Hi All, I'm working through https://www.medplum.com/docs/self-hosting/install-on-gcp and I think I've got most of it right. Terraform has run successfully and I see the redis and postgres IPs. I've also set up the secret, DNS, values.yaml, and run the helm chart. I can see my server, the hostname, and an IP address. However, I keep getting output like the attached image when I run kubectl describe pod. Looks like the server is not starting? Any ideas where to go from here?

Can you see the logs? kubectl logs -n medplum medplum-server-6d76766c49-6cgv9

kubectl logs -n medplum medplum-server-6d76766c49-6cgv9 {"level":"INFO","timestamp":"2025-01-31T18:29:59.843Z","msg":"Starting Medplum Server...","configName":"gcp:composite-store-387221:config-secret"} A new version (v3.2.29) of Medplum is available. Your current version (v3.2.28) may be missing important updates and bug fixes. {"level":"INFO","timestamp":"2025-01-31T18:30:03.306Z","msg":"Building structure definitions..."} {"level":"ERROR","timestamp":"2025-01-31T18:30:04.291Z","msg":"Database error","error":"OOM command not allowed when used memory > 'maxmemory'.","stack":"ReplyError: OOM command not allowed when used memory > 'maxmemory'.\n at parseError (/usr/src/medplum/node_modules/redis-parser/lib/parser.js:179:12)\n at parseType (/usr/src/medplum/node_modules/redis-parser/lib/parser.js:302:14)"} d [Error]: OOM command not allowed when used memory > 'maxmemory'. at normalizeDatabaseError (/usr/src/medplum/packages/server/dist/fhir/sql.js:428:12) at Repository.withTransaction (/usr/src/medplum/packages/server/dist/fhir/repo.js:1811:80) at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at async seedDatabase (/usr/src/medplum/packages/server/dist/seed.js:24:5) at async /usr/src/medplum/packages/server/dist/app.js:193:9 at async initApp (/usr/src/medplum/packages/server/dist/app.js:139:5) at async main (/usr/src/medplum/packages/server/dist/index.js:34:17) { outcome: { resourceType: 'OperationOutcome', issue: [ [Object] ] }, cause: undefined }

can you try increasing the redis max memory and restart the server?

Sure. I also set maxmemory-policy to allkey-lru Here's what I'm getting in the logs now, over and over (I assume meaning it's crashing while building the value sets and then restarting) kubectl logs -f -n medplum medplum-server-6d76766c49-8f4jp {"level":"INFO","timestamp":"2025-01-31T19:21:19.659Z","msg":"Starting Medplum Server...","configName":"gcp:composite-store-387221:config-secret"} A new version (v3.2.29) of Medplum is available. Your current version (v3.2.28) may be missing important updates and bug fixes. {"level":"INFO","timestamp":"2025-01-31T19:21:22.686Z","msg":"Building structure definitions..."} {"level":"INFO","timestamp":"2025-01-31T19:21:51.941Z","msg":"Finished building structure definitions","durationMs":29254} {"level":"INFO","timestamp":"2025-01-31T19:21:51.941Z","msg":"Building value sets..."}

phew - glad to hear it