What's New in v20.2.17

Warning:
Cockroach Labs will stop providing Assistance Support for this version on May 10, 2022. Prior to that date, upgrade to a more recent version to continue receiving support. For more details, see the Release Support Policy.

October 11, 2021

This page lists additions and changes in v20.2.17 since v20.2.16.

Get future release notes emailed to you:

Downloads

Warning:
The CockroachDB executable for Windows is experimental and not suitable for production deployments. Windows 8 or higher is required.

Docker image

icon/buttons/copy
$ docker pull cockroachdb/cockroach:v20.2.17

DB Console changes

Bug fixes

  • Fixed a bug that caused internal errors with set operations (e.g., UNION) and columns with tuple types that contained constant NULL values. This bug was introduced in version 20.2.0. #69270
  • Fixed a bug causing table stats collection issued via an EXPLAIN ANALYZE statement or via a CREATE STATISTICS statement to run into a "flow: memory budget exceeded" error if the statement was executed without specifying the AS OF SYSTEM TIME option. #69590
  • Fixed a bug causing a prepared statement to incorrectly reuse the query plan of a different prepared statement that had similar, but not identical type hints. #67687
  • Fixed a bug causing other nodes in the cluster to fail to connect to a restarted node due to their circuit breakers not resetting after a temporary node outage. This would manifest in the logs in messages of the form "unable to dial nXX: breaker open", where XX is the ID of the restarted node. Note that such errors are expected for nodes that are truly unreachable, and could still occur around the time of the restart, but for no longer than a few seconds. #70353
  • RESTORE now correctly ignores dropped databases that may have been included in cluster backups with revision history. #69795
  • DNS unavailability during range 1 leaseholder loss no longer causes significant latency increases for queries and other operations. #70133
  • Addressed a memory corruption issue in Pebble when running with CGO_ENABLED=0. #70995
  • Addressed an issue in Pebble where a key can be dropped from an LSM snapshot if the key was deleted by a range tombstone after the snapshot was acquired. #70995
  • Addressed an issue in Pebble where a crash between the creation of the WAL and the MANIFEST could leave the second most WAL with an unclean tail that would be considered corruption on a subsequent Open. #70995

Contributors

This release includes 11 merged PRs by 11 authors.

YesYes NoNo