Q: Prometheus graphs have regular gaps every 15 minutes for many targets. What do you investigate?
Regular gaps point to scrape or ingestion problems, not random application behavior.
Want to master this scenario in a live sandbox? The Linux Foundation's Prometheus Certified Associate (PCA) & Monitoring Labs covers this exact problem with hands-on terminal drills.
🛠️ Production Runbook & Step-by-Step Resolution
Direct Answer: "In an interview, I explain how we designed actionable, symptom-based alerting using the Four Golden Signals. The interviewer is testing: Scrape reliability, timeouts, sample limits, target health.. I structure my answer around systematic triage first, root cause analysis second, and permanent remediation third."
Initial Diagnostics & Root Cause Analysis
Regular gaps point to scrape or ingestion problems, not random application behavior.
upfor affected targets during the gaps.scrape_duration_secondsto see whether scrapes are timing out.scrape_samples_scrapedand sample-limit errors if exporters produce too many metrics.- Prometheus CPU, memory, WAL, and disk I/O during the gap.
Remediation & Permanent Safeguards
I would check: The fix depends on the cause: increase scrape timeout carefully, reduce exporter work, shard Prometheus, or remove expensive metrics.
- Network, DNS, service discovery, or load balancer behavior on a 15-minute schedule.
- Exporter logs for slow collection, especially exporters that call cloud APIs or databases.
- up for affected targets during the gaps.
- scrape_duration_seconds to see whether scrapes are timing out.
- scrape_samples_scraped and sample-limit errors if exporters produce too many metrics.
Senior hiring managers and bar raisers evaluate how you prioritize during outages. Avoid these common anti-patterns:
- Injecting high-cardinality label dimensions (such as user IDs or request UUIDs) into Prometheus metrics when configuring Observability Prometheus graphs High-cardinality label values cause exponential time-series growth, consuming gigabytes of TSDB memory and crashing Prometheus servers.
- Evaluating application performance solely through arithmetic mean or p50 latency while ignoring tail latency (p99) when configuring Observability Prometheus graphs Averages hide catastrophic tail latency outages where 1% of high-value user transactions suffer severe timeouts.
- Failing to inject and extract W3C distributed trace context across message brokers and RPC calls when configuring Observability Prometheus graphs Breaking distributed trace context propagation fragments end-to-end transaction waterfalls into disconnected, un-correlatable trace segments.
- Configuring alert rule evaluation intervals without sufficient duration damping (for: 5m), inducing alert flap when configuring Observability Prometheus graphs Instant alerts triggered by transient sub-second CPU or latency blips page on-call engineers unnecessarily and dilute response urgency.