Resolved Issues

Parameterized SQL queries that include LIKE, STARTING, or ESCAPE with a NULL parameter could cause the Storage Manager (SM) to shut down unexpectedly, reporting "Cannot convert <null> into string". (14774)
The issue has been resolved.

When a SQL statement results in duplicate keys error on a unique index containing BINARY, VARBINARY, or BYTES columns, the error message could include unreadable key values or misleading information related to decoding values. (14756)
The issue has been resolved.

Execution of SQL queries that include a HAVING clause on a group key could return inaccurate results.
The issue has been resolved.

Execution of a SQL query that includes 64 tables or 63 joins could report "Internal error: failed precondition _plan.expression(p).referencedRelations().is_subset_of(available), file xxx".
The issue has been resolved.

A SQL query that uses decorrelated dependent existential Right Hash Joins could return inaccurate results when NULL values are included in the existential comparison.
The issue has been resolved.

A SQL query that includes multiple join operations with at least one dependent join and a LIMIT clause could return inaccurate results.
The issue has been resolved.

A SQL UPDATE statement with a JOIN clause could result in error because update checks were also performed on joined views which are not updatable.
The issue has been resolved.

Execution of SQL queries with correlated subqueries and non-inner joins could terminate and report a precondition error.
The issue has been resolved.

Execution of SQL statements with multiple Common Table Expressions (CTEs) defined with the same name in the same scope of a SQL query used only the first definition, and all the subsequent definitions were ignored without any warning.
The issue has been resolved.
Now the execution of such statements are blocked and an error message is reported. For more information, see WITH.

Repeated execution of SQL queries that use index scans could take longer because the state of index scans was not reused.
The issue has been resolved.

When the KEY_ORDERED optimizer hint was specified in a SQL query, the hint was ignored if the optimizer elected to use Parallel query execution.
This issue has been resolved.
The optimizer now prioritizes the KEY_ORDERED hint over Parallel query execution.

When using nuosql, executing user-defined stored procedures with EXECUTE could return results without column names.
The issue has been resolved.

The STARTID field in the CONNECTIONS system table may contain -1 instead of the correct start ID of the Transaction Engines (TEs).
The issue has been resolved.

In rare cases, archive synchronization may run indefinitely while the log reports 0 catalogs remaining to sync.
The issue has been resolved.

A SQL query that includes a precision parameter in the CURRENT_TIMESTAMP function could report an error. (14473)
The issue has been resolved.

A SQL query that includes a precision parameter in the CURRENT_TIME function could return inaccurate results. (14473)
The issue has been resolved.

A SQL query that compares the result of CURRENT TIME or CURRENT TIMESTAMP with a TIME literal may return inaccurate results. (14575)
The issue has been resolved.

Executing nuoloader with the --export option without specifying a CSV file name could result in unexpected behavior. (14695)
The issue has been resolved.

Large Object (LOB) data was processed using the hardcoded chunk size of 1024 bytes and the default chunk size of 65536 bytes was ignored. (13739)
The issue has been resolved.

Building a very large index using MERGE TREE, could cause the Transaction Engine (TE) to shut down unexpectedly. (14699)
The issue has been resolved.

A SQL query that includes a parameter that was cast to a specific data type could return inaccurate results because the data type length was ignored after casting.
The issue has been resolved.

A SQL query that includes a GROUP BY clause with only constant values and an ORDER BY clause could cause the TE to shut down unexpectedly.
The issue has been resolved.

A SQL query that uses decorrelated dependent existential hash joins could return inaccurate results when NULL values are included in the existential comparison.
The issue has been resolved.

A SQL query could report inaccurate results when using the hash variants of right outer, full outer, or right existential join operators on columns that contain NULL values.
The issue has been resolved.

Executing any SELECT query from GETSTATSMETADATA () on a pseudo table could report a precondition error instead of raising an exception.
The issue has been resolved.

Executing a SQL query that performs hash join operations between string expressions of different data types could report inaccurate results.
The issue has been resolved.

Executing a SQL query on a temporary table created from an existing partitioned table using CREATE TABLE LIKE could cause the database to shut down unexpectedly.
The issue has been resolved.

Executing an ANALYZE SQL statement containing projections could report inaccurate statistics because only the accumulated number of rows of multiple executions were reported (without providing the average over all executions).
The issue has been resolved.

Executing a CREATE TABLE SQL statement with column names beginning with HELP could result in a syntax error.
The issue has been resolved.

When using the nuosql command line tool, recalling previous commands using ! commands could report an error.
The issue has been resolved.

If the parameters accessed by a pseudo table are modified during query execution, the TE could report an error.
The issue has been resolved.

When using C, C++, or Java drivers, the IS_GENERATEDCOLUMN field in the DatabaseMetaData object returned by getMetaData() could contain the value NO for columns defined as GENERATED BY DEFAULT.
The issue has been resolved.
Now the IS_GENERATEDCOLUMN field correctly returns a YES for columns defined as GENERATED BY DEFAULT.

A transaction executing a DDL operation could automatically roll back if an exception occurred due to an update conflict with another transaction.
The issue has been resolved.

In rare cases, a SQL query could return inaccurate results due to a race condition between garbage collection and index scan.
The issue has been resolved.

In rare cases, a SQL query could report "result set unusable (corrupted record must be removed)" even though no corruption occurred.
The issue has been resolved.

In rare cases, MERGE TREE index statistics were not updated after removing all the data from a table.
The issue has been resolved.

In rare cases, an object fetch operation would not be fully acknowledged by another node, causing database operations involving that object to wait indefinitely.
The issue has been resolved.

In rare cases, a Transaction Engine (TE) would shut down unexpectedly when executing a SQL query that included a UNION operator. (14666)
The issue has been resolved.

When the stats-v2-collection database option was disabled, automatic v2 statistics updates were not completely disabled causing inaccuracies in cardinality estimates and resulting in slow query performance. (14668, 14748)
The issue has been resolved.

A SQL query that includes predicates in a correlated subquery referencing literals from another subquery could report "Error 58000: Internal error: failed precondition hasValidUnfilteredCardinality(relationId)". (14684)
The issue has been resolved.

Index statistics generation was not triggered when creating or rebuilding an index. This affected offline builds of tables with records that span multiple atoms.
The issue has been resolved.

The Table Statistics (Stats v3) metadata was not updated automatically for system tables.
The issue has been resolved.

When using the v3 optimizer, a SQL query could take longer to execute because the optimizer might not select the most efficient index while executing the query.
The issue has been resolved.

If a SQL query processes a few extremely large strings (containing millions of characters) in a column that mostly contains smaller strings, the memory usage during query execution could be significantly higher than when working only with smaller strings in the column." (14508, 14744)
This issue has been resolved.

A SQL query that executes the RAND() function a large number of times may generate more duplicate values than expected. (14512)
The issue has been resolved.

If LOB streaming is enabled by configuring the lobChunkingEnabled connection property to true in the NuoDB JDBC driver, a SQL query that fetches multiple BLOBs or CLOBs could cause a Transaction Engine (TE) to shut down with "Unable to free memory for 300 seconds".
The issue has been resolved.

Using nuodump with the output-schema option failed to replace schema references within foreign key constraints with the specified output schema.
The issue has been resolved.

Executing the EXPLAIN query on a SQL statement that includes columns with case insensitive collation could return inaccurate results.
The issue has been resolved.

If a table was converted to a partitioned table, running a SQL query against the table could cause the executing TE to shut down unexpectedly.
The issue has been resolved.

A SQL query that included an ATAN2() function could return inaccurate results. (14538)
The issue has been resolved.

A newly started Storage Manager (SM) with an archive of type LSA that attempted to synchronize with an existing SM while running a write-heavy workload could shut down unexpectedly with, "Archive failure: 'write' failed with error No space left on device".
The issue has been resolved.

In rare cases, if an SM shuts down soon after the database made changes to an atom that was not previously in memory, the copy of the atom in the SM’s archive could become corrupted.
The issue has been resolved.

Write-heavy workloads (workloads that include DML statements) could take longer to execute due to the logging of non-cached user queries.
The issue has been resolved.

The log files could be named incorrectly when using nuocmd get log-messages or nuocmd show log-messages with the --engine-file option.
The issue has been resolved.

A synchronizing SM could shut down unexpectedly with "FAILED: BUGCHECK: Type of object xx/yy is unknown" because an object was deleted.
The issue has been resolved.

If index statistics were requested during the shutdown of a TE or an SM, during the execution of the DROP INDEX statement, or when rebuilding indexes for a table, a SQL ANALYZE TABLE statement could run indefinitely. This could also prevent the automatic update of statistics.
This issue has been resolved.

Executing a SQL query with more than 50 JOIN operations could incorrectly report, "java.sql.SQLException: Memory exhausted". (14405)
The issue has been resolved.

In rare cases, during indexing, the Storage Managers (SMs) and Transaction Engines (TEs) could shut down unexpectedly reporting, "FAILED: !index.isArchive || NODE_STATE_SYNCING == index.thisNode→nodeState". (14537)
The issue has been resolved.

SQL queries with INNER JOINS and OUTER JOINS could take longer to execute. (14468, 14435)
The issue has been resolved.

Inserting a row into a very large table containing hundreds of millions of rows, with a large number of record atoms with free space, could cause the TE to shut down unexpectedly with, "listener thread taking longer than XX milliseconds to apply YY message". (14439, 14454, 14458, 14463, 14453, 14514)
The issue has been resolved.

SQL insert operations into a table with a non-unique Merge Tree index could be blocked indefinitely. (14145, 14307, 14314)
The issue has been resolved.

Executing the EXPLAIN (ANALYZE_GRAPH on) query on a SQL statement that included a LEFT OUTER JOIN, RIGHT OUTER JOIN, or FULL OUTER JOIN could report, "Error 58000: op type N4Exec18RightHashOuterJoinE is not supported by execution tree walker". (14423)
The issue has been resolved.

A SQL query that included a parameterized limit could take longer to execute. (14320)
The issue has been resolved.

The execution of nuodump returned SQL statements with improper formatting. (14155)
The issue has been resolved.

When pip install 'pynuoadmin[completion]' was used to install the pynuoadmin package, a nuocmd command could report "Cannot locate NuoDB: please set the NUODB_HOME environment variable." (14287)
The issue has been resolved.

A SQL SELECT query with a LEFT OUTER JOIN could take longer to execute when using the Merge Tree indexing compared to the B Tree indexing. (14265)
The issue has been resolved.

When upgrading to NuoDB 7.0 from any prior version, if there is only one database in the domain that has never been stopped or restarted since its creation, the first SM or TE started after the upgrade will not start and report, "Connection error: Joining node requests sid:x which conflicts with a prior node ID". (14561)
The issue has been resolved.

Execution of a JOIN operation on byte type attributes of different lengths could report, "Unsupported cast from bytes(n) to bytes."
The issue has been resolved.

If a query includes null predicates (IS [NOT] NULL), an OUTER join could incorrectly be written as an INNER join, causing inaccurate results.
The issue has been resolved.

A SQL comparison operation between two numeric values using < or >, where the first value has more decimal places than the second, could return incorrect results.
The issue has been resolved.

Executing a SQL query with a correlated subquery that references columns from the outer query in the SELECT list could report, "Error 58000: Internal error: failed precondition resolution."
The issue has been resolved.

Executing a SQL query containing an outer query that groups the results of a subquery returning multiple rows of NULL values could report, "Error 58000: unknown function type: 39`".
The issue has been resolved.

Executing a SQL query containing an existential subquery that queries a table function could report, "failed precondition projection().selectionVectorId() == _producers[_projectingProducer]→projection().selectionVectorId()".
The issue has been resolved.

Executing a SQL query that references a table that does not exist in the schema could result in the log message, "table 'XXX' does not exist" appearing multiple times.
The issue has been resolved.

Executing a SQL query on a partitioned table using Merge Tree indexes could report inaccurate results.
The issue has been resolved.

Execution of a SQL query that contains a GROUP BY operation on numerics of different scales could return inaccurate results.
The issue has been resolved.

Executing SQL queries that include sorting binary values in descending order, specifically zero values, with different byte lengths could return inaccurate results.
The issue has been resolved.

When executing ALTER DATABASE RUN GARBAGE COLLECTION, the command did not wait for the garbage collection process to complete.
The issue has been resolved.

Case sensitivity comparisons between CLOB (and TEXT) data types were not performed, regardless of their collation settings.
The issue has been resolved.

Execution of the EXPLAIN SQL command could return results containing extra empty spaces.
The issue has been resolved.

SQL queries involving inverted Merge Tree indexes with a large number of index components stored at higher levels than lower, could cause a TE and a SM to shut down with an Out-of-Memory (OOM) error.
The issue has been resolved.

When a SELECT FOR UPDATE SKIP LOCKED statement was executed on a locked table, the statement could wait for an extended period until the lock was released.
The issue has been resolved.

The peak amount of memory allocated during the execution of the statement reported by the PEAKMEMUSAGE field in the LASTSTATEMENT system table could be inaccurate.
The issue has been resolved.

When executing a query using nuosql from stdin, if the query contains a multi-line string, line breaks will be ignored.
The issue has been resolved.

When using nuosql, executing a query that includes a backslash literal (\) within a quoted string could report, "unrecognized character error".
The issue has been resolved.

When using nuodump, extracting DML statements that include quotes nested inside a quoted string could return inaccurate results.
The issue has been resolved.

When using nuodump, primary keys with user-specified names were not logged accurately.
The issue has been resolved.

Execution of the nuoarchive check --rewrite command could rewrite older atoms to the default version of nuoarchive rather than the effective version.
The issue has been resolved.

In rare instances, when an SM was shut down immediately after starting, the database could not be restarted and the SM remained in the UNKNOWN state indefinitely.
The issue has been resolved.

In rare cases, a CREATE TABLE statement could block another CREATE TABLE statement.
The issue has been resolved.

In rare cases, a SQL SELECT query that includes an ORDER BY clause, OUTER joins, and NULL in the sort column could report inaccurate results.
The issue has been resolved.

In a Kubernetes deployment, starting a process before Kubernetes populated the running pod state could result in "KubernetesResourceInspector Unable to find authoritative TE\SM process for dbName=xxx".
This issue has been resolved.

In Kubernetes deployments with multiple databases, scaling down to zero SM replicas could lead to inaccurate counting of database archives. This could result in logging error messages when attempting to remove the final archive for a database.
The issue has been resolved.

When using Java drivers, the generated keys from a statement or execution could be requested even after they had been invalidated.
The issue has been resolved.

When a server was configured to use IPv6 instead of IPv4 and requireLocalDomainUser was set to true in nuoadmin.conf, the unrestricted domain user specified by the nuoadmin.conf property domainUser did not have connect access despite being on the same host as the Admin Process (AP).
The issue has been resolved.

If an SM started on a new archive was shut down before initializing the archive, restarting the SM may report, "Location xx is not a valid file archive".
The issue has been resolved.

In rare cases, after an archive syncing process, a specific sequence of events could cause slow query performance due to long queues for atom reloads on a SM. (13083, 13700, 13757, 13950, 13993, 14131, 14190, 14299, 14303, 14334, 14342, 14347, 14385, 14390, 14394, 14444)
The issue has been resolved.

If the database protocol was not upgraded to at least version 6.0, migration of data from File archives to Log-Structured Archives (LSA) could cause the SMs to shut down unexpectedly. (14254, 14368)
The issue has been resolved.

Logging the execution plan of a running SQL query could cause the TE that executed the query to shut down unexpectedly if the client process was terminated abruptly. (14381, 14403)
The issue has been resolved.

When the ALTER TABLE REBUILD INDEXES USING MERGE TREE SQL command was used, the SM could shut down unexpectedly with "ASSERT: FAILED:thisObj.isComplete() || thisObj.isPurged()". (14296, 14297, 14298)
The issue has been resolved.

In rare cases, a SQL query with an ORDER BY clause on an indexed column may not execute. (14159)
The issue has been resolved.

In a Kubernetes deployment, in rare cases, especially if rapidly deleting and recreating a database with the same name, a database would not start because the NuoDB Kubernetes Aware Admin (KAA) became out of sync with the actual Kubernetes state due to rapid removal and recreation of archive objects.
The issue has been resolved.

In rare cases, a SQL request to lock data could run indefinitely because the chairman could not correctly ignore stale pending lock versions.
The issue has been resolved.

A SQL query that included a JOIN operation performed on columns of different data types could report, " failed: Internal error: failed precondition vectorDesc.dataTypeDesc(). ".
The issue has been resolved.

In some cases, a SQL query that included window functions in JOIN conditions could not execute and reported an error.
The issue has been resolved.