Timeouts and Latency
184 incident problems that show up as “Timeouts and Latency”.
먼저 읽을 가이드
추천 문제
All problems (184)
LINUX-324A bond reports link up while only one member still honors the new VLAN tagging profile and traffic quality becomes asymmetric during a failover rehearsalThe aggregate interface remains healthy by state and not by real forwarding behavior. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.LinuxAdvanced18 minProCICD-387A canary job compares median latency while one percentile tail explodes only for a tenant class routed through a different auth provider during a staged decommissionAggregate health stays green while a routed subset proves the release is unsafe. The service still works through the primary path, but one dependency only fails when the old component is finally drained away.CI/CDAdvanced18 minProCICD-150A canary metric query excludes the holiday traffic segment and the release looks successful onlyThe analysis engine reports low error rates, yet its sampling window omits the exact traffic cohort that reveals the bug.CI/CDAdvanced18 minProLINUX-252A direct firewall rule survives live reload while the nftables backend reorders chains on restart during a failover rehearsalThe runtime policy looked right and the boot-time chain order changed how packets were really evaluated. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.LinuxAdvanced18 minProLINUX-270A network bond remains up while carrier checks miss the upstream loss mode that only changed traffic quality during a failover rehearsalThe link never reports down and the service path is still broken in practice. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.LinuxAdvanced18 minProCICD-162A release gate reads a stale error budget snapshot during a failover rehearsalThe canary automation evaluates lagged data and promotes or blocks on the wrong signal. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.CI/CDAdvanced18 minProK8S-264A source-range rule trusts node addresses while externalTrafficPolicy Local leaves one zone with no eligible ingress path during a failover rehearsalThe load balancer policy is correct and locality changes which nodes can actually receive traffic. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.KubernetesAdvanced18 minProNETWORK-101An OSPF sham link comes up, but the cost still makes the MPLS VPN choose the slower backdoor pathAdjacency is healthy, yet path selection remains wrong because the cost model still favors the unintended route source.NetworkAdvanced18 minProCICD-139Canary analysis compares the old metric label after a service rename and promotes a release on meaningless dataThe statistical gate still returns a verdict, but it is evaluating the wrong stream after an observability label migration.CI/CDAdvanced18 minProK8S-116kube-proxy in IPVS mode retains a stale destination after a rapid rollout and some clients keep hitting terminated podsThe Service endpoints update correctly, yet a subset of traffic still lands on dead backends because the node-level forwarding table lagged behind the event stream.KubernetesAdvanced18 minProLINUX-216Policy routing marks the outbound packets correctly while rp_filter still rejects the asymmetric return path during a failover rehearsalForwarding is valid in one direction and kernel anti-spoofing defeats the other. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.LinuxAdvanced18 minProK8S-115Audit log backend stalls and admission latency spikesRequests time out across the cluster, but the root cause is not the applications; it is the control plane waiting on an overloaded audit destination.KubernetesAdvanced19 minProLINUX-105nftables accepts established flows but drops the related ICMP fragmentation signal so PMTU discovery never convergesNormal traffic starts, then large payloads stall because the firewall policy forgot a control message that is not part of the main flow tuple.LinuxAdvanced19 minProK8S-074Admission latency spikesThe webhook service exists, but requests from one path stall because the backing Deployment lost zonal coverage and the API server keeps timing out on remote retries.KubernetesAdvanced21 minProSECURITY-014Container image scan passes base layer but misses runtime package driftContainer image scan passes base layer but misses runtime package drift is a hands-on troubleshooting drill. The registry scan is green, but runtime package installation changes the actual container risk profile after deploy. Incident Response Operations needs to be checked by...SecurityIntermediate21 minProLINUX-022Systemd service restarts too quickly to capture useful logsA restart loop makes the service hard to inspect because the process dies before operators can capture the right signal.LinuxIntermediate21 minProLINUX-025Socket backlog tuning hides application accept bottleneckKernel queue settings were increased, but the service still stalls because the application cannot drain connections fast enough.LinuxIntermediate22 minProLINUX-039NFS client hits stale file handle after the export path movedPermissions and network reachability look fine, yet file operations fail because the server-side export changed under a still-mounted client path.LinuxAdvanced24 minProK8S-056kube-proxy keeps sending traffic to a deleted backendEndpoints changed correctly, but some nodes still route to a dead backend because stale load-balancing state was not reconciled as expected.KubernetesAdvanced26 minProNETWORK-008Packet loss that recurs only between the application and the DBStep-by-step narrowing of a situation where the whole network looks fine but loss occurs on only a specific path.NetworkAdvanced26 minPro