Incidents
Each snapshot is ten minutes of telemetry around one injected fault, or a healthy control. pilot (n=42 snapshots)
adHighCpu-r1adHighCpu-r2emailMemoryLeak-r1emailMemoryLeak-r2healthy-r1healthy-r2hwContention-r1hwContention-r2hwContention-r3hwContention-r4hwWeakSignal-r1hwWeakSignal-r2hwWeakSignal-r3hwWeakSignal-r4hwWifiOutage-r1hwWifiOutage-r2hwWifiOutage-r3hwWifiOutage-r4imageSlowLoad-r1imageSlowLoad-r2intlShippingSlowdown-r1intlShippingSlowdown-r2kafkaQueueProblems-r1kafkaQueueProblems-r2netemLoss-r1netemLoss-r2paymentFailure-r1paymentFailure-r2paymentUnreachable-r1paymentUnreachable-r2productCatalogFailure-r1productCatalogFailure-r2productCatalogLockContention-r1productCatalogLockContention-r2recommendationCacheFailure-r1recommendationCacheFailure-r2toxiBandwidth-r1toxiBandwidth-r2toxiLatency-r1toxiLatency-r2toxiReset-r1toxiReset-r2