Version 2.2.6
Released on 2018/01/17.
Note
If you are upgrading a cluster, you must be running CrateDB Version 1.1.3 or higher before you upgrade to 2.2.6.
If you want to perform a rolling upgrade, your current CrateDB version number must be Version 2.2.0. If you want to upgrade from a version prior to this, the upgrade will introduce all of the breaking changes listed for Version 2.2.0, and will require a full restart upgrade.
Warning
Before upgrading, you should back up your data.
Table of contents
Changelog
Fixes
Fixed a race condition that caused
DELETE
orUPDATE
statements to result in an error if a shard was being relocated or in aRECOVERY
state.Improved error message when using the
DISTINCT
clause on unsupported data types. Also improved documentation to note the data type restriction.Fix log verbosity by only logging the HTTP SSL enabled/disabled message once at startup.
Fixed an issue that could cause
INSERT INTO
statements into partitioned tables to not work correctly. This only occurred if aquery
instead of theVALUES
clause was used.Fixed the evaluation of JavaScript user-defined functions that caused CrateDB to crash because of an unhandled assertion when providing the UDF with EcmaScript 6 arrow function syntax (
var f = (x) => x;
).Fixed an issue where batch operations executed using the PosgreSQL wire protocol returned 0 as row count, even though the actual row count was different.
Fixed a bug which could cause job entries in
sys.jobs
not being removed when a connection error occurred while sending the results of the job execution to the client.Fixed an issue that caused incorrect results for queries with joins on more than 2 tables with implicit join conditions using the
ON
clause and whereLIMIT
is applied. E.g.SELECT * from t1
INNER JOIN t2 on t1.id = t2.id
INNER JOIN t3 on t3.id = t2.id
LIMIT 100
Fixed an issue that could cause an
ALTER TABLE
statement to fail with an exception on partitioned tables created with CrateDB < 1.2.Enforce validation of column constraints for
INSERT
statements with a subquery as source and in combination withON DUPLICATE KEY UPDATE
.Fixed an issue that caused an error to be thrown when using
GROUP BY
orDISTINCT
on a column of typeIP
in case there are rows with null values for that column.