Arafat Khan created HDDS-14079:
----------------------------------
Summary: Fix Recon startup failures during schema upgrades due to
race conditions
Key: HDDS-14079
URL: https://issues.apache.org/jira/browse/HDDS-14079
Project: Apache Ozone
Issue Type: Improvement
Reporter: Arafat Khan
*Problem:*
During cluster upgrades, Recon fails to start due to race conditions between
schema upgrades and service initialization. Multiple components attempt to
access database columns before the upgrade framework adds them, causing
SQLSyntaxErrorException and startup failures.
*Root Causes:*
1. ReconTaskStatusUpdaterManager reads RECON_TASK_STATUS table during Guice
injection, before schema upgrades run
2. ReconTaskStatusMetrics registers and queries the table during start(),
before upgrades complete
3. NSSummaryAggregatedTotalsUpgrade requires currentOMMetadataManager which is
not set during startup
*Impact:*
- Recon starts successfully during upgrades without schema-related crashes
- Upgrade handlers can access required dependencies
- No functional changes to steady-state behaviour
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
