Skip to main content
Version: 3.18

ScalarDB 3.18 Release Notes

This page includes a list of release notes for ScalarDB 3.18.

v3.18.4​

Release date: November 11, 2026

Summary​

This release improves transaction metadata decoupling by supporting READ COMMITTED isolation on more databases and fixing issues with imported tables. It also addresses several data correctness issues, particularly those involving BLOB values in ScalarDB Cluster and ScalarDB SQL, and improves SQL parsing and validation. Additionally, this release enhances schema management, fixes various storage adapter issues, and strengthens security through dependency upgrades and vulnerability fixes.

Community edition​

Improvements​

  • Lowered the isolation level that ScalarDB requires for consistent virtual table reads to READ COMMITTED for Oracle, PostgreSQL, AlloyDB, YugabyteDB, MySQL, MariaDB, and TiDB. With this change, you can use transaction metadata decoupling on these databases at READ COMMITTED, which is the default isolation level in Oracle and PostgreSQL, and you no longer need SERIALIZABLE on Oracle. (#4148)
  • Made the schema file optional for --repair-all in Schema Loader when --coordinator or --replication-tables is specified, so that you can repair the coordinator tables and replication tables without re-applying table schemas. In addition, running --repair-all without a schema file, --coordinator, or --replication-tables now does nothing and exits successfully instead of failing with an error. (#4183)

Bug fixes​

  • Fixed an issue where a conditional put (PutIfExists or PutIf) that specifies only primary key values failed with an SQL error in JDBC databases. In Spanner, such a put still fails for tables that have only primary key columns. (#4129)
  • Fixed issues with tables imported with transaction metadata decoupling in JDBC databases. A record that no transaction had written since the import could no longer be updated or deleted after a transaction that wrote it was rolled back. Also, a transaction could recreate a record that another transaction deleted after the first transaction read it. In Spanner, the latter still occurs for tables that have only primary key columns. Records that already can't be updated or deleted because of this issue stay that way after you upgrade. To recover them, after upgrading all your applications, delete the records whose tx_id column is NULL from the <table_name>_tx_metadata table of the imported table. For example: DELETE FROM <namespace>.<table_name>_tx_metadata WHERE tx_id IS NULL. (#4131)
  • Fixed an issue where deleting a record that no transaction had written since the import failed with an IllegalArgumentException when one-phase commit was enabled, in tables imported with transaction metadata decoupling in JDBC databases. (#4133)
  • Fixed CVE-2026-91403 and CVE-2026-114820 (eMudhra CA compromise) in the ScalarDB Schema Loader and ScalarDB Data Loader CLI container images by updating the base image to Eclipse Temurin 21.0.13_9, which no longer trusts the compromised emSign root CAs. (#4143)
  • Upgraded the Guava library to fix a security issue: CVE-2026-102554 (#4144)
  • Applications that use ScalarDB as a library now get the fixed Netty and Jackson versions, unless they manage those versions themselves. Previously, fixes for vulnerabilities in these libraries (for example, CVE-2026-75595 in Netty and CVE-2026-89407, CVE-2026-89425, CVE-2026-68497, CVE-2026-91776, and CVE-2026-91777 in Jackson) applied only to ScalarDB's own builds and container images. As a result, Jackson in those applications is upgraded to 2.22. Depending on the ScalarDB version, applications built with Maven previously got much older versions, such as jackson-databind 2.7 and Netty 4.1.94. This release also fixes CVE-2026-73508 in Netty, which affected the Schema Loader and Data Loader CLI images as well as those applications. (#4146)
  • Fixed an issue in the Object Storage adapter with Amazon S3 where truncating or dropping multiple tables concurrently could hang indefinitely. (#4149)
  • Fixed an issue where importing a table from Oracle, MySQL, MariaDB, or TiDB registered the primary key columns in the alphabetical order of the column names instead of the order of the table's primary key. For tables imported after this fix, specify keys, including --partition-key for Data Loader exports, in the order of the table's primary key. Tables imported before this fix keep their existing order. (#4175)
  • Upgraded Micrometer (micrometer-core, micrometer-observation, and micrometer-commons) to 1.15.12 to fix CVE-2026-40983 and CVE-2026-40984. ScalarDB now declares them directly and excludes them from the libraries that bring them in, so applications that use ScalarDB as a library also get the fixed version, unless they manage the Micrometer version themselves. (#4192)

Enterprise edition​

Bug fixes​

ScalarDB Cluster​
  • Fixed an issue where the embedding client threw AssertionError instead of UnsupportedOperationException for an unsupported filter type, such as ContainsString.
  • Fixed CVE-2026-91403 and CVE-2026-114820 (eMudhra CA compromise) in all ScalarDB Cluster container images by updating the base images to Eclipse Temurin 21.0.13_9 and, for the UBI-based node image, java-21-openjdk 21.0.13.0.9, which no longer trust the compromised emSign root CAs.
  • Upgraded the grpc_health_probe binary to fix a security issue: CVE-2026-116587
  • Upgraded the Guava library to fix a security issue: CVE-2026-102554
  • Fixed a bug where the ScalarDB Cluster client read a BLOB value bound as a ByteBuffer to the end of the buffer each time it sent the value, so the value was sent as an empty BLOB from the second time on, such as when a statement was executed again or retried, when the same buffer was bound to more than one bind marker, or when statements in a batch shared the same value. As a result, an empty BLOB could be written, or BLOB columns could be compared with an empty BLOB, without any error. The client now copies the bytes without changing the buffer's position. This fix does not repair data that was already affected.
  • Fixed a bug where, with write buffering enabled, the ScalarDB Cluster client sent buffered INSERT and UPSERT statements with the values they had when the buffered statements were sent, instead of the values they had when they were executed. If an application reused a BoundStatement, or a byte array or ByteBuffer for a BLOB value, after executing a statement, every buffered statement that used it was sent with the last values, so rows could be written with wrong values without any error. The client now copies the values of a statement when it buffers the statement. This fix does not repair data that was already affected.
  • Upgraded Bouncy Castle (bcprov-jdk18on, bcpkix-jdk18on, and bcutil-jdk18on) to 1.86 in the Java client SDK and the embedding Java client SDK to fix CVE-2025-14813, CVE-2026-0636, CVE-2026-5588, CVE-2026-5598, CVE-2026-8763, and CVE-2026-13506. Applications that use these SDKs now get the fixed version, unless they manage the Bouncy Castle version themselves. Previously, the fixed version applied only to ScalarDB Cluster's own builds and container images other than the BYOL ones, which loaded Bouncy Castle 1.70 classes that a dependency brought in. The BYOL container images now use the fixed version, too.
  • Applications that use the Java client SDK or the embedding Java client SDK and build with Maven now get the same version for all Jackson artifacts. Previously, some of them, such as jackson-dataformat-cbor, could be at older versions that other dependencies of the SDKs request. The embedding Java client SDK no longer exposes Jackson as a compile-time dependency. If your application uses Jackson directly and relied on the SDK to provide it, declare Jackson in your application's build.
ScalarDB GraphQL​
ScalarDB SQL​
  • Upgraded the Guava library to fix a security issue: CVE-2026-102554
  • Fixed Value.hashCode() for BLOB values created from byte[] so that it is consistent with equals(). This inconsistency could cause a SELECT, UPDATE, or DELETE statement that repeats the same BLOB equality condition on a primary-key or indexed column in every OR group to be planned with a less efficient scan, or as a cross-partition scan that fails unless cross-partition scan filtering is enabled.
  • Fixed two bugs that could silently write or compare wrong BLOB values. In direct mode, a BLOB value bound as a ByteBuffer through BoundStatement.setBlob(), Value.ofBlob(), Value.of(), or the JDBC setObject() method was read to the end of the buffer each time it was used, so an UPDATE statement that updated multiple rows, or a statement executed again, wrote an empty BLOB or compared BLOB columns with an empty BLOB. In the JDBC driver, setBytes() and setObject() kept a reference to the application's array or buffer, so reusing one array or buffer for each row of a batch wrote the last contents to every row. ScalarDB SQL now copies these values instead of reading the application's buffer or array when it uses them. This fix does not repair data that was already affected.
  • Fixed a bug where the SQL parser misread some SQL. It dropped characters that it could not read, so INSERT INTO ns.té was executed on the table ns.t, and names quoted with backticks or square brackets were accepted. It also read a number followed by a word without a space, such as 5AND, as a number and a word, and ignored a block comment that was not terminated together with the rest of the statement. In ScalarDB Cluster, such SQL could make a statement outside a transaction run with the credentials in the client configuration instead of those in its WITH clause. Such SQL is now rejected with a syntax error, and -- now starts a line comment whatever follows it, as in the SQL standard. The comment /*!*/ no longer makes the parser skip the SQL up to the next */, which could make a DELETE statement delete all the rows.
  • Fixed an issue where a HAVING clause in a SELECT statement with neither a GROUP BY clause nor an aggregate function was silently ignored, so the statement returned every row that matched the WHERE clause. As in the SQL standard, a HAVING clause alone makes all the rows one group, so such a statement is now rejected with the same error as other statements that select a column that is neither in the GROUP BY clause nor in an aggregate function. A statement that returned rows before can therefore fail now; to filter rows, move the condition to the WHERE clause. Statements with a GROUP BY clause or an aggregate function are not affected. The error message for such a column now shows the column as written instead of internal details.

v3.18.3​

Release date: October 30, 2026

Summary​

This release adds support for SQL Server 2025 and drops support for SQL Server 2017. It also improves transaction reliability and error handling in Consensus Commit, optimizes table metadata operations, enhances the correctness and reliability of storage adapter operations, and improves resource management. Additionally, this release strengthens security through authentication and authorization fixes, dependency upgrades, and vulnerability mitigations, while addressing data comparison and aggregation issues in ScalarDB SQL.

Community edition​

Improvements​

  • Fixed an issue where a conflict error, such as a lock wait timeout or a serialization failure, in a single put or delete operation on JDBC databases was not classified as retriable. As a result, Consensus Commit can now report such a conflict during commit as CommitConflictException instead of CommitException or UnknownTransactionStatusException. (#3852)
  • Fixed the Consensus Commit record recovery to retry conflict errors reported by the underlying storage, and to report a conflict that persists as a conflict so that the transaction can be retried. (#3853)
  • Improved the performance of checking whether a table exists. tableExists() now reads only the metadata of the specified table instead of listing the tables in the namespace, so its cost no longer grows with the number of tables. As a result, on DynamoDB, a table that has no ScalarDB metadata is no longer reported as existing. Also fixed an issue in which getTableMetadata() for DynamoDB and Cosmos DB threw an exception instead of returning null when the ScalarDB metadata table or container did not exist. On these storages, a "table not found" error can therefore now also mean that the metadata table or container itself does not exist. If such a table is visible in the AWS console or the Azure portal, verify that scalar.db.system_namespace_name (and, on DynamoDB, the namespace prefix) match across all configurations before running repairTable(). (#3947)
  • Added support for SQL Server 2025 and dropped support for SQL Server 2017. (#3998)
  • Changed table import to reject column types that can't follow the TEXT collation order: SQL Server nchar, nvarchar, text, and ntext; Oracle NCHAR and NVARCHAR2; and Db2 GRAPHIC, VARGRAPHIC, and DBCLOB. (#4051)
  • Upgraded the SQL Server JDBC driver to 13.6.0 and fixed TIMESTAMPTZ values dated before October 15, 1582, being stored up to 10 days off in SQL Server. TIMESTAMPTZ values dated before October 15, 1582, that were stored in SQL Server by a previous ScalarDB version are read back up to 10 days earlier than the value that was written. If you have such data, consider migrating it to remove the offset. (#3971)
  • Improved the S3 and Cloud Storage adapters to retry deleting objects that failed on transient errors when truncating or dropping tables. (#4077)
  • Mitigate CVE-2026-91403 and CVE-2026-114820 (eMudhra CA compromise) by removing the affected emSign root CAs from the cacerts of all container images, and update the base image of the ScalarDB Schema Loader and ScalarDB Data Loader CLI container images from Java 8 (Eclipse Temurin 8) to Java 21 (Eclipse Temurin 21). The images set no JVM options by default, so this affects only users who pass Java 8-specific JVM options (for example, -XX:+PrintGCDateStamps through JAVA_TOOL_OPTIONS), which now make the JVM fail to start. (#4112)

Bug fixes​

  • Fixed the get operation in the Cloud Storage adapter to handle concurrent writes. (#3851)
  • Fixed an issue where rolling back a Consensus Commit transaction with an open scanner, including the automatic rollback of an expired or evicted transaction, could fail with an IllegalArgumentException and leave the remaining scanners open and the group commit slot unreleased. (#3867)
  • Fixed the Cassandra adapter so that an update applied by a conditional write is no longer silently discarded when it runs in the same millisecond as a preceding non-conditional write to the same record. The adapter now takes write timestamps from the server and sends every request for a partition to the same replica. (#3890)
  • Fixed an issue where Consensus Commit accepted invalid conditions on conditional mutations instead of rejecting them. (#3871)
  • Fixed an issue in DynamoDB where a single conditional write that the AWS SDK retried could be reported as not applied, even though the write had been applied. Such a write is now reported as having an unknown outcome. Also fixed an issue where some failures in DynamoDB operations were thrown as unchecked AWS SDK exceptions instead of ExecutionException. (#3922)
  • Fixed an issue where a scan condition with the !=, <, or <= operator was satisfied by a NULL column value in Cassandra, Cosmos DB, DynamoDB, and Object Storage. Also fixed an issue where a != condition was satisfied by a NULL column value in PutIf or DeleteIf in Cassandra, DynamoDB, and Consensus Commit, and in UpdateIf in Consensus Commit. (#3930)
  • Fixed an issue where repairTable() on DynamoDB did not enable auto-scaling for existing global secondary indexes, for example after createTable() or createIndex() failed before enabling auto-scaling (such as when Application Auto Scaling permissions were missing). repairTable() now also keeps the existing auto-scaling range of a table instead of resetting it to the given or default request unit, so running repair without --ru no longer lowers the capacity of existing tables, including the Coordinator table. To change the auto-scaling range of an existing table, use the AWS console or CLI. If createTable() or createIndex() with auto-scaling enabled failed before, run repairTable() again after upgrading, with the schema that includes the index and the same request unit as at creation. If repairTable() (or Schema Loader with --repair-all) was run before without --ru or with a smaller request unit, check the auto-scaling range of existing tables, including the Coordinator table, and restore it with the AWS console or CLI, because repairTable() no longer changes the range of existing tables. (#3949)
  • Fixed an issue in the Object Storage adapter with Amazon S3 where a conditional write that the AWS SDK retried could be reported as a conflict, even though the write had been applied. Such a write is now reported as having an unknown outcome. (#3958)
  • Fixed an issue in the Object Storage adapter with Azure Blob Storage where a conditional write that the Azure SDK retried could be reported as a conflict, even though the write had been applied. Such a write is now reported as having an unknown outcome. (#3964)
  • Fixed an issue in the Object Storage adapter with Google Cloud Storage where a deletion that the Cloud Storage client retried could be reported as a conflict, even though the deletion had been applied. Such a deletion is now sent only once, and a failure of it other than an unsatisfied condition is reported as having an unknown outcome. (#3965)
  • Fixed an issue where a scanner opened outside a transaction was never closed if a client abandoned it while active transaction management kept it across requests. Such a scanner kept its transaction and storage resources, for example, a database connection and an open database transaction. Now, such a scanner is closed when it has not been used for the active transaction expiration time, or when more such scanners than the maximum number of active transactions are open. (#3992)
  • Upgraded the Netty and Jackson libraries to fix security issues: CVE-2026-75595, CVE-2026-89407, CVE-2026-89425, CVE-2026-68497, CVE-2026-91776, and CVE-2026-91777 (#3996)
  • Fixed an issue in the Object Storage adapter with Amazon S3 and Google Cloud Storage where truncating or dropping a table could succeed while leaving some of the table's objects undeleted. Such a failure is now reported as an error. (#4043)
  • Fixed an issue where a Put operation on a Spanner table named excluded failed. (#4044)
  • Fixed an issue where the scalar.db.s3.multipart_upload_max_concurrency option limited the number of concurrent requests of all S3 operations, not only multipart uploads, in the S3 adapter. (#4061)

Enterprise edition​

Improvements​

ScalarDB Cluster​
  • Fixed OIDC JWT validation so that an exhausted JWKS retrieval rate limit is reported as an authentication failure instead of an internal error.

Bug fixes​

ScalarDB Cluster​
  • Fixed an issue where a user whose authentication methods did not include USERPASS could log in with username/password authentication. If such a user needs username/password authentication, add USERPASS to the user's authentication methods and, unless passwordless login is intended, set a password. The fix takes full effect after all nodes in the cluster have been upgraded.
  • Fixed an issue where, with authentication enabled, a client could keep reading from an open scanner after its auth token had expired or been revoked, or after its READ privilege had been revoked. Fetching from a scanner now re-checks the READ privilege. For OIDC clients, a JWT that expires during a scan now causes an authentication error at the next fetch, as with get and scan; a client that supplies the JWT through OidcJwtAccessTokenHolder can set a fresh JWT and continue the scan.
  • Fixed an issue where batch Get and Scan operations on tables with encrypted columns failed with IllegalStateException when associated_data_required was enabled and their projections did not include all primary key columns.
  • Upgraded the Jackson, Netty, and Bouncy Castle libraries to fix security issues: CVE-2026-8763, CVE-2026-13506, CVE-2026-68497, CVE-2026-75595, CVE-2026-89407, CVE-2026-89425, CVE-2026-91776, and CVE-2026-91777
  • Upgraded the grpc_health_probe binary to fix security issues: CVE-2026-56851, CVE-2026-78667, CVE-2026-78669, CVE-2026-84304, CVE-2026-84445, and CVE-2026-97031
  • Fixed an issue where closing a ScalarDB Cluster client in the direct-kubernetes client mode left behind a thread and a connection to the Kubernetes API server. An application that repeatedly created and closed clients, including through the ScalarDB SQL JDBC driver, could eventually fail to create new threads.
  • Mitigate CVE-2026-91403 and CVE-2026-114820 (eMudhra CA compromise) by removing the affected emSign root CAs from the cacerts of all container images
ScalarDB SQL​
  • Fixed BLOB comparisons that ScalarDB SQL evaluates itself, in ORDER BY and in range conditions in HAVING and in WHERE after a join, treating bytes from 0x80 up as smaller than 0x00 to 0x7F.
  • Fixed MIN and MAX on FLOAT and DOUBLE columns returning a result that depended on row order when the values included NaN, or both 0.0 and -0.0.

v3.18.2​

Release date: September 2, 2026

Summary​

This release adds support for the AWS Advanced JDBC Wrapper for fast Aurora failover, improves Cosmos DB error handling and ScalarDB Cluster's attribute-based access control metadata cache refresh, adds retry handling for conflict errors during schema operations, and upgrades the AWS SDK and grpc_health_probe to address security vulnerabilities.

Community edition​

Improvements​

  • Added support for the AWS Advanced JDBC Wrapper, which enables fast Aurora failover for Aurora PostgreSQL and Aurora MySQL. Configuring HikariCP with an exception override is now rejected at startup, since it would cause a transaction with an unknown outcome to be reported as a definite failure. (#3805)
  • Fixed an issue where the error message for a Cosmos DB error was excessively large because it included the Cosmos DB diagnostics, which could cause the error information to be lost. (#3816)

Bug fixes​

  • Fixed an issue where schema operations could fail with a conflict error, such as ORA-08177 on Oracle when using the SERIALIZABLE isolation level. Conflict errors are now retried. (#3820)
  • Upgraded the AWS SDK to fix security issues: CVE-2026-54399 and CVE-2026-54428 (#3838)

Enterprise edition​

Improvements​

ScalarDB Cluster​
  • Reduced the number of queries issued when the attribute-based access control metadata caches are refreshed, which removes a latency spike on the first request after each refresh.

Bug fixes​

ScalarDB Cluster​

v3.18.1​

Release date: August 2, 2026

Summary​

This release improves repairTable to avoid unnecessary metadata and indexing-policy updates, adds support for jdbc:spanner connection strings, fixes multiple Consensus Commit transaction handling issues, improves ScalarDB Cluster metadata repair and ScalarDB SQL SELECT execution, resolves Cluster rollback/pause and SQL statement cache bugs, and upgrades several dependencies to address security vulnerabilities.

Community edition​

Improvements​

  • Improved repairTable to skip rewriting a table's metadata when it is already up to date (and, for Cosmos DB, to skip updating the container indexing policy when it already matches), avoiding unnecessary writes. (#3613)

Bug fixes​

  • Accepts connection string starting with jdbc:spanner to connect to Spanner (emulator, omni, and cloud instance) in addition to the already supported jdbc:cloudspanner pattern. (#3559)
  • Fixed an issue in Consensus Commit where a Get operation using a secondary index could fail with an IllegalArgumentException when another record with the same indexed value was concurrently being deleted and inserted. (#3607)
  • Fixed an issue in Consensus Commit where a read-only transaction could fail when coordinator write omission on read-only was disabled and coordinator group commit was enabled. (#3614)
  • Fixed a bug where aborting an in-flight, group-committed transaction by ID via DistributedTransactionManager.rollback(String) / abort(String) could be lost when the Coordinator group commit feature was enabled. (#3619)
  • Fixed an issue in the Consensus Commit transaction manager where a read could return a stale (pre-commit) value for a record whose writing transaction committed concurrently during lazy recovery. (#3621)
  • Upgraded the Jackson, Netty, Azure Cosmos DB, and Azure Blob Storage libraries to fix security issues: CVE-2026-42579, CVE-2026-42583, CVE-2026-42584, CVE-2026-42587, CVE-2026-44249, CVE-2026-45416, CVE-2026-45674, CVE-2026-47691, CVE-2026-50010, CVE-2026-54512, CVE-2026-54513, CVE-2026-55831, CVE-2026-55833, CVE-2026-56745, CVE-2026-59901, and GHSA-r7wm-3cxj-wff9 (#3754)
  • Upgraded the PostgreSQL JDBC driver to fix security issues: CVE-2026-42198 and CVE-2026-54291 (#3758)

Enterprise edition​

Enhancements​

ScalarDB SQL​
  • Added @AutoConfigureBefore to ScalarDbJdbcConfiguration to explicitly declare auto-configuration ordering relative to Spring Boot's auto-configs.

Improvements​

ScalarDB Cluster​
  • ScalarDB Cluster now repairs its authentication and ABAC system metadata tables at node startup to create the companion before-image secondary indexes required after upgrading, so deployments upgraded from an earlier version gain them automatically without a manual repairTable().
ScalarDB SQL​
  • Enabled the read-only optimization for one-shot SELECT statements and SELECT-only batches.

Bug fixes​

ScalarDB Cluster​
ScalarDB SQL​
  • Fixed a statement cache regression where non-parameterized DML statements occupied cache slots intended for parameterized SQL. Cache eligibility is now correctly gated on the presence of bind markers in the parsed SQL.

v3.18.0​

Release date: May 1, 2026

Summary​

This release includes a lot of enhancements, improvements, security issue fixes, and bug fixes.

Backward-incompatible changes
  • JDBC connection pool migrated from Apache Commons DBCP2 to HikariCP:
    • The following configuration properties are no longer supported. They are ignored, with a warning logged, if set:
      • scalar.db.jdbc.connection_pool.max_idle
      • scalar.db.jdbc.table_metadata.connection_pool.max_idle
      • scalar.db.jdbc.admin.connection_pool.max_idle
      • scalar.db.jdbc.prepared_statements_pool.enabled
      • scalar.db.jdbc.prepared_statements_pool.max_open
    • Prepared statement pooling is no longer available because HikariCP has no equivalent feature.
    • Connection pool default behavior follows HikariCP rather than DBCP2, which may change runtime behavior even without configuration changes (for example, connection acquisition timeout and idle eviction).
  • Existing tables require repairTable() to create the new before-image secondary index. Index-based Get, Scan, and ScanAll in Consensus Commit now rely on a companion before_<col> index. Until you run repairTable() on each existing table, ScalarDB logs a warning at startup and index-based reads fall back to the previous behavior, which may miss records whose indexed column is being concurrently updated. The configuration scalar.db.consensus_commit.index.eventually_consistent_read.enabled is provided as an opt-out but is not recommended for new workloads.
  • MySQL Connector/J replaced with MariaDB Connector/J. SSL is no longer enabled by default. If you connect to MySQL 8.0 or later with caching_sha2_password authentication, add sslMode=REQUIRED or allowPublicKeyRetrieval=true to your JDBC URL. No action is required for jdbc:mysql:// URLs because ScalarDB automatically appends permitMysqlScheme for compatibility.
  • BigIntColumn.MAX_VALUE and BigIntColumn.MIN_VALUE constants have been removed. Code that referenced them must use Long.MAX_VALUE and Long.MIN_VALUE instead.

Community edition​

Enhancements​

  • Added AuthAdmin.getRole(roleName). (#3238)
  • Added operation-level attributes to allow cross-partition scan, filtering, and ordering on a per-operation basis without requiring global configuration. (#3425)
  • Added DistributedTransactionAdmin.hasPrivilege(). (#3432)
  • Added AuthenticationMethod enum and authentication-method-aware user management APIs to AuthAdmin to support OIDC authentication. (#3433)
  • Added attributes parameter to DistributedTransactionManager begin and start methods, enabling transaction-scoped attribute configuration. CRUD methods on DistributedTransactionManager also pass the operation's attributes to the internally begun transaction. Also added support for specifying the isolation level per transaction in Consensus Commit via the cc-transaction-isolation attribute. (#3466 #3517 #3540)
  • Added support for Google Cloud Spanner as a JDBC storage, via its PostgreSQL-compatible dialect. (#3510)

Improvements​

  • Extended the applicability of one-phase commit optimization in the Consensus Commit protocol. This allows one-phase commit to be used even in SERIALIZABLE isolation level when the transaction only reads records that it subsequently updates, improving performance for read-modify-write workloads. (#3295)
  • Added batch execution support for mutate operations in the JDBC adapter to improve performance when executing multiple mutations. Mutations with the same SQL statement are now grouped and executed as a batch. (#3304)
  • Migrated the JDBC adapter's connection pooling library from Apache Commons DBCP2 to HikariCP for improved performance and reliability. (#3305)
  • Changed the behavior of Consensus Commit transactions to allow Insert/Upsert/Update operations after Delete on the same record within the same transaction. Previously, this operation threw an IllegalArgumentException, but now it properly handles the case by treating it as a new record insertion with null values for unspecified columns. (#3319)
  • Added @CheckReturnValue annotation to Get/Scan builders. (#3320)
  • Added support for setting null values on secondary index columns in DynamoDB. When a null value is set, the attribute is removed from the item and the record will not appear in secondary index scans. (#3326)
  • Inserting or updating records with TIME (microsecond precision), TIMESTAMP (millisecond precision), and TIMESTAMPTZ (millisecond precision) column values will truncate out-of-range precision rather than throwing an exception. (#3393)
  • Fixed a correctness issue in index-based Get, Scan, and ScanAll operations in Consensus Commit where records whose indexed column was being concurrently updated by another transaction could be missed. ScalarDB now maintains a companion before-image secondary index for each user-defined secondary index on a non-primary-key column and uses it to recover PREPARED/DELETED records during index reads. (#3419 #3463)
  • Replaced MySQL Connector/J with MariaDB Connector/J to resolve GPLv2 licensing concerns. (#3428)
  • Shortened JDBC index names using a hash when they exceed the maximum identifier length supported by the underlying database. (#3481)
  • Relaxed the BigInt value range restriction (-2^53 to 2^53) that previously applied to all storage backends. This restriction now only applies to Cosmos DB and Object Storage. Other backends (JDBC, Cassandra, DynamoDB) now support the full Java long range for BigInt columns. (#3490)
  • Changed the Oracle BIGINT column type mapping from NUMBER(16) to NUMBER(19) to support the full Java long range. (#3507)
  • Changed the deprecation policy so that APIs and configurations marked as deprecated will now be removed in 4.0.0 instead of 5.0.0. (#3520)

Bug fixes​

  • Fixed option issues in Object Storage adapter. (#3237)
  • On Oracle, when importing a table with a column using the NUMBER(1) data type, which is usually used for BOOLEAN data, that column can now be mapped to ScalarDB BOOLEAN using ScalarDB Schema Loader override-columns-type setting. (#3239)
  • Fix to increase the maximum allowed string length with Object Storage. (#3248)
  • Updated the upper limit value displayed in the error message for data size limitation in Object Storage adapter. (#3264)
  • Added explicit commits for Oracle database when using SERIALIZABLE isolation level to ensure snapshot updates after each operation. (#3294)
  • Upgraded the Jackson library to fix a security issue: GHSA-72hv-8253-57qq (#3394)
  • Fix dropColumnFromTable() dropping the secondary index even when column drop is unsupported. (#3450)
  • Upgraded the Netty library to fix security issues: CVE-2026-33870 and CVE-2026-33871 (#3452)
  • Fixed a bug where index-based Get and Scan operations could return incorrect results after lazy recovery rolled back a PREPARED record whose after-image index value matched the query but whose before-image (restored) value did not. (#3488)

Enterprise edition​

Enhancements​

ScalarDB Cluster​
  • Added getRole() API and equivalent to retrieve a single role by name.
  • Added support for starting replication with existing backup site tables.
  • Added support for the executeBatch API in ScalarDB Cluster SQL transactions, enabling batch execution of multiple SQL statements in a single call.
  • Added client-side optimizations for SQL transactions (piggyback_begin and write_buffering) to reduce RPC overhead between the client and the cluster.
  • Added support for configuring a separate gRPC port for the admin service by using the scalar.db.cluster.node.admin.port property.
  • Added DistributedTransactionAdmin.hasPrivilege() to check if a specified user has a specific privilege on a table.
  • Added support for the attributes parameter in transaction begin/start methods, enabling transaction-scoped configuration such as isolation level.
  • Added OIDC JWT authentication support.
  • Added ThreadLocal-based user-password authentication support via CredentialsHolder, allowing multiple users to share a single TransactionManager.
  • Added property-based OIDC JWT authentication support for the client, allowing JWT access tokens to be passed via configuration properties.
ScalarDB SQL​
  • Added Metadata.getRole(roleName)
  • Added support for Spring Boot 4 in Spring Data JDBC for ScalarDB.
  • Added support for extending AbstractJdbcConfiguration for custom bean configuration in Spring Data JDBC for ScalarDB.
  • Changed the PreparedStatement API to use bind() methods that return a BoundStatement for parameter binding, instead of directly setting values on PreparedStatement.
  • Added executeBatch API to execute multiple statements in a single call, with support in the core SQL API, direct-mode, and JDBC driver.
  • Added support for BLOB literals using X'hex' syntax in SQL statements.
  • Added support for AuthenticationMethod in CREATE/ALTER USER and SHOW USERS
  • Added support for specifying transaction-scoped attributes in BEGIN and START TRANSACTION SQL statements using the WITH clause (e.g., BEGIN WITH 'cc-transaction-isolation' = 'SNAPSHOT'). Also added begin(Map), and beginReadOnly(Map) methods to SqlSession for programmatic attribute specification.
  • Updated SqlJdbcDatabaseMetaData.getUserName() to return the authenticated user name via Metadata.getCurrentUser().
  • Added support for specifying ABAC read and write tags in the WITH clause of BEGIN and START TRANSACTION statements.

Improvements​

ScalarDB Cluster​
  • Refactored internal auth token handling to use a type-safe class hierarchy instead of string prefix parsing.
  • Auth, ABAC, and Encryption modules no longer require scalar.db.cross_partition_scan.enabled=true to be enabled globally. These modules now use operation-level cross-partition scan attributes and run internal metadata transactions with Snapshot Isolation.
  • Changed the deprecation policy so that APIs marked as deprecated will now be removed in 4.0.0 instead of 5.0.0.
ScalarDB GraphQL​
  • Relaxed the range of values accepted by the GraphQL BigInt scalar to the full Java long range, aligning with the corresponding relaxation in ScalarDB. Backend-specific limits (Cosmos DB and Object Storage still restrict BigInt to -2^53 to 2^53) are now enforced by ScalarDB at the storage layer rather than by the GraphQL layer.
ScalarDB SQL​
  • Improve thread pool management in two-phase commit interface transaction to avoid potential starvation issues in Spring Data JDBC for ScalarDB.
  • Added one-operation mode support for one-shot query execution to improve performance by skipping explicit transaction begin/commit when the SQL query can be expressed as a single operation.
  • Inserting or updating records with TIME (microsecond precision), TIMESTAMP (millisecond precision), and TIMESTAMPTZ (millisecond precision) column values will truncate out-of-range precision rather than throwing an exception.
  • Fixed an issue where the shadow jar was unnecessarily published to GitHub Packages for the CLI module.
  • Update the TIMESTAMPTZ literal to make optional the space character before the UTC timezone Z character. For example, 2021-03-04 12:30:45.123Z is now accepted, in addition to the current format 2021-03-04 12:30:45.123 Z. Also, when selecting a TIMESTAMPTZ column, the value is printed without a space before the Z.
  • Refactored the SQL statement builder APIs for improved type safety and ergonomics. SelectStatementBuilder now enforces SQL clause ordering at the type level so that having() is only callable after groupBy(), and HAVING supports the same fluent .and(...) / .or(...) chain as WHERE; Having.of(...) factories have been consolidated into Having.create(...); CreateUserStatementBuilder / AlterUserStatementBuilder expose explicit withPassword() / withSuperuser() / withNoSuperuser() methods in place of the previous overloaded with(...); the UserOption enum has been removed in favor of a nullable superuser boolean on CreateUserStatement / AlterUserStatement; and FunctionRef now provides convenience factories (count(), sum(...), avg(...), min(...), max(...)) for common aggregates.

Bug fixes​

ScalarDB Cluster​
  • Made GRANT ROLE command idempotent, allowing duplicate grants and upgrading to WITH ADMIN OPTION when re-granting.
  • Fixed a bug where ScalarDB Cluster cannot be deployed in the Omnistrate environment by upgrading scalar-metering.
  • Fixed a bug where the pause functionality did not work correctly when transactions expired.
  • Fixed an issue where the batch operation with piggyback commit threw CrudException instead of CrudConflictException when a commit conflict occurred. This allows clients to properly detect and handle commit conflicts.
  • Fixed a bug where batch() with piggyback commit threw CrudException instead of UnknownTransactionStatusException on unexpected gRPC errors. This could cause incorrect error handling on the client side, as the transaction status was actually unknown when piggyback commit was enabled.
  • Upgraded grpc_health_probe to fix security issues: CVE-2025-68121, CVE-2025-61726, CVE-2025-61728, CVE-2025-61729, and CVE-2025-61730
  • Excluded com.microsoft.azure:adal4j from the Kubernetes Java Client to fix security issues: CVE-2023-52428, CVE-2021-31684, and CVE-2023-1370
  • Upgraded the Jackson library to fix a security issue: GHSA-72hv-8253-57qq
  • Upgraded grpc_health_probe to fix security issues: CVE-2025-59250 and CVE-2026-25679
  • Upgraded the Netty library to fix security issues: CVE-2026-33870 and CVE-2026-33871
  • Fixed a bug where one-shot batch operations bypassed pause control in the GateKept transaction managers.
  • Upgraded grpc_health_probe to fix a security issue: CVE-2026-34986.
ScalarDB SQL​
  • Ambiguous column names in ORDER BY and HAVING clauses were detected.
  • Fixed a ClassCastException that occurred in StatementUtils.appendTerm() when handling DATE, TIME, TIMESTAMP, and TIMESTAMPTZ values. These values are now correctly formatted as string literals instead of being incorrectly cast to String.
  • Fixed SelectStatement.toSql() to correctly generate ORDER BY clauses with aggregate functions such as SUM() and COUNT(). Previously, only column-based orderings were handled, causing incorrect SQL output when function-based orderings were used.
  • Fixed a bug where duplicate column names were allowed in CREATE TABLE statements.
  • Fixed missing strict date validation on time-related type formatters, which could allow invalid dates (e.g., February 30) to be silently accepted instead of rejected.
  • ResultSet.getObject on BLOB columns now returns a java.sql.Blob, and ResultSet.getBlob, ResultSet.getBinaryStream, and several PreparedStatement.setBlob / setBinaryStream overloads are now supported. ResultSetMetaData.getColumnType for FLOAT columns now returns Types.REAL (previously Types.FLOAT). ResultSet.getString now returns Java null for SQL NULL instead of the literal string "null", in line with the JDBC spec. As a side effect, the SQL CLI now renders BLOB values as X'...' hex literals instead of [B@xxxxxxxx.
  • Fixed an issue where CachedMetadata#invalidateNamespaceNamesCache() did not actually invalidate the cached list of namespaces, causing SHOW NAMESPACES and related operations to potentially return stale results until the cache TTL expired.

Enterprise Options​

ScalarDB Analytics​

Enhancements​
  • Password authentication with access token support.
  • Internal user directory management.
  • gRPC authentication interceptor.
  • SDK and CLI authentication with token caching.
  • Spark data source authentication support.
  • ScalarDB Cluster password authentication backend.
  • gRPC retry with exponential backoff for Client SDK.
Improvements​
  • Set scan fetch size to 4096 for JDBC and ScalarDB data sources.
Breaking changes​
  • Unify ScalarDB namespace to scalardb_analytics and add domain prefixes to table names (registry_, auth_) to avoid PostgreSQL identifier length limits.
Bug fixes​
  • Use TIMESTAMPTZ instead of TIMESTAMP for temporal columns.