10.0.x Resolved Issues¶
10.0.1-20¶
Release Date: 03/08/2022
| Issue Key | Component | Description |
|---|---|---|
| VER-79623 | Optimizer | The optimizer occasionally chose a regular projection when directed to use LAP/Top-K projections. Resolved. 优化器有时会忽略 LAP/Top-K 投影的指示,选择了常规投影。此问题已解决。 |
| VER-79883 | Tuple Mover | MaxDVROSPerContainer now controllable; formula updated to "max(2, (MaxDVROSPerContainer+1)/2)" with two-level strata to avoid excessive DVMergeouts. 用户现在可以控制配置参数 MaxDVROSPerContainer。新公式为 max(2, (MaxDVROSPerContainer+1)/2),采用两层策略以避免过多的 DVMergeout。 |
| VER-80240 | Recovery | After rebalance, Tuple Mover groups ROS containers in batches smaller than MaxMrgOutROSSizeMB; larger containers merged individually. 重新平衡后,Tuple Mover 将 ROS 容器分组为小于 MaxMrgOutROSSizeMB 的批次;较大的容器单独合并。 |
| VER-80542 | Backup/DR | CREATE TABLE AS on a replication target now returns a clearer error message. 在复制目标上执行 CREATE TABLE AS 现在返回更清晰的错误消息。 |
| VER-80747 | Execution Engine | regexp_like() with characters allowing zero-length matches (like ? and *) sometimes returned incorrect results. Fixed.当正则表达式包含允许零长度匹配的字符(如 ? 和 *)时,regexp_like() 有时返回错误结果。已修复。 |
10.0.1-19¶
Release Date: 01/11/2022
| Issue Key | Component | Description |
|---|---|---|
| VER-80160 | UI - Management Console | Apache log4j CVE-2021-45105 fixed in MC. 修复了 MC 中 Apache log4j 的 CVE-2021-45105 漏洞。 |
| VER-80170 | Kafka Integration | log4j CVEs 2021-45105 and 2021-44832 fixed in Vertica/Kafka integration. 修复了 Vertica/Kafka 集成中 log4j 的 CVE-2021-45105 和 CVE-2021-44832 漏洞。 |
| VER-80244 | UI - Management Console | Apache log4j CVE-2021-44832 fixed in MC. 修复了 MC 中 Apache log4j 的 CVE-2021-44832 漏洞。 |
10.0.1-18¶
Release Date: 12/16/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-80084 | Kafka Integration | log4j CVEs 2021-44228 and 2021-45046 fixed in Vertica/Kafka integration. 修复了 Vertica/Kafka 集成中 log4j 的 CVE-2021-44228 和 CVE-2021-45046 漏洞。 |
| VER-80086 | UI - Management Console | log4j CVEs 2021-44228 and 2021-45046 fixed in MC. 修复了 MC 中 log4j 的 CVE-2021-44228 和 CVE-2021-45046 漏洞。 |
| VER-79224 | UI - Management Console | MC failed to import a database containing a table named "users" in public schema. Fixed. MC 导入包含 public schema 中名为 "users" 的表的数据库时失败。已修复。 |
10.0.1-17¶
Release Date: 11/02/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-78728 | Optimizer | Optimizer failed when two NotNull predicates combined into a single Boolean predicate with subqueries. Fixed. 当两个 NotNull 谓词与子查询组合成单个布尔谓词时,优化器失败。已修复。 |
| VER-78904 | EON | Tuple mover mergeouts were not canceled by DML statements in some circumstances. Fixed. 在某些情况下,DML 语句未能取消 Tuple Mover 的 mergeout。已修复。 |
| VER-78947 | UI - Management Console | MC returned errors when email gateway aliases included hyphens. Fixed. 当电子邮件网关别名包含连字符时,MC 返回错误。已修复。 |
| VER-79047 | Optimizer | Queries with multiple distinct aggregates sometimes produced wrong results. Fixed. 包含多个 DISTINCT 聚合的查询有时产生错误结果。已修复。 |
| VER-79067 | Kafka Integration | Filters like KafkaInsertDelimiters returned a VIAssert error with many small messages. Fixed. 在包含许多小消息时,KafkaInsertDelimiters 等过滤器返回 VIAssert 错误。已修复。 |
| VER-79253 | Execution Engine | Sessions with user-defined load code accumulated memory on non-initiator nodes until session ended. Fixed. 具有用户自定义加载代码的会话在非发起节点上累积内存,直到会话结束。已修复。 |
| VER-79416 | Tuple Mover | DVMergeout now calculates column offsets/lengths without accessing storage files, avoiding unnecessary S3 API calls. DVMergeout 现在无需访问存储文件即可计算列偏移量/长度,避免了不必要的 S3 API 调用。 |
10.0.1-16¶
Release Date: 8/25/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-78588 | Database Designer Core | Database Designer generated files with wrong permissions. Fixed. Database Designer 生成的文件权限不正确。已修复。 |
| VER-78363 | UI - Management Console | COPY queries were not shown under Completed Queries on the Query Monitoring Activity page. Fixed. COPY 查询未显示在查询监控活动页面的已完成查询下。已修复。 |
10.0.1-15¶
Release Date: 8/3/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-78358 | Optimizer | Updating a result set using LISTAGG resulted in unequal row counts among projections. Fixed. 使用 LISTAGG 更新结果集导致投影之间的行数不相等。已修复。 |
| VER-78355 | Catalog Engine | Standby nodes did not get changes to the GENERAL resource pool when replacing a down node. Fixed. 替换故障节点时,备用节点未获取对 GENERAL 资源池的更改。已修复。 |
| VER-78348 | Data Networking | Internal connection socket could erroneously close, sending many errors to vertica.log. Fixed. 内部连接套接字可能错误关闭,向 vertica.log 发送大量错误。已修复。 |
| VER-78212 | Catalog Engine | Calling close_session() while running analyze_statistics() on a local temp table sometimes crashed Vertica. Fixed.在本地临时表上运行 analyze_statistics() 时调用 close_session() 有时会导致 Vertica 崩溃。已修复。 |
| VER-78172 | Optimizer | TABLESAMPLE option now supported for views. 视图中现在支持 TABLESAMPLE 选项。 |
| VER-78148 | Data Removal - Delete, Purge, Partitioning | Dropping a local storage location for USER data on clusters with standby nodes returned an error. Fixed. 在具有备用节点的集群上删除 USER 数据的本地存储位置返回错误。已修复。 |
| VER-78133 | Installation Program | The installer now caches get_disks information instead of calling it repeatedly.安装程序现在缓存 get_disks 信息,而不是重复调用。 |
| VER-78125 | Catalog Engine | Vertica now restarts properly for nodes with very large checkpoint files. Vertica 现在可以正确重启具有非常大检查点文件的节点。 |
| VER-77872 | Kafka Integration | The Kafka scheduler no longer hangs in rare cases. Kafka 调度器在极少数情况下不再挂起。 |
| VER-77845 | Kafka Integration | The Kafka Scheduler now allows an initial offset of -3 (begin reading from consumer group offset). Kafka Scheduler 现在允许初始偏移量为 -3(从消费者组偏移量开始读取)。 |
10.0.1-14¶
Release Date: 7/7/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-77829 | Backup/DR | When restoring in coexist mode, STDOUT now contains the correct schema name prefix. 在共存模式下还原时,STDOUT 现在包含正确的 schema 名称前缀。 |
10.0.1-13¶
Release Date: 6/23/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-77182 | Spread | Vertica now properly detects duplicate tokens with the same arq ID. Vertica 现在可以正确检测具有相同 arq ID 的重复令牌。 |
| VER-77247 | Data Removal - Delete, Purge, Partitioning | DML performance improved when loading data into partitioned tables. 向分区表加载数据时,DML 性能得到改善。 |
| VER-77296 | Backup/DR | vbr remove now works for partially deleted restore points.vbr remove 现在可以处理部分删除的还原点。 |
| VER-77390 | Installation Program | Pre-upgrade script now displays more helpful error messages for v10+ databases. 预升级脚本现在为 v10+ 数据库显示更有帮助的错误消息。 |
| VER-77392 | Data load / COPY, Security | Pre-upgrade script error messages improved for v10+ databases. 为 v10+ 数据库改进了预升级脚本的错误消息。 |
| VER-77459 | Optimizer | CASE expressions with constant conditions and string results participating in joins with aggregation could produce internal errors. Fixed. 带有常量条件和字符串结果的 CASE 表达式参与带聚合的连接时可能产生内部错误。已修复。 |
| VER-77524 | Security | Vertica now auto-creates default key projections for a user with DML access performing INSERT into a table with a primary key and no projections. 对于具有 DML 访问权限且向具有主键但无投影的表执行 INSERT 的用户,Vertica 现在自动创建默认键投影。 |
| VER-77572 | Data Collector | Custom data collector settings now properly handled when upgrading to 10.x. 升级到 10.x 时,自定义数据收集器设置现在被正确处理。 |
| VER-77635 | Data Export | More helpful warning message when using deprecated s3export on Minio storage.在 Minio 存储上使用已弃用的 s3export 时,提供更有帮助的警告消息。 |
10.0.1-12¶
Release Date: 5/25/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-76626 | Recovery | Restarting a node with --force on a node with missing/corrupted DFS file now recovers from an UP node or drops it with a warning. 在缺少/损坏 DFS 文件的节点上使用 --force 重启节点,现在可以从可用的 UP 节点恢复或发出警告后删除。 |
| VER-76993 | Optimizer | Unary operators as GROUP BY/ORDER BY in WITH clause statements caused a crash. Fixed. WITH 子句语句中将一元运算符用作 GROUP BY/ORDER BY 导致崩溃。已修复。 |
| VER-77017 | UI - Management Console | In MC Database Designer wizard, the k-safety warning OK button was unresponsive under certain conditions. Fixed. 在 MC Database Designer 向导中,k-safety 警告的 OK 按钮在特定条件下无响应。已修复。 |
| VER-77022 | Optimizer | Calling a view whose underlying query invoked a UDx with '' argument crashed if the table later changed. Fixed. 如果表后续发生更改,调用其底层查询使用 '' 参数调用 UDx 的视图会导致崩溃。已修复。 |
| VER-77080 | Machine Learning | Combined use of ROC function and predict_svm_classifier now handled properly.ROC 函数和 predict_svm_classifier 的组合使用现已正确处理。 |
| VER-77155 | Tuple Mover | Tuple Mover now considers resource pool capacity when creating a mergeout plan, adjusting ROS container count accordingly. Tuple Mover 现在在创建 mergeout 计划时会考虑资源池容量,相应调整 ROS 容器数量。 |
| VER-77192 | Execution Engine | Queries on v_internal tables with equality predicates on relname or nspname using underscore/percent or multiple AND predicates were unoptimized. Fixed.对 v_internal 表的查询在使用 relname 或 nspname 上的等值谓词,包含下划线/百分号或多个 AND 谓词时未优化。已修复。 |
10.0.1-11¶
Release Date: 5/5/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-76578 | UI - Management Console | MC now reflects changes to the resource pool queuetimeout value. MC 现在会反映对资源池 queuetimeout 值的更改。 |
| VER-76622 | Installation Program, UI - Management Console | MC installer optimized to reduce storage footprint. MC 安装程序已优化以减小存储占用。 |
| VER-76767 | Optimizer, Security | For view A.v1 on base table B.t1 with an access policy, users no longer need USAGE on schema B to SELECT the view. 对于基表 B.t1 上带有访问策略的视图 A.v1,用户不再需要 schema B 的 USAGE 权限来 SELECT 该视图。 |
| VER-76771 | Execution Engine | Variable-length-optimized joins had a small crash risk when checking NULL join keys. Fixed. 可变长度优化的连接在检查 NULL 连接键时存在较小的崩溃风险。已修复。 |
| VER-76805 | Client Drivers - JDBC | JDBC client now supports multi-realm Kerberos when user is not from the system's default realm. 当用户不属于系统的默认域时,JDBC 客户端现在支持多域 Kerberos。 |
| VER-76859 | Optimizer | Incorrect usage of EXPORT_TABLES in EXPORT TO VERTICA crashed the database; now returns an error.在 EXPORT TO VERTICA 中错误使用 EXPORT_TABLES 导致数据库崩溃;现在返回错误。 |
| VER-76900 | Optimizer | NO HISTOGRAM event was incorrectly set on the hidden epoch column; no longer set. NO HISTOGRAM 事件被错误地设置在隐藏的 epoch 列上;不再设置。 |
| VER-76940 | Optimizer | Setting a projection column comment to an empty string with COMMENT ON COLUMN caused a crash. Fixed. 使用 COMMENT ON COLUMN 将投影列注释设置为空字符串导致崩溃。已修复。 |
| VER-76972 | Backup/DR | Object restore now functions properly for objects >1 TB on Eon Mode on S3. 在 S3 上的 Eon 模式中,对象恢复现在可以正确处理大于 1 TB 的对象。 |
10.0.1-10¶
Release Date: 3/23/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-76138 | Backup/DR | Vbr now supports restores from MinIO S3 locations containing special characters. Vbr 现在支持从包含特殊字符的 MinIO S3 位置进行还原。 |
| VER-76191 | Depot | System tables reported a file existed in depot after eviction, causing "File not found" errors. Fixed. 系统表在驱逐后报告文件仍存在于 depot 中,导致 "File not found" 错误。已修复。 |
| VER-76253 | Data Export | Dates prior to year 2000 now properly handled when exporting to parquet files. 导出到 parquet 文件时,2000 年之前的日期现已被正确处理。 |
| VER-76257 | Hadoop | Queries of external tables of Parquet files with ZSTD compression now handled properly. 对使用 ZSTD 压缩的 Parquet 文件外部表的查询现已正确处理。 |
| VER-76362 | Execution Engine | REGEXP_REPLACE now properly accepts and returns LONG VARCHAR values. REGEXP_REPLACE 现在正确接受并返回 LONG VARCHAR 值。 |
| VER-76401 | Kafka Integration | Kafka UDXs now support sasl.mechanism types SCRAM-SHA-512 and SCRAM-SHA-256. Kafka UDX 现在支持 sasl.mechanism 类型 SCRAM-SHA-512 和 SCRAM-SHA-256。 |
| VER-76477 | Execution Engine | Complex memory-intensive queries with analytical functions feeding into merge operations could crash on high-core machines with high EXECUTIONPARALLELISM. Fixed. 在高核心数和高 EXECUTIONPARALLELISM 的机器上,分析函数馈送到合并操作的复杂内存密集型查询可能崩溃。已修复。 |
10.0.1-9¶
Release Date: 3/9/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-75754 | Security | x509v1 certificates now properly imported. x509v1 证书现在可以正确导入。 |
| VER-75902 | Tuple Mover | When Mergeout Cache is enabled, dc_mergeout_requests now contains valid transaction IDs instead of zeros.启用 Mergeout Cache 后, dc_mergeout_requests 现在包含有效的事务 ID,而不是零。 |
| VER-75940 | Tuple Mover | Tuple Mover logged excessive PURGE requests while another MERGEOUT ran on the same projection. Fixed. 当另一个 MERGEOUT 在同一投影上运行时,Tuple Mover 记录了过多的 PURGE 请求。已修复。 |
| VER-75974 | Hadoop | Parquet Predicate Pushdown on HDFS sometimes failed to prune correct rowgroups based on predicates. Fixed. HDFS 上的 Parquet 谓词下推有时无法根据谓词正确剪裁 rowgroup。已修复。 |
| VER-76047 | Execution Engine | export_objects accumulated memory via repeated privilege checks, risking OOM. Memory now freed more promptly.export_objects 通过重复的权限检查累积内存,存在 OOM 风险。现在内存释放更及时。 |
| VER-76052 | Execution Engine | Queries with multiple instances of the same ANY ARRAY or IN expression with common sub-expression evaluation enabled could crash. Fixed. 启用公共子表达式评估时,具有多个相同 ANY ARRAY 或 IN 表达式实例的查询可能崩溃。已修复。 |
| VER-76054 | ComplexTypes | Invalid cast on an empty array no longer causes Vertica to fail. 对空数组的无效转换不再导致 Vertica 失败。 |
| VER-76060 | Tuple Mover | Tuple Mover dequeued DVMERGEOUT and MERGEOUT simultaneously, leaving MERGEOUT pending indefinitely. Now checks for outstanding MERGEOUT after DVMERGEOUT completes. Tuple Mover 同时出队 DVMERGEOUT 和 MERGEOUT,导致 MERGEOUT 无限期挂起。现在在 DVMERGEOUT 完成后检查是否有未完成的 MERGEOUT。 |
| VER-76062 | Monitoring | get_expected_recovery_epoch output no longer truncated in clusters with more than 20 nodes.在超过 20 个节点的集群中, get_expected_recovery_epoch 的输出不再被截断。 |
10.0.1-8¶
Release Date: 2/19/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-75301 | Kafka Integration | dc_notifiers now contains root cause information for "Could not send" error.dc_notifiers 现在包含 "Could not send" 错误的根因信息。 |
| VER-75700 | Data Export, S3 | s3export no longer hangs indefinitely with wrong S3 parameter values; now errors out.s3export 使用错误的 S3 参数值时不再无限期挂起;现在会报错退出。 |
| VER-75786 | Machine Learning | RF_CLASSIFIER no longer triggers OOM with certain parameter combinations. RF_CLASSIFIER 在某些参数组合下不再触发 OOM。 |
| VER-75788 | Kafka Integration | Data ingested through Kafka during a timezone change could produce duplicate records. Fixed. 在时区变更期间通过 Kafka 摄入的数据可能产生重复记录。已修复。 |
| VER-75829 | EON | Backing files for temp tables in Eon mode now properly deleted from local storage after table drop. Eon 模式下临时表的支持文件在删除表后现在会从本地存储中正确删除。 |
| VER-75831 | Hadoop | QUERY_EVENTS table now logs events for External Table Paths and RowGroup Pruning. QUERY_EVENTS 表现在记录外部表路径和 RowGroup 剪裁的事件。 |
10.0.1-7¶
Release Date: 2/1/2021
| Issue Key | Component | Description |
|---|---|---|
| VER-75247 | Security | CREATE OR REPLACE VIEW now requires DROP privilege (plus CREATE on schema) rather than requiring ownership. CREATE OR REPLACE VIEW 现在需要 DROP 权限(加上 schema 上的 CREATE 权限),而不是要求所有权。 |
| VER-75283 | Depot | DEPOT_PIN_POLICIES system table was erroneously populated with enterprise storage policies. Now empty in Enterprise mode. DEPOT_PIN_POLICIES 系统表被错误地用企业存储策略填充。在 Enterprise 模式下现在为空。 |
| VER-75500 | Execution Engine | Subquery in an expression with correlated join and aggregation sometimes returned multiple rows per matched join key. Fixed. 在带有关联连接和聚合的表达式中,子查询有时为每个匹配的连接键返回多行。已修复。 |
| VER-75503 | Installation Program | SSH.py no longer logs a syntax error when checking package versions via ssh. SSH.py 在通过 ssh 检查包版本时不再记录语法错误。 |
| VER-75511 | Security | LDAPLink now properly supports attributes with more than 1500 members. LDAPLink 现在正确支持超过 1500 个成员的属性。 |
| VER-75631 | Execution Engine | ALTER TABLE REORGANIZE immediately after setting group partitioning with ALTER TABLE sometimes caused a crash. Fixed. 在使用 ALTER TABLE 设置组分区后立即执行 ALTER TABLE REORGANIZE 有时会导致崩溃。已修复。 |
| VER-75638 | ComplexTypes | QUERY_EVENTS system table will no longer recommend collecting stats on ARRAY & SET columns. QUERY_EVENTS 系统表不再推荐收集 ARRAY 和 SET 列的统计信息。 |
10.0.1-6¶
Release Date: 12/18/2020
Internal hotfix — no customer-facing issues.
10.0.1-5¶
Release Date: 12/8/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-74919 | DDL - Table | ALTER TABLE...OWNER TO now returns NOTICE instead of ERROR when the target table is already owned by the specified user. 当目标表已属于指定用户时,ALTER TABLE...OWNER TO 现在返回 NOTICE 而不是 ERROR。 |
| VER-74921 | ComplexTypes | Creating an external table with a malformed map data type no longer causes an assertion failure. 使用格式错误的 map 数据类型创建外部表不再导致断言失败。 |
| VER-74692 | Storage and Access Layer | Cancelling an INSERT/COPY transaction on Eon could fail under certain conditions. Fixed. 在特定条件下取消 Eon 上的 INSERT/COPY 事务可能失败。已修复。 |
| VER-75010 | Tuple Mover | When reflexive mergeout is enabled, mergeout caching (deprecated) is now automatically disabled to avoid performance issues. 当启用 reflexive mergeout 时,mergeout 缓存(已弃用)现在会自动禁用以避免性能问题。 |
| VER-75011 | S3 | Clearing session-level AWSSessionToken no longer displays a VIAssert error. 清除会话级 AWSSessionToken 不再显示 VIAssert 错误。 |
| VER-75015 | Hadoop | Querying an external table from partitioned parquet with an incorrect filter data type no longer causes node failure. 从具有错误过滤器数据类型的已分区 parquet 查询外部表不再导致节点故障。 |
| VER-75093 | Execution Engine | Inter-node encryption performance degradation addressed across various workloads. 节点间加密的性能退化已在各种工作负载中得到解决。 |
| VER-75244 | Execution Engine | Simple CASE statements could fail with an internal memory error. Fixed. 简单的 CASE 语句可能因内部内存错误而失败。已修复。 |
10.0.1-4¶
Release Date: 11/12/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-74648 | UI - Management Console | MC now prioritizes reachable IP host addresses when trying JDBC connections to the database. MC 在尝试 JDBC 连接到数据库时,现在优先选择可达的 IP 主机地址。 |
| VER-74713 | UI - Management Console | MC now confirms Agent IP address is reachable on a reachable subnet from the MC host. MC 现在会确认 Agent IP 地址在 MC 主机的可达子网上。 |
| VER-74809 | Optimizer | INSERT...SELECT with LIMIT OVER no longer unnecessarily transfers data to the initiator. 带 LIMIT OVER 的 INSERT...SELECT 不再不必要地将数据传输到发起节点。 |
| VER-74826 | Optimizer | Outer joins with LIMIT OVER (PARTITION BY) could cause GUA optimization to prune the inner view prematurely. Fixed — GUA now deferred until after attribute population. 带 LIMIT OVER (PARTITION BY) 的外连接可能导致 GUA 优化过早地修剪内层视图。已修复 — GUA 现在推迟到属性填充之后。 |
10.0.1-3¶
Release Date: 10/28/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-74508 | Backup/DR, Depot, Eon | Eon mode nodes can now recover properly from failures involving lost catalogs. Eon 模式节点现在可以从涉及目录丢失的故障中正确恢复。 |
| VER-74579 | Eon | Eon batch subscriptions now broadcast only leftover commits rather than the entire txn log. Eon 批量订阅现在仅广播剩余的提交,而不是整个事务日志。 |
| VER-74651 | Optimizer | Analytic functions no longer trigger an Optimizer error when used with OVER ORDER BY. 分析函数与 OVER ORDER BY 一起使用时不再触发优化器错误。 |
10.0.1-2¶
Release Date: 10/7/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-74176 | Backup/DR | migrate_enterprise_to_eon now correctly lists all table names with problematic projections in the log file, with proper anchor table names.migrate_enterprise_to_eon 现在在日志文件中正确列出所有具有问题投影的表名,并带有正确的锚表名。 |
| VER-74367 | Execution Engine | A change that adversely affected date() function performance has been fixed.对 date() 函数性能产生不利影响的更改已修复。 |
| VER-74407 | Backup/DR | migrate_enterprise_to_eon now excludes privileges on external procedures and checks dependency consistency.migrate_enterprise_to_eon 现在排除外部过程的权限并检查依赖一致性。 |
| VER-74444 | Execution Engine | METADATA resource pool size was calculated incorrectly in some circumstances. Fixed. 在某些情况下 METADATA 资源池大小计算不正确。已修复。 |
10.0.1-1¶
Release Date: 9/23/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-73869 | Storage and Access Layer | After reading from Google Cloud Storage, TCP connections are now closed when sessions close. 从 Google Cloud Storage 读取后,TCP 连接现在会在会话关闭时关闭。 |
| VER-74002 | Execution Engine | query_profiles now reports correct rows processed when querying external parquet tables on HDFS.查询 HDFS 上的外部 parquet 表时, query_profiles 现在报告正确的处理行数。 |
| VER-74133 | Execution Engine | Sort Order Violation error fixed when join output sorted like target projection was consumed by a partially-sorting Analytic function. 当按目标投影排序的连接输出被部分排序的分析函数消费时,排序顺序违规错误已修复。 |
| VER-74173 | Backup/DR | vbr no longer requires communal storage paths to end with /.vbr 不再要求公共存储路径以 / 结尾。 |
| VER-74174 | Execution Engine | IN () expressions with LONG VARCHAR/VARBINARY types caused slow execution and high memory usage. Fixed.包含 LONG VARCHAR/VARBINARY 类型的 IN () 表达式导致执行缓慢和高内存使用。已修复。 |
| VER-74184 | Admin Tools | MC can now properly revive Eon Mode hosts running on AWS. MC 现在可以正确恢复在 AWS 上运行的 Eon 模式主机。 |
10.0.1-0¶
Release Date: 8/19/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-40651 | UI - Management Console | A security bug on MC fixed. 修复了 MC 上的一个安全漏洞。 |
| VER-40662 | UI - Management Console | MC now disables HTTP TRACE call with embedded Jetty server. MC 现在通过嵌入式 Jetty 服务器禁用 HTTP TRACE 调用。 |
| VER-48151 | Client Drivers - Misc | Result set counts no longer limited to 32-bit integer values. 结果集计数不再限制为 32 位整数值。 |
| VER-56614 | Documentation | DROP_STATISTICS documentation updated — redundant BASE option removed. 更新了 DROP_STATISTICS 文档 — 移除了冗余的 BASE 选项。 |
| VER-60017 | FlexTable | FCSVPARSER now applies NULL parameter to loaded data. FCSVPARSER 现在将 NULL 参数应用于加载的数据。 |
| VER-60547 | Admin Tools | When package updates fail during upgrade, Vertica provides a HINT with manual update instructions. 当升级期间包更新失败时,Vertica 提供带有手动更新说明的提示。 |
| VER-61344 | Installation Program | Optionally use OS package manager to handle dependencies during installation. 可选择使用操作系统包管理器在安装期间处理依赖关系。 |
| VER-65314 | Security | Kerberos updated to 1.18.2 — fixes open file limit exceedance with Kerberos. Kerberos 更新到 1.18.2 — 修复了 Kerberos 打开文件数超限的问题。 |
| VER-65476 | Cloud - Amazon | Local verify scripts no longer error on low nofile/nproc/pid_max/max_map_count/PAM values for instances <128GB. 对于小于 128GB 的实例,本地验证脚本在 nofile/nproc/pid_max/max_map_count/PAM 值较低时不再报错。 |
| VER-69442 | Client Drivers - VSQL, Supported Platforms | VSQL on RHEL 8.0 now handles libnsl dependency properly. RHEL 8.0 上的 VSQL 现在正确处理 libnsl 依赖。 |
| VER-69860 | Optimizer | Queries on projections now use partition-level statistics as needed. 对投影的查询现在根据需要启用分区级统计信息。 |
| VER-70427 | Client Drivers - JDBC | JDBC driver no longer hangs on batch update during table locktimeout; standardized to JDBC SQL standard behavior. JDBC 驱动在表锁定超时期间不再在批量更新时挂起;已标准化为 JDBC SQL 标准行为。 |
| VER-70591 | AP-Geospatial | list_polygons parameter no longer causes initiator failure; new shape_column_length parameter added.list_polygons 参数不再导致发起节点故障;新增了 shape_column_length 参数。 |
| VER-70804 | Execution Engine | Cancelling a query no longer causes subsequent queries in the same transaction to be falsely cancelled. 取消查询不再导致同一事务中的后续查询被错误取消。 |
| VER-71728 | Data load / COPY, SDK | Fenced-mode user-defined parser no longer processes the same data multiple times. 隔离模式用户自定义解析器不再多次处理相同数据。 |
| VER-71983 | ComplexTypes, Optimizer | Directed queries no longer cause "Unknown Expr" warnings with arrays and sets. 定向查询不再对数组和集合产生 "Unknown Expr" 警告。 |
| VER-71997 | Backup/DR | Backups no longer fail to delete old restore points when backup objects exceed 10,000. 当备份对象超过 10,000 个时,备份不再无法删除旧的还原点。 |
| VER-72078 | Optimizer | copy_table now copies projection statistics from source to target table.copy_table 现在将投影统计信息从源表复制到目标表。 |
| VER-72227 | Cloud - Amazon, Security | Vertica instances can now only access the user-specified S3 bucket for communal storage. Vertica 实例现在只能访问用户为公共存储指定的 S3 存储桶。 |
| VER-72242 | ComplexTypes | ALTER TABLE ADD COLUMN with array type now works correctly. 使用数组类型的 ALTER TABLE ADD COLUMN 现在正常工作。 |
| VER-72342 | Backup/DR | Error message improved — "Trying to delete untracked object" now provides repair guidance. 错误消息已改进 — "Trying to delete untracked object" 现在提供修复指导。 |
| VER-72443 | Tuple Mover | Large ROS of inactive partitions is now merged with smaller ROS under certain conditions to avoid pushback. 非活动分区的大型 ROS 现在会在特定条件下与较小的 ROS 合并以避免反压。 |
| VER-72553 | DDL | Renaming an unsegmented non-superprojection no longer errors on conflicts; Vertica modifies the name. 重命名未分段的非超级投影在冲突时不再报错;Vertica 会修改名称。 |
| VER-72577 | Data load / COPY | Slow COPY of poorly written parquet files with wide varchar columns is now fixed. 复制包含宽 varchar 列的编写不良的 parquet 文件速度慢的问题现已修复。 |
| VER-72589 | Optimizer | Projections from pre-10.0.x with pre-aggregated data on partitioned tables with GROUP BY may require rebuilding — REFRESH identifies them. 来自 10.0.x 之前版本的、在带有 GROUP BY 的分区表上具有预聚合数据的投影可能需要重建 — REFRESH 可识别它们。 |
| VER-72603 | DDL | Setting NULL on an already-NULL DEFAULT/SET USING expression now correctly removes the expression when altering column data type. 在已经为 NULL 的 DEFAULT/SET USING 表达式上设置 NULL,现在会在更改列数据类型时正确移除该表达式。 |
| VER-72613 | Security | LDAPLinkSearchTimeout configuration parameter restored. LDAPLinkSearchTimeout 配置参数已恢复。 |
| VER-72625 | Optimizer | COMMENT ON no longer causes views (all_tables, user_functions, etc.) to show duplicate entries. COMMENT ON 不再导致视图(all_tables、user_functions 等)显示重复条目。 |
| VER-72683 | Optimizer | Common sub-expressions in INSERT...SELECT now reused; EXPLAIN rendering of common sub-expressions fixed. INSERT...SELECT 中的公共子表达式现在被重用;EXPLAIN 对公共子表达式的呈现已修复。 |
| VER-72721 | Catalog Engine, Execution Engine | Dropping a column no longer corrupts partition-level statistics. 删除列不再损坏分区级统计信息。 |
| VER-72797 | Data load / COPY | New UDLMaxDataBufferSize parameter (default 256 MB) added to avoid insufficient memory errors during User Defined Load.新增 UDLMaxDataBufferSize 参数(默认 256 MB),以避免用户自定义加载期间的内存不足错误。 |
| VER-72832 | Execution Engine | UUID data type with IN operator now performs equivalently to OR. UUID 数据类型使用 IN 运算符现在的性能与使用 OR 相当。 |
| VER-72852 | Kafka Integration, Security | Scheduler now supports CA bundles at UDX and vkconfig level. 调度器现在支持在 UDX 和 vkconfig 级别使用 CA 捆绑包。 |
| VER-72864 | Optimizer | INSERT with SELECT computing many complex expressions with small result sets — performance improved. INSERT 搭配 SELECT 计算许多具有小结果集的复杂表达式 — 性能已改进。 |
| VER-72886 | Execution Engine | CascadeResourcePoolAlwaysReplan=1 behavior improved for queries that have already started producing results. 对于已经开始产生结果的查询,CascadeResourcePoolAlwaysReplan=1 的行为已改进。 |
| VER-72930 | Monitoring | Two new columns added to dc_resource_acquisitions: session_id and request_id.向 dc_resource_acquisitions 新增两个列:session_id 和 request_id。 |
| VER-72943 | Optimizer | DEPOT_FETCH hint message referenced a deprecated alias; now references the supported hint name. DEPOT_FETCH 提示消息引用了已弃用的别名;现在引用受支持的提示名。 |
| VER-72952 | DDL - Projection | Users without required privileges could create projections using createType hint with L, D, or P arguments. Fixed. 没有所需权限的用户可以使用带有 L、D 或 P 参数的 createType 提示创建投影。已修复。 |
| VER-72991 | Hadoop | When accessing HDFS, Vertica no longer connects to the wrong cluster when multiple namenodes share the same hostname. 访问 HDFS 时,当多个 namenode 共享同一主机名,Vertica 不再连接到错误的集群。 |
| VER-73064 | Control Networking | Adding a column to a wide table with many storage containers could cause server failure. Fixed. 向具有多个存储容器的宽表添加列可能导致服务器故障。已修复。 |
| VER-73080 | DDL | Renamed schema sequences now properly updated in table DEFAULT expressions. 重命名的 schema 序列现在在表 DEFAULT 表达式中正确更新。 |
| VER-73100 | DDL - Table | Dropping a column no longer requires advancing the AHM when the column is referenced as a default expression. 当列被引用为默认表达式时,删除列不再需要推进 AHM。 |
| VER-73102 | DDL - Table | Renamed table columns no longer retain the previous name as an alias in projection DDL. 重命名的表列不再在投影 DDL 中保留先前名称作为别名。 |
| VER-73111 | Backup/DR | MIGRATE_ENTERPRISE_TO_EON now returns names of all tables with unbundled storage containers, recommending COMPACT_STORAGE().MIGRATE_ENTERPRISE_TO_EON 现在返回所有具有未捆绑存储容器的表名,推荐使用 COMPACT_STORAGE()。 |
| VER-73131 | Admin Tools | Locking mechanism changed to prevent admintools.conf.lock from blocking secondary node startup. 锁定机制已更改,以防止 admintools.conf.lock 阻塞辅助节点启动。 |
| VER-73209 | Data load / COPY | FJsonParser rejecting rows could panic the database. Fixed. FJsonParser 拒绝行可能导致数据库 panic。已修复。 |
| VER-73224 | Optimizer | PushDownJoinFilterNull can now also be set at the session level (previously database-level only).PushDownJoinFilterNull 现在也可以在会话级别设置(以前仅限数据库级别)。 |
| VER-73304 | Tuple Mover | If mergeout on a purge request failed due to table lock, later purge requests were blocked. Fixed. 如果由于表锁导致清理请求的 mergeout 失败,后续的清理请求被阻塞。已修复。 |
| VER-73314 | UI - Management Console | Issue selecting various Vertica database versions for MC Data Source fixed. 为 MC 数据源选择不同 Vertica 数据库版本的问题已修复。 |
| VER-73380 | Optimizer | Users now get a hint to increase MaxParsedQuerySizeMB when queries exceed the memory limit.当查询超过内存限制时,用户现在会得到增加 MaxParsedQuerySizeMB 的提示。 |
| VER-73516 | Catalog Engine | Very rare node PANIC during commit causing buddy projection discrepancies after recovery. Fixed. 提交期间极为罕见的节点 PANIC 导致恢复后伙伴投影不一致。已修复。 |
10.0.0-3¶
Release Date: 7/29/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-73015 | ComplexTypes | ALTER TABLE ADD COLUMN with array type now works correctly. 使用数组类型的 ALTER TABLE ADD COLUMN 现在正常工作。 |
| VER-73139 | Optimizer | Common sub-expressions in INSERT...SELECT SELECT clause reused; EXPLAIN rendering of common sub-expressions fixed. (Same as VER-72683) INSERT...SELECT 的 SELECT 子句中的公共子表达式被重用;EXPLAIN 对公共子表达式的呈现已修复。(与 VER-72683 相同) |
| VER-73172 | Hadoop | Fixed connection to wrong HDFS cluster when namenodes from different clusters share the same hostname. (Same as VER-72991) 修复了当不同集群的 namenode 共享同一主机名时连接到错误 HDFS 集群的问题。(与 VER-72991 相同) |
| VER-73261 | DDL - Projection | Users without required privileges could create projections via createType hint. Fixed. (Same as VER-72952) 没有所需权限的用户可以通过 createType 提示创建投影。已修复。(与 VER-72952 相同) |
| VER-73262 | Execution Engine | Cancelling a query no longer falsely cancels subsequent queries in same transaction. (Same as VER-70804) 取消查询不再错误地取消同一事务中的后续查询。(与 VER-70804 相同) |
| VER-73265 | Control Networking | Adding column to wide table with many storage containers could cause server failure. Fixed. (Same as VER-73064) 向具有多个存储容器的宽表添加列可能导致服务器故障。已修复。(与 VER-73064 相同) |
| VER-73442 | Admin Tools | admintools.conf.lock issue preventing startup fixed via changed locking mechanism. (Same as VER-73131) 通过更改锁定机制修复了 admintools.conf.lock 阻止启动的问题。(与 VER-73131 相同) |
| VER-73464 | Optimizer | Queries on projections now use partition-level statistics as needed. (Same as VER-69860) 对投影的查询现在根据需要启用分区级统计信息。(与 VER-69860 相同) |
| VER-73467 | Tuple Mover | Large ROS of inactive partitions now merged with smaller ROS under certain conditions. (Same as VER-72443) 非活动分区的大型 ROS 现在会在特定条件下与较小的 ROS 合并。(与 VER-72443 相同) |
| VER-73468 | DDL | Renamed schema sequences updated in table DEFAULT expressions. (Same as VER-73080) 重命名的 schema 序列在表 DEFAULT 表达式中更新。(与 VER-73080 相同) |
| VER-73470 | Data load / COPY | FJsonParser rejecting rows could panic. Fixed. (Same as VER-73209) FJsonParser 拒绝行可能导致 panic。已修复。(与 VER-73209 相同) |
| VER-73480 | ComplexTypes, Optimizer | Directed queries no longer cause "Unknown Expr" warnings with arrays/sets. (Same as VER-71983) 定向查询不再对数组/集合产生 "Unknown Expr" 警告。(与 VER-71983 相同) |
| VER-73478 | Optimizer | PushDownJoinFilterNull can now be set at session level. (Same as VER-73224)PushDownJoinFilterNull 现在可以在会话级别设置。(与 VER-73224 相同) |
10.0.0-2¶
Release Date: 6/24/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-72840 | Catalog Engine, Execution Engine | Dropping a column no longer corrupts partition-level statistics. (Same as VER-72721) 删除列不再损坏分区级统计信息。(与 VER-72721 相同) |
| VER-72904 | Data Load, COPY / SDK | Fenced-mode user-defined parser no longer processes same data multiple times. (Same as VER-71728) 隔离模式用户自定义解析器不再多次处理相同数据。(与 VER-71728 相同) |
| VER-72908 | Security | LDAPLinkSearchTimeout configuration parameter restored. (Same as VER-72613) LDAPLinkSearchTimeout 配置参数已恢复。(与 VER-72613 相同) |
| VER-72973 | Kafka Integration, Security | Scheduler loads all certs when CA alias omitted; loads only given alias when specified. 省略 CA 别名时,调度器加载所有证书;指定时仅加载给定别名。 |
| VER-72975 | Kafka Integration | Scheduler supports CA bundles at UDX and vkconfig level. (Same as VER-72852) 调度器支持在 UDX 和 vkconfig 级别使用 CA 捆绑包。(与 VER-72852 相同) |
| VER-72996 | Cloud - Amazon, Security | Vertica instances can only access user-specified S3 bucket for communal storage. (Same as VER-72227) Vertica 实例只能访问用户为公共存储指定的 S3 存储桶。(与 VER-72227 相同) |
| VER-73010 | Backup/DR | "Trying to delete untracked object" error message improved with repair guidance. (Same as VER-72342) "Trying to delete untracked object" 错误消息已改进,包含修复指导。(与 VER-72342 相同) |
10.0.0-1¶
Release Date: 6/1/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-72709 | DDL | ALTER TABLE...RENAME now maintains local snapshots to avoid projection name conflicts when renaming multiple tables. ALTER TABLE...RENAME 现在维护本地快照,以避免在重命名多个表时出现投影名称冲突。 |
| VER-72711 | DDL | Setting NULL on an already-NULL DEFAULT/SET USING expression removes the expression when column data type is altered. (Same as VER-72603) 在已经为 NULL 的 DEFAULT/SET USING 表达式上设置 NULL,会在更改列数据类型时移除该表达式。(与 VER-72603 相同) |
10.0.0-0¶
Release Date: 5/7/2020
| Issue Key | Component | Description |
|---|---|---|
| VER-68548 | UI - Management Console | JavaScript was caching TLS checkbox state while importing a database. Fixed. 导入数据库时,JavaScript 缓存了 TLS 复选框状态。已修复。 |
| VER-71398 | UI - Management Console | Exception occurred when a database with no password was used for Extended Monitoring. Fixed. 使用没有密码的数据库作为扩展监控时发生异常。已修复。 |
| VER-69103 | AP-Advanced | Queries with user-defined aggregates or ACD library occasionally returned "DataArea overflow". Fixed. 带有用户自定义聚合或 ACD 库的查询偶尔返回 "DataArea overflow"。已修复。 |
| VER-52301 | Kafka Integration | More helpful error message in rejection table when parsing Avro messages errors. 解析 Avro 消息出错时,拒绝表中提供更有帮助的错误消息。 |
| VER-68043 | Kafka Integration | KafkaAvroParser now reports a more accurate error message for non-Avro schema URLs. KafkaAvroParser 现在为非 Avro schema URL 报告更准确的错误消息。 |
| VER-69208 | Kafka Integration | Example vkconfig launch script now uses nohup to prevent premature scheduler exit. vkconfig 示例启动脚本现在使用 nohup 以防止调度器过早退出。 |
| VER-69988 | Kafka Integration, Supported Platforms | rdkafka issue with new glibc thread support on RHEL 8/Debian 10 causing database crash during COPY via KafkaSource. Fixed. rdkafka 在 RHEL 8/Debian 10 上使用新的 glibc 线程支持导致通过 KafkaSource 进行 COPY 期间数据库崩溃。已修复。 |
| VER-70919 | Kafka Integration | Patch applied for librdkafka issue #2108 (infinite loop causing COPY/KafkaSource hang until server restart). 应用了 librdkafka 问题 #2108 的补丁(无限循环导致 COPY/KafkaSource 挂起直到服务器重启)。 |
| VER-71114 | Execution Engine | Loading large Kafka messages causing an error. Fixed. 加载大型 Kafka 消息导致错误。已修复。 |
| VER-69437 | Supported Platforms | vertica_agent.service can now stop gracefully on RHEL 7.7.vertica_agent.service 现在可以在 RHEL 7.7 上优雅停止。 |
| VER-70932 | License | License auditor could core dump on tables with SET USING expression columns. 许可证审计器在包含 SET USING 表达式列的表上可能 core dump。 |
| VER-67024 | Client Drivers - ODBC | Batch insert row rejection error message improved to reference truncation or null constraint violations. 批量插入行拒绝错误消息已改进,引用截断或空约束违规。 |
| VER-69654 | Third Party Tools Integration | Output buffer now at least input length + 100 bytes for numeric format encryption. 对于数字格式加密,输出缓冲区现在至少为输入长度 + 100 字节。 |
| VER-54779 | Spread | Replacing/adding nodes with a separate control network no longer requires full database restart. 使用单独的控制网络替换/添加节点不再需要完全重启数据库。 |
| VER-69112 | Security, Third Party Tools Integration | NULL input to VoltageSecureProtect and VoltageSecureAccess now returns NULL. 对 VoltageSecureProtect 和 VoltageSecureAccess 的 NULL 输入现在返回 NULL。 |
| VER-70371 | Admin Tools | Log rotation now functions properly in databases upgraded from 9.2.x to 9.3.x. 从 9.2.x 升级到 9.3.x 的数据库中,日志轮转现在正常工作。 |
| VER-70488 | Admin Tools | Starting a subcluster in Eon Mode no longer writes duplicate entries to admintools.log. 在 Eon 模式下启动子集群不再向 admintools.log 写入重复条目。 |
| VER-70973 | Admin Tools | Database Designer no longer errors when targeting non-"public" schemas. 当目标 schema 不是 "public" 时,Database Designer 不再报错。 |
| VER-68453 | Tuple Mover | Tuple Mover can now allocate up to half of MAXCONCURRENCY mergeout threads to inactive partitions. Tuple Mover 现在可以分配最多 MAXCONCURRENCY 一半的 mergeout 线程给非活动分区。 |
| VER-70836 | Optimizer | Optimizer no longer takes excessive time generating plans for tables with complex live aggregate projections. 优化器不再为具有复杂实时聚合投影的表生成计划花费过多时间。 |
| VER-71748 | Optimizer | Queries with single-table predicates and multiple EXISTS subqueries sometimes returned incorrect results. Fixed. 带有单表谓词和多个 EXISTS 子查询的查询有时返回不正确的结果。已修复。 |
| VER-71953 | Optimizer, Recovery | Projection data now remains consistent following MERGE during node recovery in Enterprise mode. 在 Enterprise 模式下节点恢复期间执行 MERGE 后,投影数据现在保持一致。 |
| VER-71397 | Execution Engine | Malformed queries on specific columns could cause database failure. Fixed. 针对特定列的错误查询可能导致数据库故障。已修复。 |
| VER-71457 | Execution Engine | Certain MERGE operations failing to match hash values in optimizedmerge join — null hash values now handled properly. Fixed. 某些 MERGE 操作无法在优化的合并连接中匹配哈希值 — 空哈希值现在被正确处理。已修复。 |
| VER-71148 | DDL - Table | Column constraints with ASCII and non-ASCII characters in names no longer cause failure on insert. 名称中带有 ASCII 和非 ASCII 字符的列约束不再导致插入失败。 |
| VER-71151 | DDL - Table | ALTER TABLE...RENAME now maintains local snapshots for projection name conflicts. (Same as VER-72709) ALTER TABLE...RENAME 现在为投影名称冲突维护本地快照。(与 VER-72709 相同) |
| VER-62046 | DDL - Projection | Partitioning algorithm reversed — now checks projections contain required columns before calculating partition count. 分区算法已反转 — 现在在计算分区数之前检查投影是否包含所需列。 |
| VER-71145 | Data Removal - Delete, Purge, Partitioning | Clearer messages when drop_partitions fails due to insufficient resource pool.当 drop_partitions 因资源池不足而失败时提供更清晰的消息。 |
| VER-70607 | Catalog Engine | In Eon Mode, projection checkpoint epochs on down nodes now become consistent with current epoch when nodes resume. 在 Eon 模式下,当节点恢复时,故障节点上的投影检查点时期现在与当前时期保持一致。 |
| VER-61279 | Hadoop | Loading data into HDFS storage location failing with "Error finalizing ROS DataTarget" — fixed. 将数据加载到 HDFS 存储位置时因 "Error finalizing ROS DataTarget" 失败 — 已修复。 |
| VER-63413 | Hadoop | Export of huge tables to parquet failing with "file not found exception" — fixed. 将大表导出为 parquet 时因 "file not found exception" 失败 — 已修复。 |
| VER-68830 | Hadoop | Better error message and cancelable hang during SASL handshaking when HDFS times out. 当 HDFS 超时时,在 SASL 握手期间提供更好的错误消息和可取消的挂起。 |
| VER-71088 | Hadoop | Vertica now retries HDFS operations returning 403 error code. 对于返回 403 错误码的 HDFS 操作,Vertica 现在会重试。 |
| VER-71542 | Hadoop | INFER_EXTERNAL_TABLE_DDL now compatible with Parquet files using 2-level LIST encoding. INFER_EXTERNAL_TABLE_DDL 现在与使用二级 LIST 编码的 Parquet 文件兼容。 |
| VER-71047 | Backup/DR | With logging level 3, vbr no longer prints dbPassword, dest_dbPassword, or serviceAccessPass. 当日志级别为 3 时,vbr 不再打印 dbPassword、dest_dbPassword 或 serviceAccessPass。 |
| VER-71451 | EON | clean_communal_storage meta-function is now up to 200x faster.clean_communal_storage 元函数现在最多快 200 倍。 |