Good afternoon everyone! To resolve the problem, I had to delete the POD in Kubernetes that was generating the error logs, and then I used the "Run Pending Data Migration" option to manually restart the "v8" migration process. During the initial processing, we had this LOG which seems to be related to the AsyncJob feature:
{"level":"INFO","timestamp":"2026-06-18T14:27:58.488Z","msg":"PostDeployMigrationQueue worker: failed","jobId":"1","jobTimestamp":1781785626195,"processedOn":1781791687249,"finishedOn":1781792878478,"queuedDurationMs":6061054,"executionDurationMs":1191229,"totalDurationMs":7252283,"attemptsMade":1,"attemptsStarted":2,"asyncJob":"AsyncJob/019e03fc-40f3-7491-97fd-85dcd6316914","jobType":"reindex","prev":"active","error":"Constraint inv-1 not met: A parameter must have one and only one of (value, resource, part) ({\"fhirpath\":\"(part.exists() and value.empty() and resource.empty()) or (part.empty() and (value.exists() xor resource.exists()))\"}) (AsyncJob.output.parameter[4].part[2]); Invalid additional property \"valueString\" (AsyncJob.output.parameter[4].part[2].valueString)","stack":"Error: Constraint inv-1 not met: A parameter must have one and only one of (value, resource, part) ({\"fhirpath\":\"(part.exists() and value.empty() and resource.empty()) or (part.empty() and (value.exists() xor resource.exists()))\"}) (AsyncJob.output.parameter[4].part[2]); Invalid additional property \"valueString\" (AsyncJob.output.parameter[4].part[2].valueString)\n at er.validate (/usr/src/medplum/packages/core/dist/cjs/index.cjs:1:56030)\n at Xs (/usr/src/medplum/packages/core/dist/cjs/index.cjs:1:55505)\n at Repository.validateResourceStrictly (/usr/src/medplum/packages/server/dist/fhir/repo.js:559:52)\n at Repository.validateResource (/usr/src/medplum/packages/server/dist/fhir/repo.js:539:24)\n at Repository.updateResourceImpl (/usr/src/medplum/packages/server/dist/fhir/repo.js:443:20)\n at process.processTicksAndRejections