8.1.15

Bug Fixes

Misc

• Release Dry fails because of unrecognized argument (#13715)
• ZeebePartitionHealth repeatedly reports change of health status (#13650)
• Gateway readiness fails when tls enabled (#13431)
• Jobs not timing out in v8.1.12 (#13190)
• Error message truncation may be too strict (#13046)
• ExceededBatchRecordSizeException: Can't append entry (#12007)
• Truncate errorMessage to prevent exceeding maxMessageSize (#11460)

Maintenance

• Snyk monitoring wrong Docker images for stable releases (#13645)
• Investigate usage of client in gateway (#11799)

Merged Pull Requests

• fix(release): read releaseBranch from output on reusable workflow call (#13737)
• Fix backport automerge (#13680)
• Tag original PR author in backport PR (#13675)
• Auto merge backport PRs (#13667)
• test(engine): add TestWatcher to ProcessEngineMetricsTest (#13659)
• fix(broker): report health on change only (#13655)
• fix(.github): monitor correct image on release (#13646)
• ci(release): use new stable release runner pool (#13628)
• Prevent exceeding max message size on job batch activate commands (#13624)
• fix(release): increased maven & Go release timeout (#13623)
• fix(broker): ensure that failure in transition to INACTIVE isn't ignored (#13608)
• Resolve flaky EmbeddedSubProcessConcurrencyTest (#13498)
• Reset reschedule flag correctly on JobTimeoutTrigger (#13441)
• Increase error message truncation limit (#13414)
• refactor: do not upload sarif on release dry run (#13411)
• Temporarily disable TCC for integration tests (#13404)
• ci(go): fix go lint cache issue (#13308)
• Log node member ID via the scheduler name as part of actor logging context (#13298)
• fix: change shouldNotPreallocateSegmentFiles() test when running on M… (#12816)
• Limit error message size in job and incident records (#12066)