Summary

In components/job-orchestration/job_orchestration/scheduler/query/query_scheduler.py, the loop that processes pending query jobs performs a bare msgpack.unpackb(job["job_config"]) without any error handling. A corrupted or otherwise undeserializable job config could raise an exception and halt processing of all subsequent pending query jobs.

Suggested Fix

Wrap the msgpack.unpackb(job["job_config"]) call in a try/except block (similar to the guard added for the compression scheduler in PR #2178). On failure:

• Log the exception with logger.exception, including the job ID.
• Update the query job's status to FAILED with a clear status_msg indicating the config is corrupted or undeserializable.
• continue the loop so subsequent query jobs are still processed.

Context

• PR fix(package)!: Increase database column sizes for compression & query job configs; Handle corrupted job configs gracefully (fixes #2151). #2178 added this resilience pattern to the compression scheduler (around brotli.decompress + msgpack.unpackb of clp_config).
• The query scheduler's job_config is plain msgpack (no brotli compression), written by the Rust API server via rmp_serde, so it is not affected by the VARBINARY truncation issue that triggered PR fix(package)!: Increase database column sizes for compression & query job configs; Handle corrupted job configs gracefully (fixes #2151). #2178. The MEDIUMBLOB schema change in PR fix(package)!: Increase database column sizes for compression & query job configs; Handle corrupted job configs gracefully (fixes #2151). #2178 further prevents future truncation for that column.
• This issue was raised during the review of PR fix(package)!: Increase database column sizes for compression & query job configs; Handle corrupted job configs gracefully (fixes #2151). #2178: fix(package)!: Increase database column sizes for compression & query job configs; Handle corrupted job configs gracefully (fixes #2151). #2178 (comment)
• Requested by @sitaowang1998.