topic / cloud-internals
Cloud internals
Understand the kernel, runtime, and control-plane machinery behind cloud platforms.
Trace how containers, Kubernetes control planes, Linux isolation, networking, storage, and observability behave under real workloads and failure conditions.
- 0 articles
- 0 videos
- 0 traces
- 0 flows
- 8 interviews
- 8 total
Articles
// empty
No articles published in this topic yet
The queue is running. Subscribe to the dispatch to be told the moment this topic ships.
Join the dispatchRequest flows
// empty
No request flows in this topic yet
Interactive request journeys will appear here as this topic is built out.
Browse request flowsVideos
// empty
No videos yet
Video coverage for this topic is planned. Watch the library page for updates.
Browse videosTraces
// empty
No traces in this topic yet
Traces are the step-by-step walkthroughs. They land as the topic matures.
Browse tracesInterviews
easy · t1Explain the difference between Kubernetes `requests` and `limits` for both CPU (`cpu.weight` vs `cpu.max` CFS quota) and Memory (`memory.min/low` +.Contrast compressible CPU (throttled via CFS `cpu.max` every 100 ms) with incompressible Memory (killed with `SIGKILL` 137 on `memory.max`), trace why old runtimes read `/proc/cpuinfo` (64 host cores) instead of cgroup limits, and explain Kubernetes QoS classes (`Guaranteed`, `Burstable`, `BestEffort`).medium · t1Explain how the Linux Kernel eBPF architecture works from start to finish: `bpf()` syscall loading, the Static Verifier (`data_end` pointer tracking &.Contrast eBPF with `.ko` kernel modules, explain why `XDP` beats `iptables` by running before `sk_buff` allocation, explain `BPF_MAP_TYPE_PERCPU_HASH` for lock-free metrics, and show how `Tetragon` correlates kernel `task_struct` cgroup IDs with Kubernetes Pod metadata.medium · t1Trace what happens inside `kube-apiserver` and `etcd` when you run `kubectl apply -f deployment.yaml`. Explain `etcd`'s MVCC `ModRevision`.Walk through Authentication -> Authorization (RBAC) -> Mutating Admission Webhooks -> Schema Validation -> Validating Admission Webhooks -> `etcd` Raft MVCC transaction -> `WatchCache` fan-out to Informers.medium · t1Draw the full architecture of a Kubernetes Controller / Operator using `client-go` (`Reflector`, `DeltaFIFO`, `SharedInformer`, `Indexer`.Trace the 6 components in `sample-controller`, explain the `dirty`/`processing` workqueue sets, contrast in-cluster garbage collection (`ownerReferences`) with external cleanup (`finalizers`), and explain why mutating `Indexer` pointers corrupts shared memory.medium · t1Explain how a Kubernetes `ClusterIP` Service works under the hood in `iptables` mode, `IPVS` mode, and `eBPF` (`Cilium`) mode. Then explain the.Compare $O(N)$ `iptables` rule chains vs $O(1)$ `IPVS` hash tables vs $O(1)$ eBPF `cgroup/connect4` socket rewriting, and trace the parallel timeline of `Pod Deletion -> SIGTERM` vs `EndpointSlice -> kube-proxy / Ingress update`.easy · t1When someone runs `docker run` or Kubernetes schedules a Pod, explain the exact sequence of Linux kernel syscalls and namespaces (`clone`/`unshare`, `pid`.List the 7 Linux namespaces, trace `clone()` + `pivot_root` + `umount2(MNT_DETACH)`, explain why `PID 1` ignores unhandled `SIGTERM` and must reap orphaned `SIGCHLD` zombies, and explain how Kubernetes pods share `net`/`ipc`/`uts` namespaces via the `pause` infra container.easy · t1Walk through how `kube-scheduler` schedules a Pod (`PreFilter -> Filter -> PostFilter/Preemption -> Score -> Reserve -> Permit -> Bind`). How do you combine.Explain the synchronous Scheduling Cycle (`AssumePod` in `Reserve`) vs asynchronous Binding Cycle, contrast `podAntiAffinity` (hard 1-per-node limit) vs `topologySpreadConstraints` (`maxSkew`), and pair multi-AZ spreading with `PDB` (`maxUnavailable: 1`).easy · t1Trace the exact journey of a TCP packet from `Container A` (`10.244.1.5`) to `Container B` (`10.244.2.9`) on a different Kubernetes node under both a `VXLAN`.Walk through Pod NetNS `eth0` -> `veth` pair -> Host NetNS `cni0` bridge -> routing table -> `vxlan0` UDP 8472 encapsulation -> physical `eth0` -> remote `vxlan0` decapsulation, and explain AWS VPC ENI secondary IP exhaustion.