kubernetes
8 minKubernetes Requests, Limits, and Autoscaling: A Practical Capacity Model
Connect pod resource requests and limits to scheduling, throttling, eviction risk, and Horizontal Pod Autoscaler behavior.
Search the library
22 matching resultsArticles and interview answers
Q&A
Q&AWhy rotate a session identifier after authentication?
Rotation limits session fixation. A session identifies a principal, while each protected operation still requires independent authorization.
Read the concise answer
Rotation limits session fixation. A session identifies a principal, while each protected operation still requires independent authorization.
unix
18 minLinux Under Load: Epoll, File Descriptors, OOM, TCP Queues, and VFS
Connect kernel readiness APIs, resource limits, cgroups, socket queues, page cache, and inode behavior to production symptoms.
distributed systems
18 minDistributed Systems Advanced: Rebalancing, Sagas, Consensus, and Trace Context
Design around Kafka group churn, payment idempotency, coordination limits, sagas, CQRS, and context propagation across async boundaries.
kubernetes
5 minKubernetes CronJobs: Timezones, Missed Starts, and Duplicate Work
Operate a scheduled report as a reconciled workload with explicit timezone, lateness limits, and application-level deduplication.
Q&A
Q&AWhat happens when a Kubernetes container reaches its CPU or memory limit?
CPU can be throttled; exceeding a memory limit can trigger an OOM kill.
Read the concise answer
CPU usage above a configured limit is throttled, which can increase latency without stopping the process. Memory is not throttled in the same way: crossing the container limit can cause the kernel to kill the process, often reported as OOMKilled. Requests affect scheduling and utilization-based HPA calculations, so tune all values from observed workload behavior.
Q&A
Q&AWhy can a pod be CPU-throttled while its node is only 40% busy?
CPU quota is enforced per container, so node-wide unused capacity may not be available beyond its limit.
Read the concise answer
A container can exhaust its CFS quota during a period even if other node CPUs are idle. Inspect container throttling counters and request/limit values alongside latency. Decide whether the limit is needed, whether requests/HPA targets match demand, or whether CPU work should be reduced; do not infer container headroom from node averages.
Q&A
Q&AWhat does asyncio queue completion actually prove?
Every accepted item has been acknowledged, not necessarily committed successfully.
Read the concise answer
A bounded queue limits waiting items; workers hold additional active items. task_done releases unfinished-item accounting, while business success and durable retries require separate records.