What's new in the v2.14 LTS release series

Release announcements

Release notes

What follows are the release notes for the v2.14 release series. Content will be added as new notable features and changes are available in the patch releases of the v2.14 release series.

For an RSS feed of all release series to track the latest product updates, point your feed reader to the RSS feed for releases.

Technical Advisory: Use YugabyteDB Anywhere version 2.14.10.1 or later

YugabyteDB Anywhere releases 2.14.9 and 2.14.10.0 contain an issue for airgapped setups, where an airgapped YugabyteDB Anywhere installation fails to upgrade universes.

If you're using these releases, you won't be able to successfully upgrade your universes until you upgrade to YugabyteDB Anywhere 2.14.10.1 or later.

v2.14.16.0 - April 17, 2024

Build: 2.14.16.0-b17

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.16.0-b17

Improvements

DocDB

  • Facilitates a more rapid server initialization by deleting the superblock within the DeleteTablet process when the delete_type is TABLET_DATA_DELETED, reducing the number of DELETED tablet superblocks at server startup. #19840
  • Delivers automatic recovery for index tables affected by a bug previously found and addressed, preventing any future performance issues triggered by incorrectly set property values. #20247
  • Establishes a flag to manage indexing backfills, offering control over whether non-deferred indexes should be batched during the backfill operation. #20213
  • Allows tracking time spent on various operations for remote bootstrap sender and receiver, aiding future enhancement decisions. Also, increases the fsync call threshold on downloaded files to 1G from 8M, reducing overhead for smaller files. #13051

Bug fixes

YSQL

  • Corrects a backup failure issue by ensuring the function yb_catalog_version is introduced, especially in 2.4.x or 2.6.x clusters where it was previously missed due to a YSQL upgrade code bug. #18507

DocDB

  • Handles backfill responses getting interleaved across different operations more gracefully to prevent crashes caused by slow masters or network delays. #20510
  • Now ensures tablets are cleared from the pending deletes state even on task failures, preventing load balancer operations issues. Additionally, fixes a race condition to prevent indefinite retention of tablets in the set after task completion. #13156
  • Resolves issues in the under-replicated endpoint algorithm, ensuring correct counting of replicas only when the block's minimum number of replicas has not been fulfilled yet, hence offering accurate replica tally for placement blocks. #20657
  • Validates the use of two arguments for disable_tablet_splitting, addressing a previous condition where only one was required, thereby enhancing backup process reliability. #8744

Other

  • Updates the condition for HT lease reporting to ensure accurate leaderless tablet detection in RF-1 setup, preventing false alarms. #20919
  • Reduces disruptions by throttling the master process log messages related to "tablet server has a pending delete" into 20-second intervals. #19331

YugabyteDB Anywhere

Improvements

  • Changes the clock skew alert threshold from 500ms to 250ms, enabling detection of clock skew issues before TServer starts crashing, giving users more reaction time. PLAT-13249
  • Adds db_version label to universe_active_task_code for automated stats calculation related to DB version by the QA team. PLAT-12479

Bug fixes

  • Updates s3cmd tool to latest version 2.3.0 for better compatibility with AlmaLinux 8+ and other new Linux distros. PLAT-11559
  • Streamlines backup deletion process by focusing only on completed expired backups, to prevent queue buildup. PLAT-12770

v2.14.15.0 - January 17, 2024

Build: 2.14.15.0-b57

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.15.0-b57

New features

CDC

  • Introduces a flag: enable_tablet_split_of_cdcsdk_streamed_tables (default:false) to toggle automatic tablet splitting for tables in a CDCSDK stream, enhancing user control over replication processes #19482, #19579

Improvements

YSQL

  • Enables the modification of is_single_row_txn for finer control over non-transactional writes required by COPY, index backfill, or when yb_disable_transactional_writes is set, preventing issues during non-bufferable operations for single row transactions. #4906
  • Integrates an upstream PostgreSQL commit to streamline error checking in the pgcrypto:Check for error return of px_cipher_decrypt, a crucial step towards supporting OpenSSL 3.0+. #19732
  • Enables a prerequisite import of upstream PG commit Disable OpenSSL EVP digest padding in pgcrypto, paving the way for OpenSSL 3.0+ upgrade. #19733
  • Allows for the import of upstream PG commit Add alternative output for OpenSSL 3 without legacy loaded, aiding OpenSSL 3.0+ upgrade. #19734
  • Prevents potential SQL injections in extension scripts by rejecting substitutions of extension schemas or owners, mitigating vulnerability CVE-2023-39417. #14419
  • Mitigates potential security issues by importing upstream Postgres commit from REL_11_STABLE for future support of ALTER <object> DEPENDS ON EXTENSION, although it's not currently available in YugabyteDB. #14419
  • Increase the oom_score_adj (Out of Memory score adjustment) of the YSQL webserver to 900, the same as PG backends, prioritizing its termination when it significantly consumes memory. #20028
  • Enhances debugging capabilities in YSQL Webserver process by introducing two new flags for endpoint access logging and tcmalloc stats printing. #20157

DocDB

  • Prevents tservers from communicating with master leaders in a different universe, eliminating incorrect registrations and data loss risks, gated by a new autoflag master_enable_universe_uuid_heartbeat_check = true, requiring user to manually update post-upgrade. #17904
  • Prevents potential lags and errors in the WaitOnConflict workload testing by ensuring in-memory locks get properly released, even in the event of a failed conflict resolution sequence. This fix helps avoid lock-related issues that might occur with any leaked or stuck WriteQuery objects. #18770
  • Allows for more detailed server thread analysis and monitoring in live production situations via the /threadz endpoint. #16472
  • Introduces support for rocksdb_check_sst_file_tail_for_zeros flag, enabling verification of SST data file trailing zeros and reporting errors for anomalies detected. #19691
  • Adds support for retrying writes to SST files in cases of detected corruption after writes, controllable via the rocksdb_max_sst_write_retries flag. #19730
  • Rejects ConfigChange requests for system catalog while another server is transitioning, preventing potential data loss from mistaken quorum formation by new peers. #18335
  • Upgrades OpenSSL version to 3.0.8 to replace the outdated 1.1.1, incorporating updates to third-party archives and disabling Linuxbrew builds to ensure continued functionality. #19736
  • Reduces time window for retryable requests to match Write RPC timeouts, preventing the replay of unnecessary logs in YCQL tablets and offering protection against expired requests. #18736
  • Logs tablet metadata create/update instances, giving additional insights and reducing crash loops on TServer startup by avoiding duplicate tablet metadata records. #20042
  • Enables tracing of the UpdateConsensus API by turning on the collect_update_consensus_traces flag, which collects traces from remote followers and adds a trace message when appended to the local log. #19417
  • Enables automatic recovery of index tables for bugs detected in the context of all new index tables created from 2.14; prevents retention of tombstones in SST files, mitigates performance degradation and disk size leak. #19731
  • Enhances debugging experience by adding verbose logs for load balancer state changes, updating the global state in verbose logs, and logging state transitions per table state. #20289

YugabyteDB Anywhere

Improvements

  • Improves cluster resilience during cluster creation and when modifying a cluster by adding safeguards and controls to utilize a new, more robust and controlled master_join_existing_universe API. This applies only to VM-based universes (and not Kubernetes universes). PLAT-10511
  • Updates Prometheus version to 2.47.1 from 2.46.0 to mitigate security vulnerabilities. PLAT-10838
  • Upgrades node exporter from version 1.6.1 built with a vulnerable version of golang to a safer 1.7.0, enhancing security. PLAT-11234
  • Removes NGINX from Kubernetes deployments, enabling HSTS support, HTTPS on Play framework, and the addition of custom HTTP headers for enhanced security measures. PLAT-6893,PLAT-10837,PLAT-9109,PLAT-8063,PLAT-8514,
  • Upgrades node exporter from version 1.6.1 built with a vulnerable version of golang to a safer 1.7.0, enhancing security. PLAT-9635
  • Updates Java libraries, including Okio, Http3, Amazon SDKs, Akka-actor, Play, Docker, AWS-java-sdk-s3, and Setuptools, to address vulnerabilities and enhance security. PLAT-11227,PLAT-11619
  • Updates to Apache Commons libraries have increased security by addressing vulnerabilities in various packages, including commons-collections, -value, /x/net, httpclient, and -classic/core. PLAT-11657

Bug fixes

YSQL

  • Trims down the probability of inaccurate behaviour involving conflicts between single shard INSERT operations by ensuring read times are chosen after conflict resolution, enhancing data consistency. #19407
  • Addresses mismatches that occurred due to non-execution of the postprocess script on alternate expected files and default_expectfile in pg_regress. #19737
  • Resolves a bug that hindered consecutive lock acquisition due to not releasing LWLocks during AbortTransaction, preventing PostgreSQL processes from getting stuck. #19509
  • Prevents potential postmaster crashes during cleanup of killed connections by using the killed process's ProcStruct to wait on an unavailable LWLock. #18000
  • Ensures table's dependent objects and metadata, including views, materialized views, policies, statistics objects, and grants, are preserved when altering a table's primary key. #15245
  • Reduces spinlock deadlock detection time by 75% for prompt handling of potential freezes and restarts Postmaster when a process holding a spinlock is killed, ensuring successful initiation of new connections. #18272, #18265
  • Restarts postmaster for easier and more effective cleanup when a process is accidentally killed during its own initialization or cleanup, avoiding potential issues with shared memory items handling. #19945
  • Corrects an error during table rewrite processes that previously omitted the relforcerowsecurity, preserving the Row-Level Security (RLS) behavior. #19815
  • Refines background worker processes structure to prevent loss of Proc struct after killing a background worker, ensuring continued webserver functionality. #20154
  • Resolves the segmentation fault issue in postmaster, enhancing stability when acquiring already owned locks with the implementation of KilledProcToCleanup, mitigating rare circumstances of crash during lock dequeuing. #20166
  • Restart the postmaster whenever a process is killed while writing in a critical section, preventing corrupted state and CPU consumption due to infinite loops. #20255
  • Limits the number of attempts to read inconsistent back-end entries at 1000 to prevent indefinite waiting, proceeding with potentially incomplete yet safe results if consistently fails. #20274
  • Sets yb_pg_batch_detection_mechanism to DETECT_BY_PEEKING by default, for safer usage on older versions, mitigates "Illegal state: Used read time is not set" error with INSERTs. #20099, #16446
  • Resolves segmentation faults in the webserver SIGHUP handler during shutdown phase. #20309
  • Adds pg_stat_statements.yb_qtext_size_limit flag to control maximum query text file size that can be read into memory, preventing large or corrupted files from crashing the process. #20211

DocDB

  • Mitigates instances of hybrid time too low error by requiring new leaders to only accept writes when the previous leader's lease has expired. #18121
  • Allows tracing of outgoing calls only if the current RPC is being traced, reducing excessive memory consumption and logging. #19497
  • Restores the original behavior that doesn't count tablets on dead tservers towards the replica count in the under-replicated endpoint after a master leader failover. #19514
  • Resolves issues with ycql, snapshots, and metrics by migrating the TableInfo namespace_name field through the master migration framework, first introduced in version 2.3, with the namespace loader now preceding the table loader. #17713
  • Reduces index size growth and enhances query speed by properly updating the retain_delete_markers field on index backfilling, addressing an issue where delete markers persisted in SST files. #19544
  • Increases tablet_creation_timeout_ms from 30 seconds to 600 seconds to prevent tablet replacement during table creation, improving restore operations in network latency sensitive environments. #13030
  • Refines meta cache updates to avoid overwriting child tablets and consequently causing stale data, ensuring more accurate partition map refreshes. #18732
  • Resolves a use-after-free issue in RefinedStream's failure path to enhance system stability and prevent memory corruption. #19727
  • Fixes unbalanced hash codes distribution in PartitionSchema::CreateHashPartitions, ensuring equal hash code allocation across partitions thus eliminating hot shard issues. #6597
  • Fixes an intermittent index creation failure by checking the result of is_running, ensuring stability for empty YCQL tables, and preventing retain_delete_markers from being incorrectly set. #19933
  • Resolves a concurrency issue in MasterChangeConfigTest by operating on a copy of ExternalMaster* to prevent timing out when removing a follower with a transitioning server. #19927
  • Streamlines handling of single shard waiters by replicating existing conflict checking logic typically used for distributed transactions, ensuring waiters unblock properly after a blocking subtransaction rollback. #20113
  • Deprecates the enable_process_lifetime_heap_sampling flag, allowing control over tcmalloc sampling only with the profiler_sample_freq_bytes flag for better accuracy. #20236
  • Adjusts SysCatalog tablet's retryable requests retain duration to prevent the request is too old error in YSQL DDL after 60 seconds, enhancing user's cross-database operation execution. #20330

CDC

  • Eliminates deadlocks during database drops by properly handling namespace-level CDC stream deletion and setting up the database deletion process to finish in a finite time. #19879

YugabyteDB Anywhere

  • Improves cluster resilience when masters are removed from the cluster. Specifically, disables a callback for removed master processes. This avoids possible crashes & failures after a master is removed. PLAT-10401

Other

  • Avoids a segmentation fault in the stats collector following a Postmaster reset, solving potential crashes experienced when queries are terminated by SIGSEGV or SIGKILL. #19672

v2.14.14.0 - October 19, 2023

Build: 2.14.14.0-b17

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.14.0-b17

New features

N/A

Improvements

Database

  • [15909] [DocDB] Move sys catalog writes out of catalog loader
  • [17716] [DocDB] Move tcmalloc utility code out of MemTracker
  • [17758] [18561] [DocDB] Enable allocation sampling endpoint for gperftools tcmalloc
  • [18998] [DocDB] New flag to force shell mode for empty masters regardless of master_addresses
  • [19097] [DocDB] Escape output of master and tserver UI endpoints
  • [19221] [DocDB] Improved timeout handling for YCQL index scan
  • [19454] Move macOS support off of mac11

YugabyteDB Anywhere

  • [PLAT-9351] Improve backup script logging
  • [PLAT-9397] Disable TLS for s3cmd
  • [PLAT-9949] [PLAT-7719] Enhance platform logging based on runtime flag
  • [PLAT-10045] Bump up certifi version
  • [PLAT-10371] Change DB Queue Overflow alert label
  • [PLAT-10407] Add ap-south-2 region in YBA

Bug Fixes

Database

  • [15362] [DocDB] Reuse regular RocksDB upper bound for intents RocksDB
  • [18744] [DocDB] Add ability to recover from follower lag caused by stuck OutboundCall

YugabyteDB Anywhere

  • [PLAT-8852] [YBM] Incorrect conf file for tserver
  • [PLAT-9385] [PLAT-9408] Propagate --cluster_uuid gflag to all masters/tservers from YBA during master/tserver startup
  • [PLAT-10531] Better error handling for bash script as part of postgres upgrade
  • [PLAT-10628] [PLAT-10542] Missing Node UUID causes certain tasks to fail in new YBA releases
  • [PLAT-10629] Fix 296 migration

Known issues

N/A

v2.14.13.0 - September 20, 2023

Build: 2.14.13.0-b13

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.13.0-b13

New features

N/A

Improvements

Database

  • [16963] [18838] [DocDB] Add verbose logging to the load balancer.
  • [18826] [Build] thirdparty_tool --list-compilers option enhancement
  • [18668] [DocDB] Block stack collection during thread creation
  • [18770] [DocDB] Stricter memory order in PreparerImpl::Run to prevent re-ordering of instructions

YugabyteDB Anywhere

  • [PLAT-6180] Show YB DB Version on Universe tile in dashboard and universe list
  • [PLAT-9486] Write to files & run sync in liveness probe
  • [PLAT-10010] Add active universe task code metric
  • [PLAT-10037] Upgrading azcopy to 10.20.1
  • [PLAT-10039] Update node exporter version
  • [PLAT-10041] Upgrade net.minidev:json-smart to 2.5.0
  • [PLAT-10079] Bump up postgres version
  • [PLAT-10156] mask certContent in audit logs
  • [PLAT-10419] Pass schedule backup parallelism in GET schedule API response

Bugs

Database

  • [13128] [YSQL] Fix ClearSeparateDdlTxnMode error handling
  • [13171] [DocDB] reject expired retryable requests
  • [16432] [17080] [xCluster] Disallow N:1 replication topology
  • [16446] [YSQL] Make yb_pg_batch_detection_mechanism configurable at runtime for session, peek message following Execute from JDBC driver to identify batch execution
  • [17146] [xCluster] Fix missing pgschema_name during SetupUniverseReplication
  • [17148] [DocDB] Fixed wrong metric_type of prometheus metric
  • [17682] [DocDB] fix hybrid time too low due to race between TransactionParticipant::Poll and RaftConsensus::Start in rf-1 setup
  • [18281] [YSQL] Fix query layer's read restart retries in Read Committed isolation
  • [18652] [YSQL] YB Sequential Scans don't take locks in SERIALIZABLE isolation
  • [18879] [YSQL] Disable pushdown for TextSearch functions that are not thread safe

YugabyteDB Anywhere

  • [PLAT-4772] Allow stop + remove to wait for data to move off tserver
  • [PLAT-9814] Do not garbage collect the customer task which is still being referred in the universe task or the last task for a provider

Known issues

N/A

v2.14.12.0 - August 30, 2023

Build: 2.14.12.0-b19

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.12.0-b19

New features

N/A

Improvements

Database

  • [17805] [17965] YSQL Enhance TServer PG daemon robustness
  • [18039] [YCQLSH] Updating yugabyte-db-thirdparty release version
  • [18344] Test: Stabilize test org.yb.cql.TestSelect#testToJson
  • [17570] [DocDB] improve leaderless tablet endpoint

YugabyteDB Anywhere

  • [PLAT-9643] Update guava version
  • [PLAT-9787] Changed raw -> command in ansible install-s3cmd.yml playbook
  • [PLAT-9950] Nginx upgrade version to v1.25.1
  • [PLAT-9951] Up-version prometheus to 2.46.0

Bugs

Database

  • [17683] [DocDB] Invalidate YCQL system.partitions cache on table creation
  • [17867] [DocDB] Account for table placement and read replicas in underreplicated endpoint
  • [17558] [DocDB] Issue with picking local_limit during paging read
  • [14270] [DocDB] Fixes test SnapshotTxnTest.BankAccountsWithTimeStrobe
  • [12400] [11873] [DocDB] Obtain Tablet from TabletPeer as a shared_ptr to avoid incorrect memory access
  • [17848] [DocDB] fix race on RemoteTablet::replicas_ between RemoteTablet::GetRemoteTabletServers and RemoteTablet::Refresh
  • [18164] [DocDB] Ignore deleted tablets in leaderless endpoint
  • [18518] [DocDB] removed unused MemTracker::rand_ field
  • [17961] [18008] [YSQL] Fix max connection limit errors with additional locking
  • [18465] [DocDB] Dump PeerManager to tserver debug ui consensus page
  • [18592] [DocDB] Fix log message format in WritePostgresConfig
  • [18344] [YCQL] Fix TOJSON() to also unpack nested FROZENs

YugabyteDB Anywhere

  • [PLAT-9613] Ensure CPU usage deeplinks to appropriate metric panel
  • [PLAT-9642] Remove the dependency for net.minidev package
  • [PLAT-9620] EditUniverse retries after failures can cause unpredictable behavior.
  • [PLAT-9926] Stop master process before removing it from masters in 2.14 branch
  • [PLAT-5756] Fixing systemd nodeops

Known issues

N/A

v2.14.11.1 - August 7, 2023

Build: 2.14.11.1-b1

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.11.1-b1

New features

Database

N/A

Improvements

Database

N/A

YugabyteDB Anywhere

N/A

Bugs

Database

N/A

YugabyteDB Anywhere

  • [PLAT-9926] Stop master process before removing it from masters in 2.14 branch

Known issues

N/A

v2.14.11.0 - July 20, 2023

Build: 2.14.11.0-b35

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.11.0-b35

New features

Database

  • [16513] Add statistics to track docdb keys found (total, obsolete, and obsolete past history cutoff)
  • [16794] [DocDB] Make full compactions non-exclusive by default
  • [17629] [YSQL] Add waitpid status code decoded messages to log
  • [17633] [YCQL] Introduce GFlag to control local calls on RPC thread

Improvements

Database

  • [14473] [DocDB] Accept registrations from tservers on the same node with greater sequence numbers.
  • [15186] [YSQL] expression pushdown forward compatibility
  • [16157] [YSQL] Set PgClient session timeout value
  • [17398] [CQL] Account for user query and keyspace name for prepared statements
  • [17966] [DocDB] Unify ModifyPlacementInfo string parsing
  • [18064] [DocDB] enhance tserver consensus state page

YugabyteDB Anywhere

  • [PLAT-7729] Updating version of snakeyaml to 2.0 to fix CVEs, also making required code changes to make code compile
  • [PLAT-9144] Add UI validation for duplicate region on legacy provider form
  • [PLAT-9221] [PLAT-9222] Bump prometheus/nginx version
  • [PLAT-9353] Optimize find command in yb_backup.py
  • [PLAT-9375] bump up cffi version
  • [PLAT-9534] Removing thirdparty-deps from values.yaml since its unused
  • [PLAT-9282] Set context for ".ssh" directory when SELinux is enabled

Bugs

Database

  • [13395] [YCQL] Return error for unset bind PK value
  • [14762] [DocDB] Prevent crashing the master if trying to restore a deleted table
  • [14926] [DocDB] Prevent intent clean-up while reading YSQL system tables
  • [15166] [DocDB] Fix crashes due to tablet shutdown timeout
  • [16009] [YSQL] Fix the system catalog versioning behaviour on CREATE TABLE with FK reference command
  • [16367] [DocDB] Delete Snapshot keeps on retrying indefinitely after master failover
  • [16508] [YSQL] Fix incorrect tablet count assignment on user tables created within geo-partitoned tablespace
  • [17038] [DocDB] Async open table while updating indexes
  • [17074] [DocDB] Root memory consumption should not include pageheap_free_bytes
  • [17142] [YSQL] Fix incorrect results with functional index on 2.14
  • [17261] DST: PITR - Single file compaction does not reset Hybrid Time filter
  • [17318] [YSQL] Import 'Replace last PushOverrideSearchPath() call with set_config_option().' from PG
  • [17500] [DocDB] Fix test YbAdminSnapshotSchedule.PgsqlCreateTable in case of colocated DBs
  • [17574] [DocDB] Add EncryptedWritableFile::Size() to fix incorrect WAL footer start offset
  • [17592] [17760] [DocDB] Rename WAL file after writing the file header on the disk
  • [17603] [DocDB] Tablet splitting should respect FLAGS_max_create_tablets_per_ts.
  • [17663] [DocDB] Clear drive load / leader load maps before appending to them
  • [17712] [DocDB] Fix system partitions query to not return empty namespace name when its not set at table level
  • [17726] [DocDB] A workaround for memory leak from InboundCall weak pointers
  • [17895] Fix ASAN failures due to bad merge

YugabyteDB Anywhere

  • [PLAT-7236] Exception in change master config is eaten up leading to master leader election failure
  • [PLAT-9016] MinIO - Server side encryption specified but KMS is not configured
  • [PLAT-9040] Backup Prometheus snapshot data when exclude_prometheus is not specified
  • [PLAT-9155] [PLAT-9119] For Universe created using AWS provider gp2 is always used for the root volume even if gp3 is selected
  • [PLAT-9225] [PLAT-9374] Azcopy/Node Exporter vulnerabilities
  • [PLAT-9334] [PLAT-9333] Incorrect number of masters for RF=5 and numNodes=6
  • [PLAT-9350] Clean checksum files for backup and checksum/checksum-downloaded files for restore

Known issues

N/A

v2.14.10.4 - November 21, 2023

Build: 2.14.10.4-b1

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.10.4-b1

Improvements

Database

  • [19691] [DocDB] Added SST files tail check for zero bytes
  • [19730] [DocDB] Added SST write retries on corruption

v2.14.10.3 - October 3, 2023

Build: 2.14.10.3-b2

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.10.3-b2

New features

Database

N/A

Improvements

Database

N/A

YugabyteDB Anywhere

[PLAT-10644] Upgrading azcopy/node exporter

Bugs

Database

  • [18744] [DocDB] Add ability to recover from follower lag caused by stuck OutboundCall
  • [18465] [DocDB] Dump PeerManager to tserver debug UI consensus page

YugabyteDB Anywhere

N/A

Known issues

N/A

v2.14.10.2 - June 15, 2023

Build: 2.14.10.2-b1

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.10.2-b1

New features

Database

N/A

Improvements

Database

N/A

YugabyteDB Anywhere

N/A

Bugs

Database

  • [17712] [DocDB] Fix system partitions query to not return empty namespace name when not set at table level

YugabyteDB Anywhere

N/A

Known issues

N/A

v2.14.10.1 - June 8, 2023

Build: 2.14.10.1-b1

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.10.1-b1

New features

Database

N/A

Improvements

Database

N/A

YugabyteDB Anywhere

N/A

Bugs

Database

  • [14473] [DocDB] Accept registrations from tservers on the same node with greater sequence numbers

YugabyteDB Anywhere

  • [PLAT-8920] Adds the missing AWS regions in the UI
  • [PLAT-8833] Upgrade pip only during venv build time
  • [PLAT-9016] MinIO - Server side encryption specified but KMS is not configured

Known issues

  • N/A

v2.14.10.0 - May 30, 2023

Build: 2.14.10.0-b23

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Note: Use YugabyteDB Anywhere release 2.14.10.1 or later.

Docker

docker pull yugabytedb/yugabyte:2.14.10.0-b23

New features

Database

  • [17194] [CQL] Add memtracker to track CQL requests memory usage
  • [17358] [DocDB] Add tracking for memory used by YBMetadataCache

Improvements

Database

  • [16849] [DocDB] Tablet splitting: Lower the thresholds for splits on low core machines
  • [17041] [DocDB] Optimize YCQL JSONB attribute-level upsert performance
  • [17276] [xCluster] cleanup the transaction to write_id map whenever transaction is applied

YugabyteDB Anywhere

  • [PLAT-8570] Don't provide table name list to yb_backup during YCQL restore
  • [PLAT-8679] Incorporate increasing read replica RF in the YBA UI
  • [PLAT-8679] Support for Read Replica's Replication Factor greater than 7
  • [PLAT-8697] Bump up prometheus/nginx version
  • [PLAT-8697] Bump up prometheus/postgres version
  • [PLAT-8841] Track number of region which have zones configured during bootstrapping for legacy onprem provider form

Bugs

Database

  • [13255] [DocDB] Use ThreadLocal instead of ThreadSafeRandom in Trace::NewTrace
  • [13482] [DocDB] Config changes in Leader need not wait on servers in transition
  • [13862] [YSQL] Fix metric for total number of connections made
  • [16108] [DocDB] Start transaction promotion immediately even if not ready yet
  • [16131] [DST] B/R, PITR - Fix invalid enum value of restoration state object
  • [16329] [YSQL] Fix metric for rejections due to connections limit
  • [16565] [DocDB] Fixed missing data in case of RocksDB iterator error
  • [16632] [DocDB] Fix TestIndex.testDropDuringWrite
  • [16669] [DocDB] Fix tablet splitting / PITR deadlock
  • [16697] [DocDB] Disallow multiple OpenTable calls in YBMetaDataCache::GetTable
  • [16875] [DST] PITR - Extend master's sys catalog retention based on snapshot schedule
  • [16999] [DocDB] Retry read if read buffer was full on previous step
  • [17107] [YCQL] Prevent DROP TYPE for the used YCQL type
  • [17163] [YCQL] Use Metadata Cache by default in PREPARE
  • [17165] [CDCSDK] Added serve_as_proxy member to GetCheckpointRequestPB
  • [17219] [YCQL] Fixing tests AlterDropAdd*
  • [17362] Avoid double memory tracking of request data and fix tracking for plain response buffer size

YugabyteDB Anywhere

  • [PLAT-8652] Add customer in the condition of query to fetch universe definition fields while listing tasks
  • [PLAT-8896] [PLAT-5848] Fix increasing nodes for RR
  • [PLAT-8900] [PLAT-4504] Fix FATAL health check on systemd failure

Known issues

  • N/A

v2.14.9.0 - May 1, 2023

Build: 2.14.9.0-b23

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Note: Use YugabyteDB Anywhere release 2.14.10.1 or later.

Docker

docker pull yugabytedb/yugabyte:2.14.9.0-b23

Improvements

Database

  • [506] [YCQL] Disable alter table rename on columns with index
  • [3090] [YCQL] Limit topology change events to subscribed connections
  • [12826] [16571] [DocDB] Support traces to be enabled based on sampling
  • [13718] [DocDB] Remove const modifier for iterator HasNext and GetNextReadSubDocKey
  • [14886] [DocDB] Adds compaction types to compaction task metrics
  • [15705] [CDCSDK] Update CDCSDK checkpoint non-actionable replicated messages in WAL
  • [16091] [DocDB] Changing post-split compaction to run non-exclusively
  • [16581] [DocDB] Remove outstanding tablet split limit by default.
  • [16666] [YSQL] Switch to new query context for get_tablespace_distance() operations
  • [16868] [DocDB] Sanitizes debug messages

YugabyteDB Anywhere

  • [PLAT-7540] Make default_memory_limit_to_ram_ratio flag not hidden
  • [PLAT-7552] Logging of secret credentials in application.log for provider with custom ssh_key
  • [PLAT-7607] Support for table by table backup in YCQL
  • [PLAT-8319] UI Support for table by table backup
  • [PLAT-8363] Put alert rule strings in double quotes
  • [PLAT-8419] Use bulk fetch instead of SQL query for every row of task info in commissioner class

Bugs

Database

  • [15090] [DocDB] Tablet splitting: Some queries fail with "The tablet is shutting down" during the workload
  • [15220] [15786] [15947] [DocDB] Fix apply-operation timeout during schema changes
  • [15746] [DocDB] harden master leaderless tablet endpoint
  • [16444] [DocDB] Ignore heartbeats from peers not in reported config
  • [16789] [YSQL] Assume that CAPABILITY_PickReadTimeAtTabletServer is enabled all the time
  • Re-used master nodes are not reinitialized, leading to under-replication

YugabyteDB Anywhere

  • [PLAT-7269] Added a retry for ModifyMasterClusterConfig on config version mismatch
  • [PLAT-8516] Backups not getting cleared after retention is complete
  • [PLAT-8570] Don't provide table name list to yb_backup during YCQL restore

Known issues

  • N/A

v2.14.8.0 - April 4, 2023

Build: 2.14.8.0-b17

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.8.0-b17

New features

Database

  • [7614] [DocDB] Add an option to run a major compaction at a fixed interval
  • [16124] [DocDB] Add mem_tracker for RunningTransaction objects in TransactionParticipant
  • [16351] [DocDB] Add Cql Tests to Simulate Update Index Callback Leak

Improvements

YugabyteDB Anywhere

  • [PLAT-4642] Set -c huge_pages=off in yugaware helm chart
  • [PLAT-7416] [PLAT-7541] remove internal unsupported API docs.
  • [PLAT-7507] Remove deprecated route toggle_tls
  • [PLAT-7682] Bump up prometheus/postgres version

Database

  • [10696] [YSQL] Avoid reading sys catalog from followers
  • [13163] [CDCSDK] Making CDC checkpoint logs on tablet peer periodic
  • [15629] [DocDB] Update last_received_current_leader sent from replica to leader by including de-duped messages as well
  • [15794] [DocDB] Refactor AreLeadersOnPreferredOnly to Account for Geo Partitioned Tables
  • [15883] [DocDB] Check permission status of YCQL indices in CatalogManager::IsDeleteTableDone
  • [16006] [DocDB] execute callbacks for local RPC calls the same way as for remote calls
  • [16163] [DocDB] Change IsDeleteTableInProgress treatment of alive Index
  • [16474] [DocDB] Change PgClientService session_expiration_queue_ to min heap
  • [16351] [DocDB] Release RequestScope in WriteQuery::ExecuteDone()

Bugs

Database

  • [13850] [DocDB] fixed RemoteBootstrapITest.TestLongRemoteBootstrapsAcrossServer
  • [15528] [DocDB] Fix usage of Connection::outbound_data_being_processed_ during Shutdown
  • [16138] [DocDB] Tablet splitting should be blocked on system_postgres.sequences_data
  • [16457] [YSQL] Fix segmentation fault after rollback of subtransaction with deferred

Known issues

  • N/A

v2.14.7.0 - March 10, 2023

Build: 2.14.7.0-b51

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.7.0-b51

New features

YugabyteDB Anywhere

  • [PLAT-6481] Add a check for the YSQL Postgres server during the rolling restart
  • [PLAT-6358] Allow defining prometheus remote write config in helm chart

Database

  • [10595] [YSQL] Add progress reporting for create index commands
  • [13187] [YSQL] Enable critical (nailed) indices
  • [15032] [CDCSDK] Alter table support with CDC snapshot
  • [15368] [YCQL] Add mechanism to reset password as a non-authenticated user
  • [15499] Enable lease revocation

Improvements

YugabyteDB Anywhere

  • [PLAT-5324] [UI] Backup details of a scheduled backup shows backup type as On Demand
  • [PLAT-5876] [PLAT-6491] Get rid of import org.apache.commons.text usage
  • [PLAT-5876] Again remove apache-commons-text dependency
  • [PLAT-6026] [PLAT-6029] Make top K query use single query + return lines in correct order
  • [PLAT-6239] Upgrade the postgres version to 14.4
  • [PLAT-6305] Upgrade azcopy version to 10.16.2
  • [PLAT-6519] Collect only YBDB pod's metrics from K8s nodes
  • [PLAT-6537] Bump up paramiko version
  • [PLAT-6635] [PLAT-6304] resize_node REST API parameter validation does not consider multi-cluster Universes
  • [PLAT-6648] [PLAT-3971] Validation for inconsistent settings between main cluster and read replica
  • [PLAT-6754] Add warning to create universe API toggles
  • [PLAT-6755] [PLAT-6756][UI] Platform comes out from the universe page after pausing and rolling restart of universe
  • [PLAT-6862] Add hsts to replicated's nginx config
  • [PLAT-7137] Add io-netty dependency back
  • [PLAT-7286] Bump YBA postgres version to 14.6

Database

  • [12575] [DocDB] Improved CqlTabletSplitTest.SecondaryIndex stability
  • [12631] [DocDB] [YSQL] Update PGTableDesc table cache when partition list version is outdated
  • [12959] [YSQL] Refactor code to disambiguate in txn limit and index backfill read time
  • [13040] [YSQL] Prefetch pg_type table on cache refresh
  • [13295] [YCQL] Disable PAGING & LIMIT for aggregate functions
  • [13555] [YSQL] Preload rd_indexlist in case of cache refresh
  • [13658] [DocDB] Allow vmodules to be updated at runtime
  • [14206] [DST] Make certain per table LB states global
  • [14417] [DocDB] Tserver Registration hazard - UUID can be blank
  • [14614] [YSQL] Cleanup after terminated connections
  • [14789] [CDCSDK] Do not send checkpoint to tablet peers for tablets in uninitialized state
  • [14913] [DocDB] Start hinted leader election only on running replicas
  • [14939] [YSQL] Allow admin to limit the number of successive failed login attempts by a role
  • [15006] [DocDB] set same error to operations with same request id when retrying a batcher
  • [15157] [CDCSDK] Ignore intents with kRowLock in GetIntentsBatch
  • [15172] [YSQL] Move the ybRunContext to be a child of TopPortalContext
  • [15433] [CDCSDK] Only delete parent tablet after a successful GetChanges call on both children
  • [15435] [YSQL] Exclude some auth methods from profile handling
  • [15516] [CDCSDK] Data loss with CDC snapshot plus streaming
  • [15842] [DocDB] Count time from transaction init for tracing thresholds
  • [16135] [YCQL] Use Metadata Cache in IsYBTableAltered
  • [16147] [CDCSDK] Checkpoint set to 0.0 from raft superblock in upgrade scenarios
  • [16147] [CDCSDK] Correctly set cdc_sdk_min_checkpoint_op_id to -1.-1 in upgrade scenarios

Bugs

YugabyteDB Anywhere

  • [PLAT-4399] Failure to mount the first Azure disk on D16s_v5 instance types
  • [PLAT-6423] [PLAT-4689] Fix YBA vulnerability
  • [PLAT-6464] [Openshift] Support bundle creation failed with group id '1000720000' is too big
  • [PLAT-6568] Fix for Minio endpoint using HTTPS
  • [PLAT-6638] [PLAT-6520] AWS fix amis with bigger root volume
  • [PLAT-6641] [PLAT-5830] Smart resize should not be supported when storage type is changed
  • [PLAT-6645] [PLAT-3915] Read replica cluster should inherit ysql enabled property from primary cluster
  • [PLAT-6734] Schedule uuid not getting populated in the backup table entry
  • [PLAT-6763] Fix Response size graph
  • [PLAT-6764] [PLAT-4914] Fix gflags and tags handling in create universe payload
  • [PLAT-6777] Root Certificate Rotation from Platform UI Fails
  • [PLAT-6781] Fix rolling restart on read-replica universes
  • [PLAT-6865] Bump up prometheus version for replicated/yugabundle installation
  • [PLAT-6966] [PLAT-6708] Fix postgres-upgrade on OpenShift clusters
  • [PLAT-7034] [PLAT-4120] Fix unnecessary info in request
  • [PLAT-7095] Back-up/restore not using S3 Bucket Host Base when IAM role is enabled
  • [PLAT-7159] YBA security vulnerability fixes
  • [PLAT-7251] UI crashing on accessing backups tab
  • [PLAT-7456] Fix DB node restart alert
  • [PLAT-7561] Retry read write check in case of failure

Database

  • [11691] [xCluster] (Fix) Do not return error when schema name not found for YSQL tables
  • [13379] [DocDB] Fix BlockBasedTable memtracker being unregistered in Tablet destructor
  • [13891] [Backfill] Do not update index permissions when table is being deleted
  • [14005] [DocDB] Fix potential duplicate WRITE_OP due to split and rpc timeout interference
  • [14184] [YSQL] Fix the partial write issue of writev/pwritev by adding looping logic
  • [14225] [DocDB] disallow advance max_safe_time_returned_with_lease_ before NoOp of current term is replicated
  • [14489] [DocDB] Fix term mismatch caused by stale consensus meta data passed to InitTabletPeer
  • [14527] [CDCSDK] Flaky ActiveAndInActiveStreamOnSameTable fix
  • [14556] [DocDB] Prevent momentary data masking during IntentAwareIterator scans in index backfill
  • [14716] [DocDB] LDB tool is crashing while trying to dump manifest file
  • [15016] [DocDB] YSQL select max() returns inconsistent result after tablet splitting
  • [15124] [CDCSDK] Fix bug in GetTabletListToPollForCDC
  • [15286] [YSQL] Fix premature return in PG_TRY block in relcache.c
  • [15386] [15230] [CDCSDK] setcheckpoint never fail with non-primary key table
  • [15418] [YSQL] Fix rolprffailedloginattempts counter behaviour
  • [15559] [CDCSDK] Prevent updating LastReplicationTime when reporting TabletSplit error
  • [15626] Move mac release builds to mac12 nodes
  • [15649] fix auto re-running cmake
  • [15698] [DocDB] Handle null placement info due to faulty restores
  • [15820] [ysql] fix ybgin index scan after DROP COLUMN
  • [15849] [DocDB] Don't hold mutex when starting TransactionStatusResolver
  • [15863] [Backfill] Fixed race in Tablet::FlushWriteIndexBatch
  • [15874] [YCQL] Fix wrong metadata version with pagination on SELECT using Index
  • [15881] [YSQL] unpushable SET clause expressions caused unnecessary index updates
  • [16034] [15933] [YSQL] Avoid picking hybrid timestamps in YSQL backends
  • [16082] [YSQL] ysql_dump cannot correctly dump primary key with INCLUDE clause for table schema
  • [16109] [DocDB] Fix deadlock with YqlPartitionsVTable
  • [16151] [DocDB] Fix incorrect iteration over modified set in CleanupAbortsTask::Run
  • [16173] [DocDB] Cluster creation is failing on Mac
  • Set up chrony correctly during AMI build process

Known issues

  • N/A

v2.14.6.1 - January 30, 2023

Build: 2.14.6.1-b4

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.6.1-b4

Bugs

YugabyteDB Anywhere

  • [PLAT-4399] Fix failure to mount the first Azure disk on D16s_v5 instance types
  • [PLAT-4689] [PLAT-6423] Library updates to address potential vulnerabilities
  • [PLAT-5876] Remove apache-commons-text dependency
  • [PLAT-6305] Upgrade Golang third-party dependencies including Azcopy
  • [PLAT-6865] [PLAT-7042] Bump up Prometheus version for replicated/yugabundle installation
  • [PLAT-7137] Add io-netty dependency back
  • [PLAT-7153] Upgrading net.sf.ehcache to latest 2.10.x

Database

  • [15698] [DocDB] Handle null placement info due to faulty restores

Known issues

  • N/A

v2.14.6.0 - January 6, 2023

Build: 2.14.6.0-b30

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.6.0-b30

Improvements

YugabyteDB Anywhere

  • [PLAT-6239] Upgrade the PostgreSQL version to 14.4 (in K8s installations)
  • [PLAT-6358] Allow defining Prometheus remote write configuration in helm chart
  • [PLAT-6537] Bump up Paramiko library version to 2.12.0

Database

  • [13555] [YSQL] Preload rd_indexlist in case of cache refresh
  • [14140] [YSQL] Import Make contrib modules' installation scripts more secure
  • [14939] [YSQL] Allow admin to limit the number of successive failed login attempts by a role
  • [15032] [CDCSDK] Alter table support with CDC snapshot

Bugs

YugabyteDB Anywhere

  • [PLAT-583] Pass scrape interval from replicated to yugaware config
  • [PLAT-3971] [PLAT-6648] Validation for inconsistent settings between main cluster and read replica
  • [PLAT-4914] [PLAT-6764] Fix GFlag and tag handling in create universe payload
  • [PLAT-5012] Add more attribute to consumed cloud resources
  • [PLAT-5324] [UI] Backup details of a scheduled backup shows backup type as On Demand
  • [PLAT-5530] Rename "Full Name" to "Customer Name"
  • [PLAT-5838] [xCluster] Create xCluster always uses /home/yugabyte as the yugabyte home directory of the DB nodes
  • [PLAT-5876] [PLAT-6491] Get rid of import org.apache.commons.text usage
  • [PLAT-6026] [PLAT-6029] Make top K query use single query + return lines in correct order
  • [PLAT-6037] Prevent GFlag deletion through the Non-Restart upgrade option
  • [PLAT-6042] Fix Kubernetes CPU usage graph
  • [PLAT-6205] Prevent commenting of trust-based auths for PostgreSQL connection
  • [PLAT-6285] Fix currentInstanceType assignment in universe form
  • [PLAT-6342] ReadOnlyClusterCreate: Task retry leads to error "node name cannot be preset"
  • [PLAT-6464] [Openshift] Support bundle creation failed with group ID '1000720000' is too big
  • [PLAT-6481] Add a check for the YSQL PostgreSQL server during the rolling restart
  • [PLAT-6519] Collect only YBDB pod's metrics from K8s nodes
  • [PLAT-6520] [PLAT-6638] AWS fix AMIs with bigger root volume
  • [PLAT-6568] Fix for Minio endpoint using HTTPS
  • [PLAT-6734] Schedule UUID not getting populated in the backup table entry
  • [PLAT-6763] Fix Response size graph
  • [PLAT-6781] Fix rolling restart on read-replica universes

Database

  • [8095] [10753] [DST] Data race in blacklistPB present in Cluster Config
  • [11691] [xCluster] Do not return error when schema name not found for YSQL tables
  • [12782] [DST] Backups - Bloated SnapshotInfoPB for backup with large number of colocated tables
  • [12802] [14668] [DST] PITR - Refresh pg_table_cache on catalog version increment
  • [13031] [DST] Backup/Restore - Throttle create table requests during ImportSnapshotMeta of restore (YCQL)
  • [13040] [YSQL] Pre-fetch pg_type table on cache refresh
  • [13102] [DocDB] Fix deadlock in DoSplitTablet.
  • [13187] [YSQL] Enable critical (nailed) indices
  • [14005] [DocDB] Fix potential duplicate WRITE_OP due to split and RPC timeout interference
  • [14027] [DST] Catalog Manager BG tasks thread creating a deep copy of schema per tablet every second
  • [14042] [DST] PITR - Fix PITR race with tablet splitting
  • [14044] [DST] LB allocates deep copies of all tablet replicas every second
  • [14184] [YSQL] Fix the partial write issue of writev/pwritev by adding looping logic
  • [14225] [DocDB] Disallow advance max_safe_time_returned_with_lease_ before NoOp of current term is replicated
  • [14298] [14668] [DST] PITR - MetaCache flips replica state from failed to not failed for hidden tablets
  • [14417] [DocDB] TServer Registration hazard - UUID can be blank
  • [14614] [YSQL] Cleanup after terminated connections
  • [14771] [YSQL] Remove unneeded NOTICE in yb_server_* functions
  • [14876] [CDCSDK] Change logging level for CDCSDK related client errors
  • [14909] [CDCSDK] Fix issue where cdcsdk_change_event_count metric is incremented even on empty batches
  • [14931] [CDCSDK] YBClient unable to find a TabletClient for child tablets which are split further
  • [15004] [CDCSDK] Data loss when streaming transaction intents for large tables
  • [15006] [DocDB] Set same error to operations with same request id when retrying a batcher
  • [15025] [YCQL] Fixing Index updating by DELETE for a deleted row
  • [15124] [CDCSDK] Fix bug in GetTabletListToPollForCDC
  • [15157] [CDCSDK] Ignore intents with kRowLock in GetIntentsBatch
  • [15230] [15386] [CDCSDK] setcheckpoint never fails with non-primary key table
  • [15418] [YSQL] Fix rolprffailedloginattempts counter behaviour
  • [15435] [YSQL] Exclude some auth methods from profile handling

Known issues

  • N/A

v2.14.5.0 - November 17, 2022

Build: 2.14.5.0-b18

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.5.0-b18

Improvements

YugabyteDB Anywhere

  • [PLAT-4413] [PLAT-5831] Add support for EBean's @Encrypted annotation

Database

  • [14431] Add a metric "ql_write_latency" to track write latency, similar to ql_read_latency

Bugs

YugabyteDB Anywhere

  • [PLAT-5639] Fix deletion of expired scheduled backups
  • [PLAT-5825] Raise exception in case node is unreachable via ssh
  • [PLAT-5964] Put database name in separate field in metric query response
  • [PLAT-5970] Resize node API uses yb.internal.allow_unsupported_instances flag
  • [PLAT-5977] Fix range bounded queries for query search bar
  • [PLAT-6028] Fix cluster average calculation for metrics with single function
  • [PLAT-6186] Map extra K8s secret to YB DB pods in user-specified namespace

Database

  • [12190] [12191] [DocDB] YBOperation.partition_key encoding should respect null value
  • [13367] [YSQL] Avoid using plain text LDAP password via env variable
  • [13465] [YSQL] Enable batch upserts when creating namespace. This provides a significant speedup for the CREATE DATABASE comand, especially in multi-region environments.
  • [13633] [DocDB] Filter snapshot details in server
  • [13808] [DocDB] Backups fail when two same-name namespaces exist, one RUNNING and another DELETED
  • [13815] [YSQL] Enhance passwordcheck with more checks
  • [14128] [DocDB] Track ql_read_latency for SQL too
  • [14201] [YCQL] Fixing partition_hash() with multiple indexes
  • [14289] [DocDB] Include subtransaction_id in value of weak intents
  • [14595] [CDCSDK] TServer crash during snapshot mode
  • [14643] [DocDB] Fix strongly typed bool constants. Metrics would not report quantiles (P95) due to this bug.
  • [14823] [CDCSDK] Change log level of line which indicates fetching next block from WAL in cdcsdk_producer
  • [14846] [CDCSDK] Add support for tablet split for newly added tables

Known issues

  • N/A

v2.14.4.0 - October 27, 2022

Build: 2.14.4.0-b26

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.4.0-b26

Improvements

YugabyteDB Anywhere

  • [PLAT-5495] Update create xCluster config target universe filter
  • [PLAT-5684] List Delete backups task at universe level
  • [PLAT-5705] Improve the command to check the available memory on the DB nodes for software upgrades

Database

  • [10774] [YCQL] Refreshing Prepares after Changes (Alter) to Table Schema. This ensures that prepared statements for YCQL get invalidated and refreshed correctly after an ALTER table command.
  • [12919] [YSQL] Populate query termination entry in pg_stat with temp_file_limit errors
  • [12953] [YSQL] Populate query termination entry in pg_stat with OOM and seg fault errors
  • [13609] [YSQL] Adding Geo-distribution helper functions
  • [13860] [YSQL] Import Free correctly LDAPMessage returned by ldap_search_s() in auth.c
  • [13970] [CDCSDK] Alter table support with CDCSDK
  • [14106] [CDCSDK] Add Java level APIs to fetch tablets from the cdc_state table
  • [14136] [YSQL] Enable yb_db_admin to ALTER/DROP/CREATE triggers like a superuser
  • [14446] [YSQL] Enable yb_db_admin to SELECT/INSERT/UPDATE/DELETE rows in tables like a superuser
  • [DB-2705] [YSQL] Introduce gflag to control local file access

Bugs

YugabyteDB Anywhere

  • [PLAT-430] [PLAT-5783] Update Target name for backup deletion customer task
  • [PLAT-4318] [PLAT-5760] [PLAT-5850] Platform UI does not allow to add more nodes in Replica Cluster of On-Premises Data center
  • [PLAT-4598] [K8s] Fix simultaneous sacling down and instance type changes in edit universe.
  • [PLAT-4654] Edit primary cluster of kubernetes can delete the primary cluster.
  • [PLAT-5136] Top K nodes/tables/databases functionality
  • [PLAT-5638] Fix failing yqslsh connectivity health checks
  • [PLAT-5771] Issues when universe is created with custom ports

Database

  • [10194] [YSQL] Dump tablespaces attached to tablegroups
  • [10735] Fix TransactionLoader race on tablet bootstrap vs tablet delete
  • [10921] [CDCSDK] Populate CDC stream cache with newly added table details
  • [10921] [CDCSDK] Support for newly added table to active stream's namespace
  • [10935] [CDCSDK] Add Active time when adding child tablets to cdc_state table
  • [10935] [CDCSDK] Do not update checkpoint if tablet split was not verified
  • [10935] [CDCSDK] Fixing bug which added tables without primary keys to CDCSDK stream metadata
  • [10935] [CDCSDK] Provide tablet split support to CDCSDK Service
  • [12898] [13317] [YSQL] [Colocation] Fixed KvStoreInfo not loading colocation map. This improves to resiliency of DDLs on tablegroups and colocated tables.
  • [13444] [DocDB] Compact child tablets of a split after a restore
  • [13480] [CDCSDK] Pack all of the changes of an update operation corresponding to a row in one CDC record
  • [13603] [YSQL] Preserving order of UNIQUE CONSTRAINT during ysql_dump. This fixes backup and restore as well as ysql_dump for specific schemas (concretely if the schema has an explicit unique constraint with ASC column).
  • [14130] [14172] [CDCSDK] Remove stale stream entries from cdc_state table, flaky test fix
  • [14144] [YSQL] Disable txid_current() function on Yugabyte code-path
  • [14204] [YCQL] Create "null" and empty JSONB object only once
  • [14288] [DocDB] Fix TSAN issue in snapshot schedules test that restart the cluster
  • [14402] [CDCSDK] Add Java tests for new table additions to existing streams
  • [14424] [YCQL] Persist WHERE predicate in IndexInfo for PartialIndexes
  • [14479] [CDCSDK] Add bool parameter to error out if UpdateCdcReplicatedIndex RPC fails on any TServer
  • [14535] [YSQL] Remove unused field backend_max_mem_bytes

Known issues

  • N/A

v2.14.3.1 - October 12, 2022

Build: 2.14.3.1-b1

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.3.1-b1

Bugs

Database

  • [13111] [DocDB] Enabling tablet splitting on an already loaded cluster with materialized views causes TServer crash (null frontiers in compaction path fix)

Known issues

  • N/A

v2.14.3.0 - October 7, 2022

Build: 2.14.3.0-b33

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.3.0-b33

New features

YugabyteDB Anywhere

N/A

Database

N/A

Improvements

YugabyteDB Anywhere

  • [PLAT-4179] [xCluster] Show schema name for tables
  • [PLAT-4339] Show schema name to YSQL tables on UI
  • [PLAT-5106] Upgrade Java dependencies for scan

Database

  • [8558] [14062] [DocDB] Added manual remote bootstrap tool
  • [12394] [YSQL] Simplify PGConn::Connect overloadings
  • [13017] [YSQL] Add metric for rejected connections due to CAC_TOOMANY
  • [13364] [YSQL] [Upgrade] Option to limit upgrade_ysql command to 1 connection
  • [13862] [YSQL] Track active new connection metric
  • [14124] [YSQL] Import Fix buffer overflow when parsing SCRAM verifiers in backend
  • [14125] [YSQL] Import Fix buffer overflow when processing SCRAM final message in libpq
  • [14139] [YQL] Import Ignore attempts to \gset into specially treated variables
  • [14141] [YSQL] Import Require the schema qualification in pg_temp.type_name(arg)
  • [14145] [YSQL] Import Fix choice of comparison operators for cross-type hashed subplans
  • [14147] [YSQL] Import Fix security checks for selectivity estimation functions with RLS

Bugs

YugabyteDB Anywhere

  • [PLAT-3951] Add RocksDB compaction tasks metric
  • [PLAT-4745] [UI] SSH connection to node failing from platform due to special characters in provider name
  • [PLAT-4932] [PLAT-5312] [PLAT-5026] Stop exposing container ports on host interface by default
  • [PLAT-5226] Make YBClient use netty 4 to fix TLSv1.3 issue
  • [PLAT-5343] Make default alert notification templates more reasonable
  • [PLAT-5373] For k8s universes Yedis port is open without AUTH even if Yedis is NOT enabled.
  • [PLAT-5464] [PLAT-5301] On-prem Node Pre Check API returns None value
  • [PLAT-5552] Added ssh2_key check in try/catch block
  • [PLAT-5619] [xCluster] Do not use sudo user for transferring source root cert

Database

  • [816] [1588] [DocDB] always sync RocksDB files
  • [11691] [xCluster] Also check for schema name match when setting up xCluster replication
  • [12816] [YSQL] Restore of backup created with pg_hint_plan enabled fails
  • [12880] [DocDB] Handle mix of matched and mismatched schema versions
  • [13027] [DocDB] Splitting: Incorrect split_encoded_key re-compute in case of child tablet already registered
  • [13389] [DocDB] Fix initial leader election at table creation time
  • [13611] [DocDB] Disable enable_lease_revocation by default
  • [13708] [YSQL] Don't set ListTabletServersResponsePB::millis_since_heartbeat in case of no heartbeat
  • [13731] [13887] [YCQL] Fix invalid memory access in Selectivity::Analyze()
  • [13796] [YSQL] Fixed template1 connection crashing when using older initdb
  • [13798] [DocDB] Re-run sys catalog snapshot after failover during initdb
  • [13805] [YCQL] Fix in GetTableSchema to return correct schema version
  • [13861] [YSQL] Upgrade PostgreSQL JDBC to 42.5.0
  • [13880] [YSQL] Exclude orphaned pg_temp_1xxx tables from snapshot creation, clean up orphaned materialized views
  • [14040] [YSQL] fix EXEC_STATUS_SWITCH_CASE macro
  • [14129] [YSQL] Enable yb_db_admin to alter system triggers
  • [14155] [YSQL] Fix create database time out regression

Known issues

  • N/A

v2.14.2.1 - October 21, 2022

Build: 2.14.2.1-b1

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.2.1-b1

Bugs

Database

  • [14424] [YCQL] Persist where predicate in IndexInfo for PartialIndexes

Known issues

  • N/A

v2.14.2.0 - September 14, 2022

Build: 2.14.2.0-b25

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.2.0-b25

New features

YugabyteDB Anywhere

N/A

Database

N/A

Improvements

YugabyteDB Anywhere

  • [PLAT-4063] Added ulimit preflight check
  • [PLAT-4960] Initial metric collection profiles
  • [PLAT-5014] Adding pricingKnown to instance price API
  • [PLAT-5265] Add additional AZs in AWS ap-northeast-2 region

Database

  • [13617] Import In extensions, don't replace objects not belonging to the extension

Bugs

YugabyteDB Anywhere

  • [PLAT-511] Adding ARM node exporter to yugabundle support packages
  • [PLAT-4458] [UI] Remove extra space in Edit universe appearing below Use AWS Time Sync
  • [PLAT-4494] [xCluster] Unable to delete a failed XClusterConfig
  • [PLAT-4754] [DB-2913] Support expanding multi-AZ universe while a AZ is down
  • [PLAT-4815] Record old and new gflag values in audit payload for GFlag Upgrade tasks
  • [PLAT-4842] Set initial value of ybSoftwareVersion to current version in upgrade form
  • [PLAT-4938] Add null checks for removed/released instances
  • [PLAT-4954] Added PreValidator Hook for runtime config value
  • [PLAT-4960] Remove priority_regex from prometheus config
  • [PLAT-4979] Nodes unreachable after switching HA active/standby
  • [PLAT-5035] Fix missing scheduleUUID
  • [PLAT-5087] Collect PostgreSQL logs in support bundle
  • [PLAT-5195] [PLAT-5211] Master flag lists shows incorrect default value, ap-northeast-2 (Korea) region is not available in YB Anywhere Platform
  • [PLAT-5207] Fix connectivity issue between YBA and Master with latest JVM version
  • [PLAT-5225] Change pricing design for Azure
  • [PLAT-5235] SQL admin password validation exception due to missing database

Database

  • [7216] [12490] [12692] [YSQL] Avoid undefined behavior on postgres shutdown
  • [11769] [YCQL] Fixed TS crash due to NULL PTBindVar::hash_col_
  • [12003] [xCluster] Allow dropping of YSQL tables under replication
  • [12184] [YSQL] Flush and wait for buffered ops before executing statements with non-transactional side-effects in functions and procedures
  • [12480] [DocDB] Check master address in either rpc_bind or broadcast addresses
  • [12539] [DocDB] prometheus-metrics consumes a lot of CPU on servers with a lot of tablets
  • [12795] [DocDB] Refactor compaction metrics for priority thread pool
  • [13025] [DocDB] Tablet is trying to split infinitely
  • [13250] [YSQL] ysql_dump should set up a sequence's current value
  • [13342] [DocDB] Fix issue which fail to restart tserver if crash after tablet split applied
  • [13428] [YSQL] Disable lazy evaluation in a SQL fucntion for Read Committed isolation
  • [13434] [xCluster] Dropping a YSQL table under replication should delete its CDC streams
  • [13504] [YSQL] Count OID query failing for pg_type table
  • [13525] [YSQL] [Upgrade] Upgrade doesn't work if YSQL was initialized by a manual initdb run
  • [13558] [YSQL] Override mem_tracker_tcmalloc_gc_release_bytes for PG backends
  • [13610] [DocDB] Don't replicate write batch with size exceeds rpc message limit, fail the batch instead
  • [13653] [CDCSDK] Deleting stream IDs leads to stale entries in the cdc_state table causing tserver crash
  • [13659] [DocDB] Fix Prepare for the failed transaction
  • [13670] [DocDB] Ensure that we use tserver_tcmalloc_max_total_thread_cache_bytes if it is set
  • [13693] [CDCSDK] Add last_active_time to cdc_state table
  • [13769] [YSQL] Put OverrideTcmallocGcThresholdForPg under TCMALLOC_ENABLED
  • [13770] [CDCSDK] Intents are getting GCed after Tablet LEADER changes
  • [13799] [YSQL] Adjust backward scan factor
  • [CDCSDK] Retry for yb-client for a CDC Error and bootstrap option for CDC console client

Known issues

  • N/A

v2.14.1.0 - August 9, 2022

Build: 2.14.1.0-b36

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.1.0-b36

New features

YugabyteDB Anywhere

N/A

Database

N/A

Improvements

YugabyteDB Anywhere

  • [PLAT-3560] [PLAT-3950] Add ability to run platform as a non-root user
  • [PLAT-4121] Add UI for managing HA peer certificates
  • [PLAT-4200] Added SSH-2 Keys support in the platform
  • [PLAT-4310] Add GET endpoint to return accepted components
  • [PLAT-4657] Add advanced master metrics
  • [PLAT-4856] Add support for editing User Tags for GCP
  • [PLAT-4857] [LDAP] UI support for search+bind
  • [PLAT-4890] Add search and bind functionality for Platform LDAP

Database

  • [13265] [DocDB] Add option for server level aggregation
  • [13346] [DocDB] Wait for parent to be hidden/deleted before splitting tablet
  • [13441] [CDCSDK] Modifications to send namespace ID as a part of GetDBStreamInfoResponse in Java

Bugs

YugabyteDB Anywhere

  • [PLAT-2086] [Platform] Allow retention of platform logs on replicated after upgrades
  • [PLAT-2175] Change manual install script execution step for on-premise
  • [PLAT-2540] [PLAT-4819] [PLAT-4820] Fix for issue that HA certs are ignored after restart
  • [PLAT-2609] [Backups] Handle case when TS Web UI is not available
  • [PLAT-3156] [PLAT-4534] [PLAT-4619] Fetch slow query and reset query across all universe nodes with SSH instead of JDBC
  • [PLAT-3994] Non-Credential Values Written to GCP GOOGLE_APPLICATION_CREDENTIALS file.
  • [PLAT-4187] Fix stdDevTime formula
  • [PLAT-4304] [PLAT-4894] Fix and enable support bundle creation on k8s
  • [PLAT-4304] Add support bundle for k8s universes
  • [PLAT-4381] [PLAT-4790] Removing pricing requirement for GCP
  • [PLAT-4460] Edit universe tags is broken
  • [PLAT-4484] [PLAT-4486] Allow resuming paused universes with EAR enabled.
  • [PLAT-4498] Log an error and skip placement regions not in cluster
  • [PLAT-4586] Correctly clean up yb_univ_health_status metric on universe pause/removal/configuration changes
  • [PLAT-4602] Add scroll bar in case health check output does not fit
  • [PLAT-4650] [PLAT-4658] Rotating CA through UI results in loss of GFlags and sets unexpected flags
  • [PLAT-4657] Add missing metrics to priority regex, few metric dashboard fixes
  • [PLAT-4694] [PLAT-4891] Mask password field on replicated settings UI
  • [PLAT-4696] Temporary Fix for IAM backups deletion and validation
  • [PLAT-4810] Enable refresh button on live query tab
  • [PLAT-4847] Show cost appropriately when pricing is unknown for a running Universe
  • [PLAT-4906] Update LDAP DN for search and bind only if not empty
  • [PLAT-4936] Restore Failing with ssh2_enabled

Database

  • [12587] [YSQL] Add yb_make_next_ddl_statement_nonbreaking
  • [12613] [CDCSDK] setCheckpoint fail when number of TServers are more than tablet split count
  • [12627] [DocDB] Fix collecting table metrics by metrics snapshotter
  • [12636] [DocDB] Reduce verbosity of GetNamespaceInfo log
  • [12738] Add static libraries based on shared library dependencies during LTO
  • [12858] [YSQL] Implement GC for Postgres backend
  • [12909] [YCQL] [Backups] Support YCQL user defined types in backups
  • [12978] [YSQL] Import Fix bogus casting in BlockIdGetBlockNumber()
  • [13022] [DST] [PITR] Allow ability to restore to different points of time in the past when tablet splitting was ongoing
  • [13060] [YCQL] Disallow IN clause combined with ORDER BY
  • [13116] [DocDB] Changes the default for enable_multi_raft_heartbeat_batch gflag to false
  • [13127] [DocDB] Add field to control waiting for parent deletion in IsTabletSplittingComplete.
  • [13180] [CDCSDK] Fix for EnumCache not getting updated if new enum types are added to the same namespace
  • [13220] [12937] [YSQL] Fix incorrect max memory calculation
  • [13222] [YSQL] Ensure that the list of aborted sub-txns is not spuriously removed by asynchronous heartbeats
  • [13245] [YSQL] Import Don't use_physical_tlist for an IOS with non-returnable columns.
  • [13280] [DocDB] Fix behavior when outstanding_tablet_split_limit_per_tserver = 0.
  • [13294] [DocDB] Restore some fields from tablet metadata after restoring a snapshot
  • [13341] [DocDB] PITR could restore DB to state that contains records with garbage collected schema packing
  • [13361] Fix backups for YSQL tables only present in DocDB

Known issues

  • N/A

v2.14.0.0 - July 14, 2022

Build: 2.14.0.0-b94

Third-party licenses: YugabyteDB, YugabyteDB Anywhere

Downloads

Docker

docker pull yugabytedb/yugabyte:2.14.0.0-b94

New features

YugabyteDB Anywhere

  • [PLAT-2653] [PLAT-4108] YugabyteDB Anywhere support for ARM and AWS/Graviton is now GA
  • [PLAT-3709] Add under-replicated masters alert
  • [PLAT-3725] Add k8s_parent.py entrypoint script, and use to start DB processes
  • [PLAT-4085] Remove Backup V2 Feature Flag
  • [PLAT-4140] Remove YB beta for NTP
  • [PLAT-4143] Backport NTP Changes to 2.14
  • [PLAT-4293] OIDC, including dual login, is now available
  • [PLAT-4361] HashiCorp Certificates are now GA, and no longer beta

Database

  • [12737] [YSQL] hide superuser connections under normal user
  • [13176] [DocDB] Generate GFlags metadata XML at build time

Improvements

YugabyteDB Anywhere

  • [PLAT-2836] Removing dateutil requirement for health checks
  • [PLAT-2934] Add user friendly formatting for lag metric
  • [PLAT-2967] Add YSQL connection metrics to the platform metrics dashboard
  • [PLAT-3194] Redact the default alerts email to prevent phishing attacks
  • [PLAT-3749] Add Prometheus expression link to alert details
  • [PLAT-3791] Ensure errors in ansible/python and background Java threads are propagated to the task and API error message correctly
  • [PLAT-3819] Move health checks to the node + move custom node metrics collection to node health script
  • [PLAT-3932] Enable Request ID logging for platform
  • [PLAT-4015] Set up NTP for old providers
  • [PLAT-4059] [Backup v2] Can we have client side check for "Minimum schedule duration is 1 hour"
  • [PLAT-4079] Universe Creation UI Changes for NTP Enhancements
  • [PLAT-4151] Update universe status classification logic
  • [PLAT-4171] feat Display releases according to region architecture
  • [PLAT-4219] feat Display supported releases by provider in upgrade software form
  • [PLAT-4222] Fetching releases by provider endpoint
  • [PLAT-4267] Update most used GFlags list
  • [PLAT-4304] Enable support bundle for on-prem universes
  • [PLAT-4357] [OIDC] remove banner that says platform restart required
  • [PLAT-4364] Allow custom AMI image upgrade in addition to base image when yb.upgrade.vmImage is set

Database

  • [10186] [xCluster] Locality aware mapping for tables with different tablet counts in xCluster
  • [12327] [YSQL] Workaround for FK constraint violation in case of dynamic table split
  • [12548] [YSQL] Make permission error strings more informative
  • [12710] [CDCSDK] CDC Upgrade Path
  • [13146] [CDCSDK] Add CDCSDK checkpoint check before removing transactions from CleanupAbortsTask

Bugs

YugabyteDB Anywhere

  • [PLAT-2077] Fix node count inconsistency
  • [PLAT-3217] [UI] Default interval of rolling restart delay is not loaded for resize nodes
  • [PLAT-3460] Fix under-replicated and leaderless tablet alerts
  • [PLAT-3625] Filter target universes dropdown to contain only ready/good universes
  • [PLAT-3791] Ensure errors in ansible/python and background Java threads are propagated to the task and API error message correctly
  • [PLAT-3953] Fix backup-restore for universes with read replicas
  • [PLAT-3982] update log line start pattern for filtering logs
  • [PLAT-4029] Fix schedule response to display cron expression.
  • [PLAT-4073] Hide Password field while adding users when OIDC is enabled
  • [PLAT-4078] Allow the creation of providers with showSetupChrony false via the API
  • [PLAT-4115] Install s3cmd correctly on CIS hardened image
  • [PLAT-4117] "Use TimeSync" toggle does not disappear for new providers
  • [PLAT-4146] [PLAT-4144] [UI] [Backup] We don't list YCQL keyspace if it has a same name as of its YSQL counter part.
  • [PLAT-4147] Disable backup button when backup is disabled
  • [PLAT-4148] [PLAT-4303] [xCluster] Current Lag is 0 even there is replication is in progress
  • [PLAT-4154] Avoid multiple delete expired backup tasks
  • [PLAT-4182] Fix health checks on Ubuntu
  • [PLAT-4189] [xCluster] Get replication lag using table id
  • [PLAT-4194] [HA] Disable version check by default
  • [PLAT-4196]The Advanced Restore button is not visible if there are no backups in the universe.
  • [PLAT-4201] [Backup]Automatically calculate parallelism for restore modal
  • [PLAT-4214] Re-upload health check script to the node after VM image upgrade + upload it on each check to k8s pod.
  • [PLAT-4221] Set highest possible value to max_tables_metrics_breakdowns to make sure all per-table metric values are returned if metric name matches regex
  • [PLAT-4231] [PLAT-4314] Allow user to log in with credentials even if SSO is enabled
  • [PLAT-4234] Rollback old behaviour for resource metrics to show avg over time and nodes
  • [PLAT-4256] Add default bridge network gateway to trusted proxies in case of replicated
  • [PLAT-4266] Fetch runtime configurations with current universe scope in universe details page
  • [PLAT-4289] Custom date range setting issue
  • [PLAT-4329] Allow read-only user to read backups and schedules
  • [PLAT-4331] [xCluster] Table's Current Lag is incorrect value
  • [PLAT-4332] [OIDC] SuperAdmin role gets revoked when we log in as superAdmin when OIDC is enabled
  • [PLAT-4351] Allow backup admin users to use new backup APIs
  • [PLAT-4357] [OIDC] remove banner that says platform restart required
  • [PLAT-4410] Fix software upgrade task progress
  • [PLAT-4415] Fix health checks for arm builds
  • [PLAT-4421] Change lag metrics label to "Unreachable" when metrics array is empty
  • [PLAT-4428] Fix intermittent backup deletion failure in GCS
  • [PLAT-4495] Fix CQL Live query /rpcz response processing for missing connection_details + IPv6 hostname
  • [PLAT-4500] Fix ybp_universe_encryption_key_expiry_days metric to match Prometheus naming convention
  • [PLAT-4528] [Backup/Restore] Create backup is failing with YCQL table and index
  • [PLAT-4562] [xCluster] Lag value is overlapping on other text.
  • [PLAT-4581] Restore is not happening if user uses the existing Keyspace/Database name.
  • [UI] Backup/Restore - Delete modal shown under sidebar
  • [UI] If a Universe does not have any tables ,the "backup now" button should be disabled API for tablespaces creation is failing with Invalid Token Fix issue with saving provided private key

Database

  • [8869] [12584] [YSQL] fix TRUNCATE involving indexes + tablegroups
  • [9588] [10039] [DST] Ignore intents from aborted sub-transactions of other transactions during conflict resolution
  • [10154] [DocDB] fixed intensive intents cleanup rescheduling in case of concurrent compaction
  • [10333] [DocDB] Use table's HasOutstandingSplits function for splitting + PITR.
  • [10333] [DocDB] Wait for splitting to be done before backfilling.
  • [10360] [DST] PITR - Disable Tablet Splitting during PITR restores
  • [10840] [YSQL] Use column statistics to calculate selectivity to enable accurate row estimations while choosing query plans
  • [11030] [xCluster] [TabletSplitting] Handle master failovers during ProcessSplitTabletResult
  • [11064] [DocDB] Improve split performance by adding TServer flag to prioritize automatic compactions
  • [11132] [DocDB] Add materialized view table type, store materialized view's PG table OID in memory, fix REFRESH failure handling.
  • [11343] [YSQL] Avoid wait for RPC response on process termination
  • [11460] [DocDB] Add per-tserver split limits
  • [11461] [DocDB] Allow splitting for SST files that have only 1 block
  • [11617] [YSQL] Prefer custom plans for prepared statements using partitioned tables if partition pruning is possible
  • [11641] [DocDB] Make Follower reads work with Read Committed isolation.
  • [11642] [DocDB] Fix transaction manager shutdown
  • [11715] [YSQL] Fix bug in statement_timeout rollback
  • [11738] [DST] PITR - Block concurrent DDLs during restoration
  • [11846] [DocDB] Simple check to fail PITRs that would roll back a YSQL upgrade
  • [11849] [YSQL] Explicitly indicate aggregate push down in EXPLAIN output
  • [11928] [YSQL] Import Report progress of COPY commands
  • [12004] [YSQL] Explicit locking statement doesn't create a transaction when using READ COMMITTED in YSQL that maps to REPEATABLE READ
  • [12077] [YSQL] Add PG max memory stat to EXPLAIN output
  • [12207] [YSQL] Expand yb_db_admin function permissions
  • [12258] [YSQL] Keep ybctid consistent across all databases for shared insert
  • [12293] [YSQL] Re-enable new copy command default
  • [12386] [xCluster] Fix sanitizer errors in twodc_output_client.cc
  • [12460] [YSQL] Populate idx_scan in pg_stat_user_indexes
  • [12476] [xCluster] [TabletSplitting] Handle parent tablet deletion for xCluster related tablets
  • [12478] [YSQL] Add support for copy options DISABLE_FK_CHECK
  • [12478] [YSQL] Add support for REPLACE copy option
  • [12478] [YSQL] Add support for skipping rows in copy
  • [12478] [YSQL] Skip report error on skipped rows
  • [12482] [DocDB] Adding tserver-side tablet split metrics and logging
  • [12483] [YSQL] Allow yb_extension to create pg_trgm and PGAudit extensions.
  • [12483] [YSQL] Allow yb_db_admin to run REASSIGN OWNED BY
  • [12484] [12487] [DST] PITR - Disable PITR on tablegroups
  • [12496] [YSQL] Fix DROP MATVIEW failure after snapshot schedule creation
  • [12508] [DST] PITR - Disable drop tablespace on clusters with PITR Schedules
  • [12509] [YSQL] Enable yb_fdw to use orafce extension
  • [12526] [CDCSDK] Delete cdc_state table metadata on dropping a table
  • [12537] [DocDB] Do not retry RPC on Incomplete error.
  • [12605] [YSQL] Prevent non-superusers from reassigning objects owned by superusers.
  • [12611] [YSQL] Enable yb_db_admin to set session authorization
  • [12616] [DocDB] Change drive_fault metrics to use MetricEntity
  • [12624] [YSQL] Set read time on YSQL side in case of non-txn writes (yb_disable_transactional_writes = 1)
  • [12625] [YSQL] increase backfill client timeout to 1d
  • [12661] [Geo] Put use of local transaction tables for YCQL behind GFlag
  • [12673] [DST] PITR - Fix Segmentation fault on replaying snapshot op during tablet bootstrap
  • [12691] [YSQL] Initialize disable_fk_check field
  • [12729] [DocDB] Fix crash while calculating delete marker retention time
  • [12730] [CDCSDK] Set OpId of APPLY Record in the RunningTransaction during tablet bootstrap
  • [12740] [YSQL] Don't store index using per-tuple memory context for temp table
  • [12741] [YSQL] Reset pg_hint_plan state after ExecutorEnd
  • [12744] [CDCSDK] Enum support
  • [12753] backport fix for GitHub issue #12619 into 2.14 branch.
  • [12762] [YCQL] TServer FATAL when running Cassandra stress test
  • [12767] [YSQL] Send list of aborted sub transactions to the status tablet during a savepoint rollback + 2 bug fixes
  • [12775] [YSQL] Show transaction priority of the active transaction in current session
  • [12779] [DocDB] Change enable_automatic_tablet_splitting to false by default.
  • [12783] [YSQL] Set the default temp_file_limit to 1GB
  • [12784] [YSQL] Hard kill all hung PG processes during shutdown
  • [12790] [CDCSDK] Handles errors in enum UDT columns
  • [12799] [YSQL] Import Make relation-enumerating operations be security-restricted operations.
  • [12801] [DocDB] Change default value of automatic_compaction_extra_priority to 0.
  • [12804] [YSQL] Fix read time used by PrecastRequestSender to read rows written by buffered operations
  • [12810] [DocDB] Apply prioritize_tasks_by_disk gflag to entire group_no codepath in PriorityThreadPool
  • [12810] [DocDB] Disable automatic_compaction_extra_priority when auto splits are disabled.
  • [12810] [DocDB] Disable disk prioritization for compactions/flushes by default to improve performance
  • [12814] [CDCSDK] Backfill primitive type in tablet metadata for UDTs on upgrade
  • [12821] [YSQL] Add logic to support status column which indicates the status of the copy command
  • [12822] Handle nullptr consensus round in transaction coordinator's replication handler
  • [12843] [YSQL] Fix fclose bug when fopen does not succeed while changing oom_score_adj
  • [12845] [YSQL] Deactivate TOAST compression for YB relation and change index tuple encoding
  • [12972] [YSQL] Avoid erroring out on rolling back sub transactions of an expired transaction
  • [13006] [xCluster] Fix removing table from alter_universe_replication blocking drop table
  • [13029] [YSQL] Fix corner case of bare Param node pushdown
  • [13042] [DocDB] fixed clearing pending config for aborted CONFIG_CHANGE_OP
  • [13048] [DST] Flush all YSQL system tables metadata together during CREATE DATABASE
  • [13101] [YSQL] Change default setting of flag that restarts postmaster when Postgres backend dies
  • [13138] [YSQL] Upgrade failure to master build 88 and greater or to 2.15.0.1-b1
  • [13195] [DocDB] Increased retryable_request_timeout_secs to 660 seconds
  • [13215] [CDCSDK] Handling TServer crash for accessing uninitialised log object pointer for SetCDCCheckpoint API.
  • [CDCSDK] [13153] wal_retention_secs Flag reset to 0 after Tablet Bootstrap, causing WAL GCed
  • [CDCSDK] Add a retry in GetChangesForCDCSDK, where transaction's batch size is large
  • [CDCSDK] Data inconsistency in CDC after restart of TServer
  • [DocDB] Adds the task_ignore_disk_priority gflag to disable disk prioritization.
  • [xCluster] Fix Replication for For Transactions Spanning Multiple Write Batches

Known issues

  • N/A