2081389 - Targeted CR List for ASE 15.5 ESD#5

CR number

Description

466609

Repartitioned exchange can cause poor performance and in some cases we do not need to incur that performance penalty. Instead, it can be handled during a selective index scan, provided the column on which the data needs to be partitioned is part of the index.

475469

Message 4207 Dump transaction is not allowed because a non-logged operation was performed on the database will be raised if the SELECT INTO EXISTING TABLE syntax is used even if the SELECT INTO / BULK COPY dboption is turned off. The SELECT INTO EXISTING syntax was meant for internal use and not intended to be documented or used by end-users.

488602

Enable memusage and enhance who command in sybmon

571074

Specifying fill factor for a table has no effect on the subsequent reorg rebuild.

575761

The index_union algorithm can be invoked when queries contain the OR operator or IN list operator. This algorithm was not permitted when selecting data at isolation level 0 (dirty reads) on dol tables. This algorithm will now be permitted on DOL tables with dirty reads when optcriteria cr575761 is active, and/or trace flag 16734 is on. The behavior change will be turned off if trace flag 16733 is on. If the algorithm is chosen then index_union appear as a hash union of several index scans followed by rid join in the showplan output. The abstract plan would contain a hash_union_scan if this algorithm was chosen with dirty reads on DOL tables.

589686

Feature request to have Msg 935 (WARNING - the timestamp in database '%.*s' is approaching the maximum allowed) always be sent to the errorlog when it is raised.

590312

If reformating strategy is used and an index key has BIT datatype, it could return wrong result sometimes.

608323

The message "Warning: Row size ( < value > bytes) could exceed row size limit, which is 1962 bytes." is raised when installing the system stored procedures sp_help, sp_helpdb, sp_helpindex and sp_help_rep_agent.

618881

An infected with 11 involving operator LeHashOp or LeGroupSortedOp may be raised when selecting an aggregate function such as count() from a DISTINCT derived table based on an equijoin between local and proxy tables and the equijoin is on char/varchar type columns.

620806

For insert..select queries with an order-by clause, the order-by clause may be ignored if the table being inserted into has referential integrity checks.

621863

When a DISTINCT view (or derived table) is queried in a parent DISTINCT query or is part of an EXISTS join (e.g. under IN/Any subquery), signal 11 stacktrace could happen in CgpPop::_CgpPePredBmToScalar() for distinct operator (CgpDistinctHashing/CgpDistinctSorting/CgpDistinctSorted) when the following three conditions are all true: 1. The DISTINCT view/derived table has at least one expression SELECT column; 2. The parent SELECT DISTINCT columns are only a subset of the DISTINCT view columns. Or if the EXISTS join is on one of the DISTINCT view columns; 3. One of DISTINCT view columns which is not part of 2) has a predicate on it in the parent query block.

631758

Varchar field for value1 and value2 in sp_dbcc_run_differentialreport have insufficent length, thus causes conversion error.

632076

Alter table statement containing modify clause and a clause for addition of a table level constraint together hits sig 11.

635663

Trace 15331 with parallel select on syscomments stacks with 1289 error

636277

Avoid interrupting the load of a 12.x dump on a 15.x server, as this can lead to reports of 605 errors on a subsequent attempt to load the database.

636684

sp_unbindexeclass fails with error 512 when more than one row qualify for the delete operation.

637395

In the Cluster Edition, when monitoring is enabled, the MaxIOServiceTime in monOpenPartitionActivity table may sometimes have value less than the AvgIOServiceTime.

637589

Signal 11 in function ScalarNormalizing() (amongst others) may be seen in the error log and the spid terminated when attempting to insert < table > ( < column-list > )..select < col > from < table2 > union/union all < select > in a procedure where < table > is created inside the procedure and < table > contains non-nullable columns that are not in < column-list > .

637600

Signal 11 stacktrace with error msg 416 may be seen in the error log and the spid terminated when attempting to insert < table > ( < column-list > )..select < col > from < table2 > union/union all < select > in a procedure where < table > is created inside the procedure and < table > contains non-nullable columns that are not in < column-list > .

638556

A 623 error, "Adaptive Server failed to retrieve a row via its RID in database ' < dbname > ' because the target page is not a data page... pageno = 0 ..." may be reported during the execution of LOAD TRANSACTION command using the WITH UNTIL_TIME option from a transcation dump that was dumped using the WITH NO_TRUNCATE option.

639045

Under some rare condition, Msg 913 might raise when pagesize() is used with other builtin functions or UDF functions in a query.

639969

Msg 530 might raise when trying to execute queries involving derived table and aggregation function.

641091

When running 'sp_configure 'mnc_full_index_filter', 3' which will set the optimizer criterion full_index_filter=1, you may see a poorly performing plan in circumstances whereby a query is running under rowcount (or top) and the lead order by column doesn't have an entry in the predicate and is the lead key column of an index. In this circumstance, the index will not be costed and if this table is placed outermost in the plan you may see a complete scan of the table and a sort to generate the order prior to application of the row count.

641535

sp_sysmon output may sometimes show negative value for the total task context switches due to "Other causes".

642345

Under some rare circumstances when the "no free space acctg" database option is not set, the time taken to recalculate the free space counts after an impolite shutdown can be unnecessarily long.

646024

Sort operators are undercosted in fastfirstrow scenarios. Fastfirstrow scenarios include min/max optimizations, exists scans and the fastfirstrow optgoal. This can lead to discarding fastfirstrow strategies in preference to other less efficient strategies incorporating sorts. The fix for this is enabled by the cr646024 optcriteria.

646141

Print warning if timestamp of a joining instance is not in sync with that of the existing instances.

647239

dbcc proc_cache("free_unused") doesn't refill other engine's ELC.

647569

Certain parallel query plans available in ASE releases before 15.0 are not supported in ASE 15.0. Cases include joins between partitioned and unpartitioned tables. This change adds capabilities and costing support for these queries. Changes are enabled using trace flags 16970 and 15331. You may also need to sp_chgattribute to set the plldegree on a per table basis.

647801

A 691 error, "Encountered invalid logical page '0' while accessing database 'db_name' (id), object 'table_name' (id), index 'index_name' (255),..." or a 14108 error, "Could not find partition descriptor for objid < ... > ..." together with a stack trace which includes the modules 'get_txtpage' and 'instext' may be reported in the error log when running a SELECT INTO UNION which requires to use an internal worktable (for example, because of an ORDER BY clause) and a NULL literal value is used in one of the UNION sides for a TEXT, UNITEXT or IMAGE column.

648511

If sp_passwordpolicy "expire login passwords" is executed without specifying a login or wildcard string, it expires every login on the server.

648848

OpenSSL error while dumping the tran log SSL or Crypto Error Info: psn -1, vsn -1, sockp (nil) error id 96, severity -2, provider id 2. SSL or Crypto Error Message: 'An unexpected failure occurred while performing an OpenSSL cryptography operation. Root error: error:0606506D:digital envelope routines:EVP_DecryptFinal_ex:wrong final block length.'.

649038

Under some rare circumstance, parallel queries using expression involving LIKE and variables or parameters could cause stacktrace and the stack will show modules copy__eval() and copy__eoplike().

650076

Cluster lock space rebuild takes time and as CLM rebuild is blocking operation cluster is hung for the duration.

650839

sp_plan_dbccdb may raised error 3606 arithmetic overflow if the server contains very large databases or very large devices. The device size values displayed for recommended devices is also not properly adjusted for server page size.

653157

In Adaptive Server/Cluster Edition, the opening of databases when booting a non-coordinator instance, is performed serially so that making all databases available takes an unnecessarily long period of time. The opening of databases is now done in parallel. The previous behavior has been retained under the 2796 trace flag.

653722

An infected with 11 occurs in NormalizeScaler when inserting the results of a UNION into a table with a materialized functional column.

654233

Adaptive server goes down after an assertion is hit in the module th__refresh_threshcache(), when running alter database command if systhresholds system table is corrupted and diagserver is used.

654617

Enable CIPC native thread on HP platform.

654653

The database option "no chkpt on recovery" may be left set for a given database mistakenly when ALTER DATABASE and QUIESCE DATABASE are executed by two sessions concurrently.

657522

After a polite ASE shutdown, no_recovery durability databases whose log segment is mixed with data segments may have inaccurate free space counts.

659910

Query with top, union all and order by clause could return fewer rows.

660311

Less rows than expected may be returned when executing a query which joins 3 or more tables with All Pages Locking schema, there are some duplicate conditions in the WHERE clause and the SET SHOWPLAN ON output shows that it used a N-ARY NESTED LOOP JOIN Operator.

661265

A 7105 error, "TEXT/IMAGE/UNITEXT page < page > does not have a next page" together with a stack trace may be reported when input text string containing partial character at the end is inserted into Adaptive Server which has the same character set as client by BCP...IN.

661494

Under multi-byte character sets, selecting rows from text columns may cause more IOs comparing to single-byte character set.

661508

A 3478 error: "During undo the page timestamp value is less than new timestamp from log ... ", may be reported during reboot if a REORG REBUILD command was active at the time the server was killed.

661677

After a cross-platform dump-load, some of the histogram statistics, used by the optimizer, may be incorrect.

663109

Under some very rare circumstances, one of the cluster instances may abort with an infected with 11 message together with a stack trace showing module name 'dbt_rdump_action_pcm_callback'. This may happen after a DUMP DATABASE command was interrupted or hit an exception.

663203

Error 2502, "Table Corrupt: A page is linked in more than one chain; check this page:page number= < page number > allocation status= < status value > ", followed by error 7968, "The object allocation map for tabid < object id > indid < index id > partition < partition nid > is corrupt, therefore the optimized report cannot be generated. Contact Technical Support for assistance.", will be reported when running "dbcc textalloc" with "noreport" option multiple times.

663342

There could be a memory corruption during page allocation either in case of a MLD database when another task is doing the allocation on the same allocation page or with any database if the buffer for the new page could not be allocated.

663359

A 9236 error, "Unknown dbid '0' specified. Please check the dbid specified." may be reported when running SP_HELP_REP_AGENT with no parameter values.

663606

BCP-in performance on a large table having many indices is significantly worse than the performance achieved using 12.5.4

663628

Worktables will not be shown in monProcessObject.

663945

When creating a functional based index, data may get truncated when using CONVERT().

663994

Signal 11 could occur in CgEvals::_CgCompVcol(tree*) during DistinctHashing operator code generation when there are two distinct expression columns and one is becoming dependent on the other through equijoin transitive closure. Using diagserver, the assertion "Assertion (GcvIsVirtualColumn(pNode))" could fail in cgevals.cpp line 202.

664112

sqllocres does not check for valid default language/character set/sort order attributes in the resource file.

664161

The message "current process infected with 11" in the module 'fglk__send_prom_attn' together with a stack trace which includes the modules 'fglk__promote' and 'getnext' may be reported in the error log and may bring the server down if a query involving a big number of tables (more than 50) was being executed in parallel and due to the number of locks acquired it attempted to promote the locks to a lock of a higher level.

664703

A query against an MDA table such as monLocks may encounter a timeslice error in VTABRemoteAccess::getNext() during a period of heavy user activity.

664984

When executing an INSERT-SELECT query from a proxy table having a LOB column into a local table, a NULL value may be inserted into a LOB column on the local table if a cursor is opened on any proxy table.

665170

sp_shmdumpconfig raises message 18502 when an "error" dump type is specified when "message" should have been specified. This CR is a feature request to find ways to reduce the confusion between the two types of dump conditions.

665639

When 'statement cache' and 'literal autoparam' are configured on ASE the @@identity retains a non-zero value after an INSERT into a table that contains no IDENTITY column.

665651

If server has 'capture missing statistics' enabled, and if multiple parallel threads execute queries involving a column for which stats are missing, then all such threads would try to insert the missing stats; one of the threads would succeed, and the others would get 2601 error leading to transaction and command rollback.

666362

A SIG 11 may be reported with a call stack involving do_bt_release_rlocks()and rec_undo_session() functions while accessing the buffer address from an address lock of an APL page.

666371

TF2792 blocking on user table for select operation.

666460

Parallel BCP connections on an indexed DOL table might block each other while loading data into the same data partition.

666577

Cannot access Monitor Server counters from Sybase Central when SQL Server name are different in client and server side interfaces file.

666720

If user defines an index whose length is larger than 32767 on DOL table, there may be some problems encountered during index DML.

667025

sqllocres reports sort order "binary" is not valid for character set "utf8".

667065

Under some rare condition, queries with computed columns could cause stack traces with reference to add__computedcol module when using the table with computed columns.

667552

sp_sysmon might report wrong engines when some engines are offline.

667726

sp_helprotect truncates object and column names which are greater than 30.

667923

A 290024 error, "Invalid column length: < number > Value must be between < number > and < number > at offset < number > ..." may be reported in the errorlog when a session executed a trigger which contains a SELECT FOR XML referencing INSERTED or DELETED virtual tables and previously the same session had been executed a different trigger for a different table but with the same SELECT FOR XML statement.

667943

In some cases, low throughput or performance of partially logged operations like BCP on indexed table or reorg rebuild. sp_sysmon output shows high contention on dbt_seqspin spinlock.

667977

When a table in a view is modified, multiple select statements on the view could return different results.

668380

When free lock pool resource goes below critical threshold, server goes to slow response state due to heavy spinlock contention.

668502

For a DOL partitioned table which has text/unitext/image column(s), after alter table dropping a column, dbcc textalloc for this table which still have some text/unitext/image column(s) will report 15918 error.

668846

PIOs, LIOs, and CPU for nested loop join index scans are not reported in the estimated cost diagnostics (statistics plancost) for parallel optimized strategies.

668909

When creating a lightweight procedure, if all negative object ids are used up, error msg 2012 is not returned.

669413

Backup Server support for MOUNTWait option for Tivoli Storage Manager

669429

Infected with 11 in function testpasswd following an upgrade.

669736

Large text and data segments of the binary are not supported on hpia64 platform.

669932

Sybase Central, ASE plugin: Web Services SSL Configuration wizard fails with: Using SYBASE_JRE6 as null ERROR: Could not find java executable at null/bin/java. Please verify that a valid JRE installation exists.

670074

High spinlock contention on Resource- > rdbts_spin and Dbtable- > dbt_seqspin when running concurrent bcp-in jobs against a partitioned table.

670169

Sybase Central: ASE Plugin is not able to browse 'User Tables' in a database when it contains a table that is marked 'for load'.

670199

Infected with signal 11 could occur in function ResRelOp() called from CgpDistinctSorted::_cgpInit() when a DISTINCT view with an expression distinct SELECT column is queried with a constant equality predicate over the expression column and the final plan chooses DistinctSorted operator.

670310

The message "PCI(M1): pcidb_get_pci_directive_arg() : Unable to select the arguments for PCI Bridge directive PCI_BRIDGE_X_OPT." followed by the message "current process infected with 11" in the module 'pci__cfg_plb_options' together with a stack trace which includes the modules 'dsinit' and 'pci_cfgmgr_init' may be reported in the error log during ASE startup whose logical page size is greater than 8K if ASE first reported an error during the recovery of a user database and the configuration option 'enable pci' is active.

670907

ASE may crash unrecoverably with signal 11 in ct_poll when a connection to a remote server running an SQL query is killed.

670973

When more than one column is encrypted with same key encrypted by user password, even if wrong password is specified for one of the columns during bcp out, no error is given and bcp succeeds.

671437

Parallel query plans are not chosen over serial query plans when using a large number of engines due to over optimistic costing of serial query plans.

671440

Under some rare circumstances, "online database" or "load tran" command may fail with a stack trace showing module xls_pre_alloc_numclrs and plc__flush. This may happen after loading a transaction dump including log records written by a 'create index' command run for a DOL table.

671536

In certain circumstances, a query that contains an ‘or’ clause with multiple joins may generate a sub-optimal plan, with respect to result accurracy, when executed with the optimiser level of ‘ase_default’.

671592

Lots of concurrent logins result in error 290024 and SIG11 during ASE authentication.

671752

Truncation error occurred at Data Cache Manangement section causes spinlock contention information to not be reported on sp_sysmon

672144

A 2540 error may be reported by DBCC CHECKALLOC following instance failover recovery when a REORG REBUILD (index), CREATE INDEX, DROP TABLE or TRUNCATE TABLE command had completed on the instance just prior to its failure.

672181

Backup Server: On 64-bits Windows platforms, dumps to existing files do not reset the file size when the dump size is less.

672196

MERGE UNION DISTINCT plan could return duplicate rows incorrectly when the union view/derived_table is participating in an equijoin join with another table and there is also a constant sarg (col= < const > ) on the other table's equijoin column.

672829

There is high spinlock contention on network memory pool spinlock.

673179

A user defined function or a stored procedure that calls the radians() function with numeric parameter may start returning NULL for all executions once it has been called once in a way that passed NULL to the radians() function.

673379

Under some rare circumstances, select-into on union view may fail with error 225 (Cannot run query--referenced object (name NAME NOT RECOVERABLE) dropped during query optimization.)

673487

In limited cases, parallel group-by queries that use 2phase grouping strategies and have two or more xchg operators beneath the 2phase grouping operator may return duplicate groups with partial aggregate results.

673494

Historical Server will hang if one attempts a recording session and the monitor server name doesn't start with a letter or this name does't exist in the interface file.

673625

Sub optimal plan with INSERT and/or SELECT INTO using an ORDER BY occurring when optcriteria avoid_bmo_sorts is on

674235

A 8213 error, "Failed to acquire address lock on object < value > " may be reported in the error log preceded by a stack trace which includes the modules 'des_idtnewval' and 'des__errhdl'. The stack trace information should not be reported.

674251

A roundhash N to roundrobin 1 NLJ strategy may return too few rows if compile-time thread adjustment is done. Typically, compile-time thread adjustment might occur if "max parallel degree" is not configured at least one (or more) higher than "max scan parallel degree".

674278

In limited scenarios, right nested outer joins optimized in parallel mode could generate a SEGV. This would occur if the outer stream of the right nested OJ had replicated partitioning. The stack would contain a calls to OptDeepPDTransform() and eqcGenNlJoinPop().

674607

The threads required to run a parallel query may be under calculated. The problem occurs in plans that contain store_index or N-to-1 Nlj strategies. This can lead to the selection of plans that require more threads than the system/query was configured for. This fix is enabled with optcriteria cr647607.

674622

Query got incorrect result when like clause with long like term and having escape char.

674849

If the SYBASE_SAM_CAPACITY environment variable is set prior to running the ASE installer in a dedicated CPU logical partition environment, you can't proceed to next panel after specifying the SySAM License Type.

674877

Performance issue on select into from encrypted tables for non-privileged user

675179

Possible incorrect index selection when using OR predicates with equalities referencing multiple attributes of an index. For example, (a = 5 and b = 6) OR (a = 15 or b = 16) when an index exists on (a,b) could be incorrect (notice the multiple attributes, the issue does not exist for IN lists). The change is enabled thru optcriteria cr675179

675990

After cross-platform dump and load of a database containing a view defined on a base table with an encrypted column and a decrypt default, the decrypt default information may be lost leading to a 3762 error: "The decrypt_default for column id 27 was not found in the system catalogs."

676195

Following installation of ASE 15.5 ESD#3 on Windows, the gui tool Server Config appears to be unaware of character set cp1250.

677059

When parse_text() builtin with execute immediate is run concurrently in many sessions with each session running the query multiple times in a loop, it will result in segmentation violation in prot_desc().

677101

In a high volume server with many processes accessing and updating a DRL table with a TEXT/IMAGE or off-row JAVA column, sometimes it could return inconsistent data.

677153

When loading a dump generated by an ASE 12.5 (which is not AIX) to an ASE 15.5 AIX server, an invalid column length error may be raised raised.

677580

When executing SP_SETREPTABLE with the parameter value 'USE_INDEX', the indexes created for the replication of off-row text/image columns may become inconsistent when the table does not have any other non-clustered index. In that case 'DBCC CHECKTABLE' would report error 7928. 'UPDATE' or 'WRITETEXT' may fail with error 644 or error 7165 respectively.

677991

Sometimes after a failed attempt to drop a large buffer pool in a busy system, the pool may still be be available for use even if sp_cacheconfig shows the pool status as "Unavailable/deleted". However, subsequent use of the pool may, under rare circumstances, lead to timeslice error with modules like pg_allocate(), bufnewpage() and cm_cache_wash() reported in the stack.

678184

In rare circumstances, checkstorage may report the error 835, "Illegal attempt to change contents of buffer" followed by a stack trace that includes the module pchk_copy_page().

678278

BCP into a range partitioned DOL table can result in excessive unused space.

678287

Memory allocated for the internal ASE server DES is not freed after a failed connection attempt that results in error 7218. The entire memory pool runs out with an error 11223 after 1024 failed connection attempts.

678432

If a target list for a view maps to more than one encrypted column in the base table with the decrypt_default attribute, Adaptive Server may fail to check the user's decrypt permission and incorrectly return the decrypt_default value instead of the actual column value.

678639

Closing a connection in ASE may cause high cpu usage.

678736

Upgrading from 15.5 ESD to 15.7 release, if tables have partitions, ASE may generate seg fault involving functions s_readparseProcText() - > s_cpytxt().

678812

Sybmigrate hits a 247 error (Arithmetic overflow during explicit conversion of NUMERIC value to a INT field) on tables with more than 2.1 billion rows.

679134

JS fails to execute job with "Multiple password found" error even when sp_addexternlogin was done only once for a user.

679449

Installer displays some messages in English when run in non-English environment.

679497

Linux/pSeries installer throws java.lang.UnsatisfiedLinkError exception when verify SySAM license.

679807

For a select statement in a stored procedure, if the table has more then one partition keys and not all the partition keys in the predicates are constants, Neither compiler time partition elimination nor dynamic partition elimination occured.

680360

In Adaptive Server version 15.5 ESD#3 or later, failure to move requested memory between buffer pools using the sp_poolconfig system stored procedure, may lead to error 2747: "Incorrect position specifier in format string of PRINT command" followed by message: "Transaction count after EXECUTE indicates that a COMMIT or ROLLBACK TRAN is missing. Previous count = X, Current count = X+1".

680424

Sybase Central Editor used by ASE Plug-in does not have the help menu correctly configured.

681107

In Sybase Central ASEPlugin, Customer click database - > right click Property - > device tab then got Java Error(java.lang.Integer cannot be cast to java.lang.String) when the database has user-defined segment.

681538

When installing ASE using one off release EBF #19426, misleading warnings about connectivity may be reported. These can be ignored.

681666

The message "Error #0; Result Table was expected but not received" may be reported in sybcentral when running "Object Page IO Monitor" in monitors folder.

681817

When configuration parameter 'FIPS login password encryption' is enabled and there is no valid ASE_ASM license, sp_configure unexpectedly succeeds. License checks done at ASE startup correctly report that no license is available, but unexpectedly allow continued use of the FIPS login password encryption configuration option.

682864

Error 17290 can be reported by syconfig when creating a backupserver even though it is correctly created and setup in the ASE.

683062

The description in the installer Configure Self Management and other panels are truncated.

683192

After downgrade from ASE 15.7 to ASE 15.5, the booting 15.5 server can encounter a segmentation fault during startup, while attempting to reconcile downgraded configuration options. This prevents successful completion of the server boot.

683455

If user has multiple Sybase products installed to the same directory, uninstaller may throw java.lang.NullPointerException exception.

684722

On non-English locale, uninstall prompt java.exe not find error during dbisql and sc uninstall

685000

Sybmigrate: if the version of source server is ASE 15.0, the exception happens "Error while getting the configuration of net password encryption reqd for server".

685300

If user installed only the Sybase Control Center Framework, installer reports SCC failed to start even though it was successfully started.

686131

I/O fencing on AIX uses raw SCSI-3 PGR commands which may not be supported on some IBM storage devices. ASE will display I/O fencing reservation errors and will fail to boot.

686160

In the case of working with DOL tables that are marked as hot objects using dbcc tune(..des_bind) there is still contention on the locks.

686186

Wrong result could happen when using an IN clause over a complex union view column and one side of the union is directly projecting a constant for the column and the constant is also part of the IN clause list.

686209

When statement cache is on and after a variable assignment from a column of a table statement is cached, it will return wrong results for the subsequent select the variable value if the lwp for the statement need to be recompiled.

686385

HA SC 3.x: kill_ase_engines may spend lots of time constructing the pid list.

686481

311 errors can occur on every second execution of a stored procedure that uses dirty reads on a #temptable/index created within the stored procedure.

687106

In syconfig.exe, many sort order options are missing if user chooses to set utf8 as the default character set.

687735

Populating a proxy table that contains TEXT columns with several hundred rows of data using INSERT INTO... SELECT may fail with Error 3651, "This transaction has been rolled back, rather than only the current statement."

688338

Query involving column relop null clause could hit 403 error msg if column projected NULL and relop could be =, > =, > , < =, < .

691255

Grouping by univarchar column may return wrong result if the univarchar column contains NULL and pure blank values and a group hashing plan is chosen.