---
title: "observability"
lang: "en"
author: "Mohammad Abu Mattar"
canonical: https://mkabumattar.com/incidents/categories/observability
---

[Home](/)›[All Incidents](/incidents)›[All Categories](/incidents/categories)›[Observability](/incidents/categories/observability)

Category

1Incident

# Observability

CategoryAllReliability (2)Architecture (1)CI/CD (1)Data Engineering (1)Observability (1)Performance (1)Security (1)TagAllPostmortem (4)Admission Control (1)AWS (1)Caching (1)Cron (1)Data Pipeline (1)Event Driven (1)Failover (1)Idempotency (1)Kubernetes (1)Latency (1)Monitoring (1)Multi Region (1)Payments (1)PostgreSQL (1)Provenance (1)Redis (1)Silent Failure (1)SLSA (1)SQS (1)Supply Chain (1)SortNewest firstOldest firstA to Z

## [Three Weeks of "Sometimes Doesn't Load", While p50 Stayed Green](/incidents/post/p99-tail-latency-three-week-blind-spot)

-   Sev2
-   About three weeks
-   [Performance](/incidents/categories/performance)
-   [Observability](/incidents/categories/observability)
-   14 Oct 2026

A largest-account customer's dashboard intermittently failed to load for about three weeks while the p50 dashboard showed green. p99 was 4.2 seconds against a 90ms median, driven by a nine-table join running 214 times per request.

[#Postmortem](/incidents/tags/postmortem)[#Latency](/incidents/tags/latency)[#Caching](/incidents/tags/caching)+3 tags

[Read Postmortem](/incidents/post/p99-tail-latency-three-week-blind-spot)