What was visible during the incident, such as pods Pending or DNS failures.
service-unavailable9 cases · 2017 to 2025
high-error-rate7 cases · 2017 to 2023
no-available-ip5 cases · 2020 to 2025
oom-killed4 cases · 2019 to 2023
pod-pending4 cases · 2020 to 2025
high-latency3 cases · 2019 to 2024
pods-evicted3 cases · 2019 to 2025
upgrade-stuck3 cases · 2019 to 2025
control-plane-unresponsive2 cases · 2019 to 2024
dns-resolution-failures2 cases · 2019 to 2024
gpu-unavailable2 cases · 2024
image-pull-failures2 cases · 2023
intermittent-connection-failures2 cases · 2019 to 2020
link-flapping2 cases · 2024 to 2026
monitoring-down2 cases · 2019 to 2023
nccl-timeout2 cases · 2024 to 2026
node-not-ready2 cases · 2023 to 2024
resources-deleted2 cases · 2021 to 2023
autoscaler-misbehaving1 case · 2019
connection-resets1 case · 2020
control-plane-slow1 case · 2019
cpu-throttling1 case · 2020
credentials-unavailable1 case · 2019
database-overloaded1 case · 2024
intermittent-timeouts1 case · 2020
uneven-load1 case · 2019
requests-misrouted1 case · 2025
node-disk-full1 case · 2023
nodes-recreated-in-loop1 case · 2019
pod-stuck-container-creating1 case · 2023
readiness-probe-failures1 case · 2020
response-quality-degraded1 case · 2025
autoscaler-scale-up-failed1 case · 2020
stale-service-routing1 case · 2017
throughput-degraded1 case · 2024
training-job-failed1 case · 2026