrect used/unused counts in the OAM for the table. This is reflected by a later 'dbcc checkalloc' on the table reporting error messages 7949 and 7940.

723244

After downgrading from ASE 15.7 to ASE 15.5GA, when booting the ASE 15.5GA server, a segmentation fault may be encountered during start up while attempting to reconcile the downgraded configuration options.

723356

A segmentation fault in tally_fetch() after a 249 error may occur during BCP IN due to invalid column default. Previous CR number is 552227 and the fix is under trace flag -T2781. Please turn on the trace flag to check column default data type during table creation time.

725872

Print extra information for home location of DOL forwarded rows in DBCC PAGE output.

730215

The message, "Current process infected with signal 11" followed by a stack trace that includes the modules d_deleteplan() and proc_plan_get() may be reported in the SAP ASE errorlog when executing DBCC DELETEPLAN.

732613

The Job Scheduler fails to start if quoted_identifier property is set for the connection (which is used to start the Job Scheduler). The following error is logged in the SAP ASE errorlog: "Job Scheduler: IS_ISQL_EXT failed [set nocount on use sybmgmtdb print ":s: %1!", "sybmgmtdb"]"

732693

ALTER TABLE < TABLENAME > SPLIT PARTITION may fail with errors such as 2601 (Attempt to insert duplicate key row in object < name > with unique index < name > ) when there is a non-clustered index.

733194

Incorrect disk activity may be reported by sp_sysmon_diskio when there is no disk I/O in given sample period on this device.

735036

When running concurrent SELECT commands and DELETE/UPDATE commands on TEXT/IMAGE/UNITEXT columns, the client connection may hang or report a TDS token error or raise error 7128 state 6.

735692

The function row_count() when used on system tables returns zero after upgrade from SAP ASE 12.5.4

736622

When running REORG REBUILD against a table whose name or any column name might form part of a reserved double or triple keyword phrase it is possible for an error 156 error to be reported along with error 11934 and for the REORG REBUILD command to abort.

738218

A 1209 error, "Internal Error in lock manager routine: Attempted lock upgrade from shared intent to shared table in parallel query thread is illegal", may be reported in the error log during a SELECT or SELECT-INTO query on a table with parallel plan if there are concurrent DMLs or utilities operating on the table.

740137

The message, "Current process infected with signal 11" (SIGSEGV), followed by a stack trace that includes the module stu_get_next_active_hist_slot() may be reported in the SAP ASE error log when using HASH based statistics gathering and the column has more than 65535 unique values.

743433

Performing decryption of encrypted column data can be much slower for standard users when decrypt default is used.

743558

Error 14613 (The ' < name > ' field in usage anchor has ' < n > ' which does not match with ' < n > from sysusages. Dbid ' < id > '. Lstart ' < n > '.) formats the last 4 parameters incorrectly so the output appears garbled. The field name is correct.

744011

When a user database is binded to a named cache, create index in it might report Error 693 while a dump database process running simutanously.

746028

If a UDR roleid is greater than 1024 then a timeslice error may be reported in the SAP ASE errorlog.

746101

When using sp_downgrade to downgrade an SAP ASE from version 15.7 or higher to an SAP ASE that is version 15.5 based, the keycnt in sysindexes for index sysattributes.ncsysattributes will become incorrect.

747591

MERGE into statement with columns having decrypt defaults in their ON clause may not produce expected results when the user does not have decrypt permission.

748320

Alter table split/ move partition on a table followed by drop column with no datacopy can cause data corruption.

748851

'Invalid pointer value' error messages may occur when a non-database owner accesses the SAP ASE database.

749274

When the monDeviceSpaceUsage table is queried concurrently from different processes, SAP ASE running with threaded mode might report a SIGSEGV (infected with 11) error that mentions the function mda__init_FSList() on the IBM AIX platform.

749279

When a rule is created which is not compatible with the column to which it is bound, SAP ASE generates an error message when it tries to insert a value. After raising the error, an infected with 11 (SIGSEGV) along with a stack trace is reported in the SAP ASE errorlog.

749312

When using a large IN list and the statement cache is on, if the plan chosen has > 96 tables, the query may stack trace during compilation.

750500

Stored procedures with application context functions take a very long time to execute because SAP ASE accesses system catalogs each time for protection checks.

750645

ASE hit SEGV in ct__tds_fetch_cache_cols() and crashes.

750683

In rare circumstances, the execution of Java User Defined Functions (UDF) could mistakenly ignore the arguments it expects when these are presented to it through a single argument that holds a white space separated string.

750719

Error 247:"Arithmetic overflow during implicit conversion of < type > value < value > to a < type > field ." can be seen while running stored procedure sp_spaceusage on large table.

750807

Correct an 814 error situation (Keep count of buffer '0x%lx' in cache '%.*s' holding logical page '%u' in database '%.*s' has become negative).

750937

Sometimes a long running transaction can be seen in tempdb for $sort_local which results in tempdb filling up if a large data set needs to be sorted.

751820

New feature request to allow DUMP TRANSACTION to use the options WITH VERIFY=CRC and VERIFY=READ_AFTER_WRITE as well as LOAD TRANSACTION WITH VERIFY[ONLY]=CRC. These options are applicable only at DUMP TRANSACTION time if the option WITH COMPRESSION is also used. Likewise, the CRC verification at LOAD TRANSACTION time is only applicable if the transaction dump is created with the CRC verification option as well.

751829

Alter table split/ move partition on a table followed by add non materialize column can cause data corruption.

751900

'REORG REBUILD WITH ONLINE' may fail with error 806 (Could not find virtual page for logical page < page# > in database ' < dbname > '.), if the table has undergone an 'ALTER TABLE DROP COLUMN WITH NO DATACOPY' operation.

752115

When the SAP ASE kernel mode is configured as process and an SSL session/connection is abruptly killed/aborted, a SIGSEGV (infected with 11) with a stacktrace may be reported in the SAP ASE errorlog. The functions drop_connection() and ssl_nclose() are likely to be seen as part of the stacktrace.

752391

In rare case, if DOL table has unique index, concurrent dmls may hit index corrupt(error 12313 "Insert into page id < page # > of index id < index # > of table ' < table name > ' (id = < id # > ) in database ' < db name > ' tried to replace undeleted RID ( < > , < > ) at slot < slot # > and ridposn < # > . Try dropping and recreating the index.")x.

752408

In rare circumstances, a 3474 error, "During redo the page timestamp value is less than old timestamp from log....", may be reported during boottime or LOAD DATABASE recovery, when a transaction that allocated a page, rolled back that allocation while a checkpoint was active.

752591

A segmentation violation occurs when the config option enable xact coordination is set to 0 and a connection is established between a primary and secondary server.

752639

When the SAP ASE server is out of locks, a permission check can cause the error message "ulpspinlock: spinlock order violation" to be reported followed by a crash the SAP ASE server.

752660

In rare circumstances, an infected with 11 (SIGSEGV) may be reported in the SAP ASE error log together with a stack trace that includes the modules 'rec_dynamic_prefetch()' and 'rec_undo_session()' when recovery is being run at boot time or during LOAD DATABASE or LOAD TRAN.

752839

The message "current process infected with 11" in the module 'cpynodes' together with a stack trace which includes the modules 's_copytmps' and 'batch_loopend' may be reported in the SAP ASE error log when a BATCH INSERT into a temporary table is executed, followed by 6103 and 703 error while SAP ASE attempts to free procedure cache memory.

753022

SAP ASE running on Windows platforms may report the following message in the SAP ASE errorlog early on during startup: "kernel KERNEL structure not padded correctly - size = 1736". This message may be safely ignored, and this CR prevents it being reported.

753079

For internal use.

753098

In a long running SAP ASE server configured for SSL, SSL handshake errors due to memory allocation failures appear for new incoming connections.

753140

The message "current process infected with 11" in the module 'plc__discard' together with a stack trace which includes the modules 'xact_endupdate' and 'reorg_main' may be reported in the error log during 'REORG REBUILD < table > WITH ONLINE' while this command is executed on a system where high concurrent DML activity is happening on the same table. Alternatively this error can occur during a high concurrent DML activity in a low durability database.

753147

In an HA environment, when the same user tries to login on both primary and companion server at the same time then the login process occassionally hangs.

753461

A segmentation violation occurs when sp_listener starts an SSL listener where no SSL listeners had been started previously.

753525

SAP ASE running on Unix/Linux platforms only prints the top 25 stack frames in a stack overflow stacktrace report which could mask the root cause and also provide insufficient diagnostics when a stack overflow is caused by a recursive function. If the stack overflow backout code uses too much stack, the backout process can cause stack guard word corruption resulting in server shutdown. This CR removes the limitation on number of frames in the stacktrace and if the stack overflow backout code use of the stack reaches within 4K bytes of the stack boundary, it then initiates forcible termination of the task to prevent stack guardword corruption. The default value of 'stack guard size' is increased by 4K bytes to retain earlier level of stack availability. Traceflag 3687 provides a stacktrace when a transaction is aborted due to insufficient stack space that can provide additional diagnostics on stack usage by the query. Traceflag 3686 disables this change. This change is not applicable to the Microsoft Windows platform.

753753

In rare circumstances Online REORG / ONLINE CREATE INDEX may hit SIGSEGV when table contains compressed row but table is marked as not compressed.

753905

When using ALTER DATABASE, an 806 error, "Could not find virtual page for logical page < page# > in database ' < dbname > '..." may be reported if ALTER DATABASE removes more than a single fragment from the end of the given database.

753956

When an insert trigger exists on the target table of the MERGE statement, the wrong result may happen.

753996

An SAP ASE shutdown occurs after hitting a SIGSEGV (infected with 11). In the SIGSEGV message in the SAP ASE errorlog the faulting address may reference the function uppushaffinity().

754075

In some rare conditions, error 8201 - "Keep count of descriptor (objid= < objid > , dbid= < dbid > ) was expected to be < n > . Instead < m > was found.", error severity 26, state 1 might be raised during high availability failover accompanied by a stack trace in the SAP ASE errorlog with the functions des_do_get()/ha_session_check() in it.

754095

On the AIX platform, when SSL is enabled, after a few thousand connections, connection failures occur and error message "ks_falloc: Cannot allocate using ubfalloc" appears in the SAP ASE errorlog.

754273

Client connections to SAP ASE may hang when SSL is enabled and the network memory pool is exhausted.

754283

Backup Server on HPUX may die with error messages: A00: SYBMULTBUF ERROR: Emulator interprocess communication failed with error state = 1, error code=0, system message=Pipe I/O returned 0 bytes. D00: SYBMULTBUF ERROR: Emulator interprocess communication failed with error state = 9, error code=0, system message=Pipe I/O returned 0 bytes while trying to print out a legitimate API error.

754542

Under some rare condition, an infected with 11 (SIGSEGV) error in the function querytuning_ins_sysoptions() may be reported in the SAP ASE errorlog.

754622

Performance of concurrent insert is slow.

754679

In some cases, a DML command is not correctly replicated because SAP ASE Replication Agent may mistakenly generate the table schema definition for metadata reduction for a table that is marked for replication. Schema building is needed when the table is created or altered in the same user transaction as the DML which affects the same table.

754856

When we try to alter temporary table for the database which has proxy database mapped on secondary companion, we hit error 177: CREATE PROXY_TABLE cannot create a temporary object (with '#' as the first character name.

755230

For tables with several indexes and TEXT/IMAGE/UNITEXT columns, a 605 error, "An attempt was made to fetch logical page < page id > from cache < cache > . Page belongs to database < db > , object < object > , index < index > , partition < partition > and not to database < db > , object < object > , index < index > , partition < partition > .", may be reported when running 'SELECT...INTO...FROM ... WHERE condition1 or condition2 or ...' command on it with concurrent DELETE/UPDATE/TRUNCATE TABLE commands.

755287

An assertion will occur if encrypted columns is enabled and there are external passwords that need to be recreated.

755330

When using LDAP, an assertion message and a SIGABRT/SIGIOT (infected with 6) stacktrace may be reported in the SAP ASE errorlog. The assertion will say "result == SYBCSI_RESULT_OK, file core/source/csimemory.c". The stacktrace will include the function login__ldapauth().

755424

When RepAgent is configured for Multiple Paths Replication and verbose LTL traces are active through Repagent traces 9201, 9227 and 9228, a SIGSEGV may occur together with a stack trace showing the module ra__sender_ltl_cmds.

755644

Ocassionally, a heap memory spinlock violation and followed by various stacktraces may happen for session running multiple XML queries with errors.

755682

An ALTER TABLE SPLIT PARTITION command may report a SIGSEGV (infected with 11) in the SAP ASE errorlog if the destination partition name is not specified.

755806

Under extreme OLTP activity in a mixed-log and data database (like tempdb), a task doing CREATE INDEX may report "Stack overflow detected: limit: 0x0x*, guardpage: 0x0x*, sp: 0x0x*" and "*** Stack guardword corrupted" together with a stack trace that includes the modules 'pg__updateoam', 'pg_oamspace', and 'pg_allocoam'.

755887

The error 1120, "Adaptive Server failed to access page < pageno > in database ' < database name > ' because it is not an allocation page." can be raised if the commands DBCC CHECKALLOC or or DBCC TABLEALLOC(syslogs) are executed on an archive database.

756274

When reporting on the dump header of a dump that was taken with a higher SAP ASE version that has used new metadata in the dump, an error 3264, "The dump header contains device information with invalid token length ...", possibly followed by errors such as 3208 “Unexpected end of file while reading beginning of dump...” will be reported. An unknown element warning message will now be reported instead.

756321

The message "Invalid column length: < num > . Value must be between 0 and 255 at offset 0 for 'data-only' row with minimum row length of 70." together with a stack trace including the modules 'collocate' and 'ra__new_schema' may be reported if a row for a replicated table with a name of the maximum length (255 char) created by a user with a name of the maximum length (30 char) is encountered by Replication Agent.

756618

Duplicate identity values may be generated after loading a transaction log dump containing transactions that inserted identity values reserved using multiple calls to RESERVE_IDENTITY().

756631

A 644 error, "Index row entry for data row id ( < pageid > , < rnum > ) is missing from index page < index_pagenum > of index id < indid > of table ' < tablename > ' in database ' < dbname > '. Xactid is ( < xactid > ). Drop and re-create the index." may sometimes be reported in heavily concurrent systems using DATAROWS LOCKED or DATAPAGES LOCKED tables.

756648

A 213 error, "Insert error: column name or number of supplied values does not match table definition." may be reported in a stored procedure, when deferred compilation is active, if a column is added to a table and there is a dependency on "select * " behavior to ignore the extra column. Trace flag 7798 will ignore this error if deferred compilation is turned on and allow normal compilation to process the statement. If normal compilation still sees an inconsistency, or some other activity causes re-resolution of the stored procedure, then errors may be still be reported.

756819

Arithmetic overflow occurred in procedure 'sp_sysmon_kernel'.

756875

When executing a DUMP command with more than one VERIFY option, the backup server will ignore all but the last one. Also, on Windows platform, if the archive is bigger than 4GB and we are executing WITH VERIFY=READ_AFTER_WRITE or VERIFY=CRC and the verification fails when reading a block that is at an offset greater than 4GB, the error message will report an offset value that is only the two lower bytes of the actual offset, for example "There was a failure decompressing < # > bytes at offset 804831240 when processing stripe ' < the stripe name > '" instead of the offset value for 5099798536.

757061

When shared memory size exceeds 1TB, sp_shmdumpconfig might report error 265, "Insufficient result space for explicit conversion of INT value '1382954' to a CHAR field.".

757215

In rare circumstances, a 1105 message "Can't allocate space for object 'syslogs' in database 'tempdb' because 'logsegment' segment is full" may incorrectly be printed for tempdb when the first pages of tempdb are not mapped in sysusages.

757310

Cannot use syconfig to do backup server upgrade on Windows 64bit

757459

A timeslice error, followed by a stacktrace that includes the module memshrink(), may be reported in the SAP ASE errorlog during a sort operation on a very large data set.

757531

RepAgent may fail with error 9275 "RepAgent ( < dbid > ): Could not locate the before image for the INSERT log record at ( < page# > , < rowid# > )." while processing the log records for a MERGE command leading to inserts and updates.

757824

When running in SAP ASE process mode under heavy load in a multiengine environment, an engine may become CPU bound when handling network cancel requests arising from attentions received either from the client or from kill commands. In the event that the network listener was on the run queue of the engine which has become CPU bound as above, it may be impossible to establish new logins to the server.

758353

When using data encryption functionality on Windows platforms, encryption operations can make an engine stall for short periods of time.

758408

Stored procedures using "execute as caller" with queries that create and query #temporary tables fail with error 208, "Server ' < servername > ', Line 26: < #temporary_table > not found after LOAD DATABASE".

758938

When a process that is executing a DUMP DATABASE command is killed either using the KILL command or using CTRL-C or the SQL session goes away, the Backup Server will not release its associated resources until it finishes copying the whole database. One of the consequences is that the archive devices can remain locked for a long time because the sybmultbuf processes stay alive.

758973

A SIGSEGV (infected with 11) can be logged in the SAP ASE errorlog on the Windows platform if the loginfo() built-in is called with three parameters and for whatever reason the built-in returns an error.

759377

In very rare situations, SAP ASE may panic and shutdown if sp_sysmon is being run and a connection disconnects at the same time.

759501

Allow the command LOAD WITH VERIFYONLY to be executed using less stripes than the number used at DUMP time.

759624

On ASE Cluster Edition insert jobs may unexpectedly continue to run with no new rows inserted into the target table. There are NO lost inserts but, dbcc checkalloc or tablealloc may report Msg 2546 or we may encounter deadlocks which can be seen in the error log.

759637

The message, "Current process infected with signal 11", followed by a stack trace that includes the module LeScanOp::_LeOpNext() may be reported in the error log when an online utility (e.g. REORG REBUILD) is in progress on a table that contains forwarded rows.

759646

Arithmetic overflow may occur when executing procedure sp_sysmon_locks.

760187

The index ID is incorrectly reported as 0 in the SAP ASE error message number 12301, "Internal error: the index (id = < ID > ) and the data for row ( < page > , < row > ) in table ' < table > ', database < db > are inconsistent; data is marked deleted but index is not. Aborting the transaction".

760390

A DUMP WITH COMPRESSION could fail with the message "SYBMULTBUF ERROR: Emulator interprocess communication failed with error state = 9, error code=0, system message=Pipe I/O returned 0 bytes". If a core file is produced the stack will show the module sm__do_allocations().

760539

SAP ASE fails to boot on Windows with the message "FATAL: shared memory region is being held at xxx bytes but yyy bytes are required" being reported in the SAP ASE errorlog.

760672

A 3478 error, "During undo the page timestamp value is less than new timestamp from log." or other fatal errors may be reported in the error log when SAP ASE is rolling back a transaction following an 863 error, "Buffer resources in cache < cache_name > are unavailable. Please re-run this query or ask the system administrator to re-configure buffer cache memory." that is accompanied with messages "Named cache < cache_name, cache_id > is configured too small for current demands".

760673

If the data cache is configured small, SAP ASE may report a 'cache too small' error when the new sort buffer algorithm is used.

760875

sp_sysmon and monEngine do not report DiskIOChecks in process mode.

760876

Deprecate plan sharing feature.

761117

DML on tables with Encrypted Columns are significantly slower than SP50.

761155

For SAP ASE Cluster Edition, when table creation is in progress, the checkpoint process may hit an infected with 11 (SIGSEGV) in the module 'ocm_hold()' and report a stacktrace in the SAP ASE errorlog that includes the modules 'des_has_onlutl_active()' and 'cm_writedes()'.

761275

When there are several scans on the same index and some of the scans hit the same index page needed to be shrunk concurrently, the index page may be shrunk twice. The second attempt will fail and a 2630 error, "In database < db > , page < page#2 > is linked backward to page < page#1 > , but that page is linked forward to page < page#3 > . Please report this internal error to SAP Technical Support". Although a 2630 error is raised, the index is not corrupted.

761385

The message, "Current process infected with signal 10 (SIGBUS)", followed by a stack trace that includes the modules 'loginfo_builtin' and 'exec_lava' may be reported in the error log when the builtin loginfo(' < dbname > ','oldest_active_transaction_date') is executed.

761410

UPDATE STATISTICS WITH SAMPLING may report error 614: "Adaptive Server accessed a row that has an illegal length of < n > while in data base ' < db > ." or error 806: "Could not find virtual page for logical page < n > in database ' < db > ', while concurrent DML is going on the same object.

761518

A 7786 error, "Unable to turn role off since user is not a valid user in the current database", may be reported incorrectly when executing the SET ROLE OFF command.

761555

Some of the cache buffers used for sort are not claimed and unclaimed correctly. This can lead to a performance impact if cache swap occurs.

761836

New Feature: Bucket Pool memory manager to enable ASE scale over large CPU core configuration.

761881

BCP in to a table with encrypted columns is much slower with the change from the use of the Certicom libraries.

762223

When granular permissions is enabled and the user who has "own database" permission has executed setuser " < another_user > ", the user is not able to select from < another_user > .table.

762266

Under a rare condition, the fix for CR 750937 could cause a performance regression. The fix for 750937 is now only enabled under traceflag 219.

762281

The version of OpenSSL used has been upgraded to OpenSSL 1.0.1g

762960

When permission on built-in function is granted to a role which is granted to a user, then executing the built-in function may result in permission error.

763247

Using a NOT LIKE clause to compare with a local variable assigned to a constant string could return different results than comparing to the constant string directly if the column has NULL values in it.

763271

For the diagserver on Solaris platforms, if SAP ASE is running in threaded kernel mode, setting the congiguration option "number of disk tasks" to any value greater than 1, will result in an assertion failure being reported in the SAP ASE errorlog.

763301

The ASE COTS JavaVM version is updated to version JRE 6 update 45 for each supported platform in the sybpcidb database.

763376

A 6908 error, "During redo the page timestamp value is less than old timestamp from log. Page .. object id = 99, page timestamp=0000 00000000. Log old timestamp=0000 00000001." may be reported during load tran in a database that has been altered with the "async_init" option.

763514

A 2510 error, "Key mismatch between index page < value > , partition ID < value > ;... Drop and re-create index ID < value > of table ' < value > ' in database ' < value > '." followed by the message "current process infected with 11" in the module 'byte_eq' together with a stack trace which includes the modules 'compress_page' and 's_xoltp_dml' may be reported in the error log when an INSERT is executed and while the configuration option "streamlined dynamic SQL" is active. Additionally a 631 error will be reported while SAP ASE is rolling back the current transaction.

764862

The message text for error 3543 is enhanced to include the database id value such as "One or more entries in the data segment free space counts in database ID %d was found to be negative and has been set to 0."

764945

In rare circumstances, the execution of Java User Defined Functions (UDF) could mistakenly ignore the arguments it expects when these are presented to it through a single argument that holds a white space separated string.