Fleet overview
What is broken and who caused it: cluster health, incidents, Azure events and the control plane, then five core health measures across all production clusters.
Regions
Azure Service Health eventTile colour is the worst cluster health in the region. Select a tile to scope the page.
Likely Azure-caused
Degraded clusters where Resource Health, a Scheduled Event, disk throttling or Blob errors occurred within ±10 minutes of the symptom.
- ANQ-0142 · Media & Entertainment41m agoRead latency p95 27 ms
- −4 minThrottlingdisk-anq-c0142-data-07 IOPS consumed 100%
- −3 minResource Healthdisk-anq-c0142-data-07 Degraded
- ANQ-0217 · Life Sciences14m agoRead latency p95 21 ms
- −6 minScheduled EventFreeze on vm-anq-c0217-n02
- +2 minBlob errorsstanq0217data02 server errors 1.8%
No Azure signal nearby for ANQ-1236. Check the cluster itself first.
3 of 200 clusters breach at least one health threshold
Fleet averages in all production clusters look healthy, but 5 of 5 measures have clusters past their threshold. Start with the worst cluster on each card.
Read latency (p95) · all production clusters
The slower end of file-read response times on each cluster.
Qumulo telemetry · 200 clusters · 2-hour intervals, UTC
Clusters by value
Worst firstOpen the cluster and correlate with data-disk latency and queue depth.
Early-warning indicators
All risk signalsTop 10 noisy clusters · 7 days
Lowest quota headroom
Capacity pressure regions
- West EuropeHighL32s_v3 constrained in zone 24 allocation failures (7d)VM create time 1.8× baseline
- Southeast AsiaHighVM create time 1.5× baseline, quota 71% used1 allocation failure (7d)VM create time 1.5× baseline
- East US 2ModeratePremium SSD v2 provisioning slower in zone 10 allocation failures (7d)VM create time 1.3× baseline