A civic open-data InfiniBand pod finishes HA provisioning. What should on-call documentation include?
Select an answer to reveal the explanation.
Short Explanation
Night-shift staff cannot guess where the SM lives. Document InfiniBand HA management endpoints and reachability for the civic open-data pod’s on-call runbook.
Full Explanation
After a civic open-data InfiniBand pod finishes HA provisioning, on-call documentation must include InfiniBand HA management endpoints and reachability so night-shift staff can find and contact the SM pair. Runbooks should list master/standby (or VIP) management addresses, how to verify which node is active, and how to reach them from the municipal ops network. Listing only the primary subnet manager address with local console recovery details fails because on-call needs the full HA management endpoint pair and reachability, not just a single primary address plus console recovery paths that may not be accessible during a fabric incident. Listing InfiniBand fabric partition membership and PKey assignments for tenants fails because those are tenant fabric configuration details, not the operator-facing HA management endpoints and reachability on-call needs during an incident. Providing Ethernet fabric NetQ monitoring URLs and spine login inventory for operators fails because those are Ethernet telemetry and access records, not the InfiniBand HA management endpoints and reachability that on-call staff need for the IB fabric. Exam caveat: HA provisioning is incomplete without operator-facing endpoint documentation distinct from Ethernet fabric NetQ monitoring URLs and spine login inventory. Operational check: from a jump host, open the documented IB HA management endpoints, confirm master/standby identity matches the runbook, and have on-call practice the reachability path before the open-data pod is declared production-ready.