See Waz work
A real infrastructure event just came in. Watch Waz correlate alerts across systems, identify root cause, and draft the procedures your operator needs.
Inlet temperature exceeding ASHRAE envelope: Row C, Rack C-12
3 Correlated Alerts Detected
Topology
DM-Akron · Row C · Rack C-12 (42U). 4 production hosts in U18-U24 (db-prod-akron-04 in U22). Primary cooling: hot-aisle CRAC-08 (Schneider Galaxy VS InRow). Redundant: CRAC-07 (currently 78% load).
via Source of Truth
Health Signals
C-12 inlet temp 29.4°C (ASHRAE A1 ceiling 27°C); CRAC-08 supply airflow 62% of nominal; db-prod-akron-04 CPU package 84°C, entering thermal throttle; humidity 47% (normal).
via Prometheus + Netdata
Recent Logs
14:21:08 Schneider crac-08 alarm: SUPPLY_AIRFLOW_LOW; 14:21:33 prometheus node_thermal_throttle{host="db-prod-akron-04"}=1; 14:22:14 sunbird rack rack-c-12 inlet_temp_c=29.4 (threshold 27.0); 14:22:51 ipmi sensor cpu_temp warning on 3 hosts in C-12.
via Splunk
This walkthrough runs entirely in your browser. The full investigation runs live in Spektro.
[01] The Relationship Graph
Waz maps how your facility connects
Every signal is clustered by relationship, power, network, cooling, so when an incident fires, Waz already knows its blast radius and isolates it instantly.
[02] The Core
Unified Architecture
[03] Knowledge Base
A living knowledge base
Every procedure Waz drafts becomes a searchable, shareable record, an MOP, EOP, JHA, SOP, or postmortem, grounded in the incident that produced it.
Replace CRAC-08 Return-Air Filter Under Reduced Cooling
crac-akron-08
- Generated by
- Waz
- Incident
- CRAC-08 supply airflow restricted
- Risk class
- Medium
- Approval
- Operator review required
1. Summary
Return-air filter on CRAC-08 has reached end of life (differential pressure 1.42" w.c.), cutting supply airflow to 62% of nominal and driving Rack C-12 inlet temps past the ASHRAE A1 ceiling. This procedure restores cooling capacity while protecting the throttled workloads in Row C.
2. Procedure
Spin up CRAC-07 to full and stage portable spot-cool at the Row C cold-aisle endcap. Drain db-prod-akron-04 to the secondary cluster. LOTO CRAC-08, replace the return-air filter, vacuum the coil face. Restore CRAC-08 and verify supply airflow at or above 95% of nominal.
[04] Operators Always Approve
Humans + Agents
Spektro absorbs the hours of triage, correlation, and documentation. Your operators stay in the driver's seat for every production change. Every artifact is reviewed, every action logged.
- ✓Zero-Touch Context
- ✓Audit-Ready Logs
- ✓Human-in-the-loop Control
Infrastructure Context
Sunbird + NetBox Synced[05] The Payoff
What operators get
Cut MTTR
Diagnosis and the procedure land in seconds, not hours.
End alarm fatigue
Correlated root causes replace thousands of raw alarms.
Protect uptime
Catch the failing unit before it cascades across the row.
Reclaim operator hours
The mechanical triage work stops landing on your team.