AWS redshift: Redshift patch notes: resize and Lambda UDF reliability fixes
Summary
Adds release note entries describing fixes for online resize operations stalling in read-only mode and for Lambda UDF queries exhausting internal process resources causing unexpected cluster restarts.
Security assessment
The change documents availability/reliability bug fixes (resize stalls, resource exhaustion from Lambda UDFs). While resource exhaustion could be abused for DoS, the note frames it as a reliability fix with no stated vulnerability, CVE, or security control, so it is not security documentation.
Evidence
* Fixed an issue where queries that call AWS Lambda user-defined functions (UDFs) could exhaust internal process resources and cause an unexpected cluster restart.
Diff
diff --git a/redshift/latest/mgmt/cluster-versions.md b/redshift/latest/mgmt/cluster-versions.md index 6900fbe..1bf0f7e 100644 --- a//redshift/latest/mgmt/cluster-versions.md +++ b//redshift/latest/mgmt/cluster-versions.md @@ -151,0 +152,2 @@ Cluster versions in this patch: + * Changed how tables load after a cluster restart so that commits no longer wait for all tables to load. The first commit can complete up to 15 minutes sooner, with minimal impact on query performance. This improvement is currently enabled for a limited set of Amazon Redshift Serverless workgroups and will be extended in future patches. + @@ -167,0 +170,4 @@ Cluster versions in this patch: + * Provided a fix to improve reliability of online resize operations. A failure where the transition to read-only mode could stall and never complete is fixed, so these operations no longer get stuck partway through and complete successfully without manual intervention. + + * Fixed an issue where queries that call AWS Lambda user-defined functions (UDFs) could exhaust internal process resources and cause an unexpected cluster restart. +