# 土法炼钢 - 系统与基础设施 Linux 内核、存储与网络、可观测性、系统架构与大模型基础设施的工程笔记：机制拆解、踩坑复盘与可核对证据，少空谈。 This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## 【kube-apiserver】List、Pagination 与一致性 List：continue token 与 etcd Range 成本 DevFeed: [【kube-apiserver】List、Pagination 与一致性 List：continue token 与 etcd Range 成本]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [Cache](), [version]() Tags: [410-gone](), [apiserver](), [cacher](), [continue-token](), [distributed](), [etcd](), [etcd-range](), [http](), [kubernetes](), [limit](), [list](), [pagination](), [range](), [resourceversion](), [v1-30-3]() ### AI overview This article examines Kubernetes v1.30.3 List pagination in kube-apiserver, explaining continue-token semantics, limit-and-continue etcd Range behavior, resourceVersion consistency, and the different costs of cacher and etcd3 paths. It also explains how label and field selectors affect scanning and filtering. ### Source excerpt 钉 Kubernetes v1.30.3 List 分页的 continue token 编码语义、limit+continue 多轮 etcd Range 行为、resourceVersion 对一致性语义的影响，以及 label/field selector 在 cacher 与 etcd3 路径上的不同成本。 ## 【kube-apiserver】Admission 链概览：内置插件顺序与 webhook 边界 DevFeed: [【kube-apiserver】Admission 链概览：内置插件顺序与 webhook 边界]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes]() Tags: [admission](), [api](), [apiserver](), [authz](), [controllers](), [distributed](), [kubernetes](), [mutating](), [namespace-lifecycle](), [openapi](), [resourcequota](), [timeout](), [v1-30-3](), [validating](), [webhook]() ### AI overview This tutorial explains the Kubernetes v1.30.3 admission chain in kube-apiserver. It places Admission between authorization and storage, describes the Mutating-then-Validating order, and covers built-in plugins, webhooks, timeout behavior, and fail-open versus fail-closed considerations. ### Source excerpt 钉 Kubernetes v1.30.3 Admission 链的阶段位置、Mutating 与 Validating 两阶段顺序、内置插件注册路径与典型示例、webhook 边界与 timeout 语义，以及 fail-open vs fail-closed 的工程争议。 ## 【kube-apiserver】选型收束与开放问题：排除树、Kine 与 events 分集群 DevFeed: [【kube-apiserver】选型收束与开放问题：排除树、Kine 与 events 分集群]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [Caching](), [etcd](), [SQL](), [MySQL](), [PostgreSQL](), [SQLite](), [Raft]() Tags: [apiserver](), [churn](), [distributed](), [errcompacted](), [etcd](), [events](), [exclusion-tree](), [kine](), [kubernetes](), [mysql](), [open-questions](), [postgresql](), [quota](), [raft](), [selection](), [sql](), [sqlite](), [storage](), [ttl](), [v1-30-3](), [watch-cache]() ### AI overview This concluding article presents a mechanism-based decision tree for diagnosing Kubernetes control-plane issues across kube-apiserver and etcd layers. It explains Kine's differing Watch semantics when using SQLite, PostgreSQL, or MySQL backends, discusses routing Kubernetes Events to a separate etcd cluster, and identifies open questions around watch-cache SLOs, compaction, linear reads, and Lease fencing. ### Source excerpt 用机制排除树收束何时查 apiserver 轴、何时穿透 etcd 轴、何时两轴联查；回收 etcd/13 写下的 apiserver 停损线；Kine SQL backend 的 Watch 语义差与 Jepsen 覆盖空白；events --etcd-servers-overrides 分集群运维；列出 watch cache SLO 联合模型、线性读期望与 Lease+fencing 全链路三个开放问题；以 ADR 语言关闭系列边界。 ## 【kube-apiserver】storage.Interface 与 etcd3：codec、prefix 与 CRUD 路径 DevFeed: [【kube-apiserver】storage.Interface 与 etcd3：codec、prefix 与 CRUD 路径]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [Encryption](), [Raft]() Tags: [api](), [apiserver](), [codec](), [crud](), [distributed](), [encryption](), [encryption-at-rest](), [etcd3](), [guaranteed-update](), [kubernetes](), [optimistic-concurrency](), [raft](), [rest](), [storage](), [v1-30-3]() ### AI overview This tutorial explains how kube-apiserver's storage.Interface integrates with the etcd3 backend in Kubernetes v1.30.3. It covers key prefixes, codec serialization, value transformation and encryption-at-rest boundaries, optimistic concurrency through GuaranteedUpdate and transactions, and the mapping from etcd3 watch streams to apiserver events. ### Source excerpt 拆解 kube-apiserver 的 storage.Interface 契约与 etcd3 实现：codec 序列化、pathPrefix/resourcePrefix、value.Transformer 加密边界、GuaranteedUpdate 乐观并发，以及 Watch 到 etcd3 的完整路径。版本锚定 Kubernetes v1.30.3 / etcd v3.5.33。 ## 【kube-apiserver】运维与升级：HA、flags、graceful shutdown 与 etcd 联检 DevFeed: [【kube-apiserver】运维与升级：HA、flags、graceful shutdown 与 etcd 联检]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [Encryption]() Tags: [api](), [apiserver](), [audit](), [backend](), [config](), [distributed](), [encryption](), [etcd](), [flag](), [graceful-shutdown](), [ha](), [kubernetes](), [leader-election](), [load-balancer](), [operations](), [retry](), [servers](), [upgrade](), [v1-30-3]() ### AI overview This Chinese-language operations tutorial explains kube-apiserver high availability and upgrades for Kubernetes v1.30.3. It clarifies that multiple apiserver instances run concurrently without internal leader election, sharing an etcd cluster, and covers key flags, graceful shutdown, encryption configuration, audit webhooks, upgrade checks, and version-skew guidance. ### Source excerpt kube-apiserver 高可用模式：多实例共享 etcd、无需内部 leader election；核心 flag 语义（--etcd-servers、--etcd-servers-overrides、--shutdown-delay-duration、encryption-provider-config）；与 etcd/14 的联合升级检查单；Kubernetes 版本偏差策略与 etcd 矩阵指针。 ## 【kube-apiserver】控制面全景：缺口、五轴坐标系与 16 篇路线 DevFeed: [【kube-apiserver】控制面全景：缺口、五轴坐标系与 16 篇路线]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [etcd]() Tags: [admission](), [apf](), [apiserver](), [auth](), [control-plane](), [distributed](), [etcd](), [kubernetes](), [network](), [overview](), [storage](), [v1-30-3](), [watch](), [watch-cache]() ### AI overview This article maps gaps in existing coverage of the Kubernetes control plane and proposes a 16-part reading route organized around Storage, Watch, Admission, Auth, and APF. It anchors the discussion to Kubernetes v1.30.3 and etcd v3.5.33, focusing on how requests travel through kube-apiserver and how to distinguish admission, caching, backend, and timeout failures. ### Source excerpt 相对 etcd/13、distributed/50、k8s-network 补齐 kube-apiserver 生产内核缺口；以 Storage/Watch/Admission/Auth/APF 五轴为坐标系定义 16 篇阅读路线；版本锚定 Kubernetes v1.30.3。 ## 【kube-apiserver】Authentication：SA、Bearer、OIDC 边界 DevFeed: [【kube-apiserver】Authentication：SA、Bearer、OIDC 边界]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Authentication](), [Kubernetes](), [OpenID connect (OIDC)](), [JSON Web Tokens](), [authz]() Tags: [api](), [apiserver](), [authentication](), [authorization](), [bearer-token](), [distributed](), [jwt](), [k8s](), [kubernetes](), [oidc](), [serviceaccount](), [v1-30-3]() ### AI overview This Chinese-language tutorial explains the kube-apiserver authentication chain in Kubernetes v1.30.3. It distinguishes X.509 certificates, ServiceAccount tokens, static and bootstrap tokens, OIDC JWTs, webhook authentication, and anonymous requests, and separates authentication failures such as 401 from storage-layer failures such as 503 or 504. ### Source excerpt 钉 K8s v1.30.3 的认证链：X509、SA token（bound token）、Bearer、OIDC、Webhook token、匿名用户；ServiceAccount bound token 与 TokenRequest 语义；apiserver 认证 flags 边界；401 与存储层故障的分列；AuthZ 路径见第 11 篇。 ## 【kube-apiserver】Watch 路径（服务端）：长连接、410 Gone 与 timeout DevFeed: [【kube-apiserver】Watch 路径（服务端）：长连接、410 Gone 与 timeout]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: article Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [HTTP](), [kubectl]() Tags: [410-gone](), [apiserver](), [bookmark](), [cacher](), [distributed](), [etcd](), [etcd3](), [informer](), [kubernetes](), [resourceversion](), [rest](), [timeout](), [v1-30-3](), [watch]() ### AI overview This article explains the Kubernetes v1.30.3 apiserver Watch path, from the REST handler through cacher and etcd3 watchers. It distinguishes 410 Gone caused by cacher ring-buffer eviction from errors propagated from etcd compaction, and covers resourceVersion semantics, timeouts, bookmarks, connection lifecycle, and the boundary with client-go Informer. ### Source excerpt 钉 Kubernetes v1.30.3 服务端 Watch 路径：从 REST handler 到 cacher/etcd3 watcher 的建立过程、resourceVersion=0 与具体 RV 的语义差异、410 Gone 与 etcd ErrCompacted 的映射、bookmark 推送与连接生命周期，以及与 client-go Informer 的边界。 ## 【kube-apiserver】进程与请求路径：generic apiserver、HandlerChain 与 REST 路由 DevFeed: [【kube-apiserver】进程与请求路径：generic apiserver、HandlerChain 与 REST 路由]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API]() Tags: [admission](), [apf](), [apiserver](), [audit](), [authn](), [authz](), [distributed](), [generic-apiserver](), [gvr](), [handlerchain](), [kubernetes](), [request-path](), [rest](), [v1-30-3]() ### AI overview This tutorial explains the Kubernetes v1.30.3 kube-apiserver process model, the generic apiserver framework, HandlerChain ordering, GVR routing, and the points where requests can be rejected before reaching storage. It also distinguishes failures caused by APF, authentication and authorization, admission, and storage. ### Source excerpt 拆解 kube-apiserver 进程模型与 generic apiserver 框架；钉 HandlerChain 各插槽顺序与失败落点；说明 GVR 路由机制与请求在到达 storage 前可能被拦截的位置。版本锚定 Kubernetes v1.30.3。 ## 【kube-apiserver】CRD / aggregation / 扩展边界：API 扩展停损线 DevFeed: [【kube-apiserver】CRD / aggregation / 扩展边界：API 扩展停损线]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: article Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [etcd](), [Extension](), [Server]() Tags: [aggregation](), [api](), [api-server](), [apiserver](), [apiservice](), [authorization](), [crd](), [distributed](), [encryption](), [etcd](), [extension](), [kubernetes](), [v1](), [v1-30-3](), [v3-5](), [v3-5-33](), [webhook]() ### AI overview This article explains the boundary between Kubernetes CRD and Aggregated API extension paths in kube-apiserver. CRD requests remain within the apiserver and use its storage and admission paths, while Aggregated API requests are proxied to an external Extension Server. The two paths have different failure modes and troubleshooting points, although both can appear as 503 errors or timeouts. ### Source excerpt 厘清 CRD v1 与 Aggregated API 两条扩展路径在 kube-apiserver 中的存储与请求分界：CRD 对象存 etcd、conversion webhook 失败如何体现在 Storage 轴，APIService 则把请求转发到外部 Extension Server 可引发 503。明确 scheduler/controller/kubelet 为扩展停损线之外的数据面与控制循环。 ## 【kube-apiserver】resourceVersion 与 Revision 映射：mod revision、continue 与一致性读期望 DevFeed: [【kube-apiserver】resourceVersion 与 Revision 映射：mod revision、continue 与一致性读期望]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [Caching]() Tags: [410-gone](), [api](), [apiserver](), [cache](), [consistency](), [continue-token](), [distributed](), [etcd](), [kubernetes](), [pagination](), [resourceversion](), [revision](), [v1-30-3](), [watch]() ### AI overview This article explains how Kubernetes resourceVersion maps to etcd mod_revision, including the different semantics for individual objects and List responses. It covers Watch starting points, continue-token pagination and its costs, compaction-related 410 Gone errors, and consistency differences between watch-cache and etcd read paths. The discussion is anchored to Kubernetes v1.30.3 and etcd v3.5.33. ### Source excerpt 钉 Kubernetes resourceVersion 字段与 etcd mod revision 的对应关系；分析 continue token 的分页语义与成本；说明不同 List 路径的一致性期望差异；以及 410 Gone 与 ErrCompacted 的分列。版本锚定 Kubernetes v1.30.3 / etcd v3.5.33。 ## 【kube-apiserver】Mutating / Validating Webhook：timeout、failurePolicy 与可用性门 DevFeed: [【kube-apiserver】Mutating / Validating Webhook：timeout、failurePolicy 与可用性门]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [configuration](), [dry-run](), [TLS (Transport Layer Security)](), [API]() Tags: [admission](), [api](), [apiserver](), [cel](), [configuration](), [distributed](), [kubernetes](), [kyverno](), [mutating](), [timeout](), [tls](), [v1-30-3](), [validating](), [webhook]() ### AI overview A tutorial on configuring and troubleshooting Kubernetes v1.30.3 mutating and validating admission webhooks. It explains configuration fields, timeout and failurePolicy behavior, production availability requirements, reinvocation and idempotency, CEL ValidatingAdmissionPolicy boundaries, and an evidence-based troubleshooting approach. It emphasizes that slow webhooks add latency to the admission path rather than indicating etcd lag. ### Source excerpt 钉 K8s v1.30.3 的 MutatingWebhookConfiguration / ValidatingWebhookConfiguration v1：timeoutSeconds、failurePolicy、sideEffects、reinvocationPolicy 字段语义；webhook 慢如何表现为写路径延迟而非 etcd lag；生产可用性门选取；CEL ValidatingAdmissionPolicy 作为内置替代路径；排障证据包。 ## 【kube-apiserver】排障五轴：Storage、Watch、Admission、Auth、APF DevFeed: [【kube-apiserver】排障五轴：Storage、Watch、Admission、Auth、APF]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [etcd]() Tags: [admission](), [apf](), [apiserver](), [auth](), [authz](), [distributed](), [etcd](), [five-axes](), [kubernetes](), [metrics](), [storage](), [troubleshooting](), [v1-30-3](), [watch]() ### AI overview A troubleshooting guide for kube-apiserver organized around five diagnostic axes: Storage, Watch, Admission, Auth, and API Priority and Fairness. It maps symptoms such as 504, 410 Gone, 401, 403, webhook timeouts, list storms, and OOM to likely axes, explains when to investigate etcd, and emphasizes that commands and metrics were not executed in the author's environment. ### Source excerpt 按 Storage/Watch/Admission/Auth/APF 五轴做症状否证；给出完整五轴命令表与症状->轴映射（504、410、401、403、webhook 超时、List 风暴、OOM）；说明 apiserver_request_duration_seconds 等核心 metrics 语义；并提供决策树：何时穿透到 etcd/15，何时留在 apiserver 轴。 ## 【kube-apiserver】Watch cache / cacher：dispatch、bookmark 与穿透 etcd DevFeed: [【kube-apiserver】Watch cache / cacher：dispatch、bookmark 与穿透 etcd]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: article Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Caching](), [Kubernetes](), [API](), [Kubernetes v1.32]() Tags: [410-gone](), [apiserver](), [bookmark](), [cache](), [cacher](), [distributed](), [kubernetes](), [metrics](), [resourceversion](), [storage](), [v1](), [v1-30-3](), [watch-cache]() ### AI overview This article explains the Kubernetes v1.30.3 apiserver watch cache and cacher implementation. It covers the watchCache sliding window, the Ready gate during startup synchronization, in-memory event dispatch, bookmark progress notifications, and the conditions that cause cache misses to fall back to etcd. ### Source excerpt 钉 Kubernetes v1.30.3 cacher 对 storage.Interface 的包装：watchCache 滑动窗口与 Ready 门、bookmark 推送路径、cache miss 打穿 etcd 的触发条件，以及 List 风暴成因与 watch cache SLO 与 compaction 间隔之间的开放问题。 ## 【kube-apiserver】Authorization 与 Audit：RBAC、SAR 与 403 分列 DevFeed: [【kube-apiserver】Authorization 与 Audit：RBAC、SAR 与 403 分列]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Authorization](), [Kubernetes](), [audit](), [API]() Tags: [403](), [admission](), [api](), [apiserver](), [audit](), [auth](), [authorization](), [authz](), [distributed](), [etcd](), [http](), [k8s](), [kubernetes](), [policy](), [post](), [rbac](), [subjectaccessreview](), [v1](), [v1-30-3](), [webhook]() ### AI overview This tutorial explains Kubernetes v1.30.3 authorization and auditing. It distinguishes the RBAC, Node, Webhook, and legacy ABAC authorizer paths; describes SAR and SSAR access checks; outlines audit policies and stages; and explains why user-facing 403 errors come from Kubernetes authorization rather than etcd permissions or Admission failures. ### Source excerpt 钉 K8s v1.30.3 的授权链：RBAC、Node、Webhook、ABAC（遗留）；SubjectAccessReview / SelfSubjectAccessReview 用法；Audit policy 四级；403 与 etcd 权限错误、Admission 拒绝的分列；不把 etcd 的 mTLS 错误写成 K8s RBAC 故障。 ## 【kube-apiserver】APF 与 max-in-flight：公平排队、504 与 etcd lag 分列 DevFeed: [【kube-apiserver】APF 与 max-in-flight：公平排队、504 与 etcd lag 分列]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: article Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [SRE](), [Linux]() Tags: [apf](), [api](), [apiserver](), [distributed](), [etcd](), [fairness](), [fairqueuing](), [flowcontrol](), [k8s](), [kubernetes](), [lag](), [max-in-flight](), [priority](), [timeout](), [v1-30-3]() ### AI overview This article explains how Kubernetes v1.30.3 protects kube-apiserver from overload through API Priority and Fairness (APF) and the older max-in-flight limits. It distinguishes 429 responses, APF queue timeouts that can produce 504 responses before storage is reached, etcd latency, and admission webhook delays, and describes APF's FlowSchema, PriorityLevelConfiguration, fair queuing, and shuffle sharding mechanisms. ### Source excerpt 钉 K8s v1.30.3 的 API Priority and Fairness（APF）：FlowSchema 匹配、PriorityLevelConfiguration 公平排队（SFVR）、与旧 max-in-flight flag 的共存关系；429/timeout/504 在 APF 排队、etcd_request_duration_seconds、Admission Webhook 三轴的分列；pkg/util/flowcontrol 路径；APF 与简单 max-in-flight 的运维复杂度争论。 ## 【etcd】MVCC 数据模型：Revision、keyIndex 与 generation DevFeed: [【etcd】MVCC 数据模型：Revision、keyIndex 与 generation]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Raft](), [Kubernetes]() Tags: [compaction](), [distributed](), [etcd](), [generation](), [keyindex](), [kubernetes](), [mvcc](), [raft](), [revision](), [treeindex](), [v3](), [v3-5](), [v3-5-33](), [watch]() ### AI overview This article analyzes etcd v3.5.33's MVCC data model through three concepts: globally ordered revisions, keyIndex generations that track key lifecycles, and the division of responsibilities between the in-memory treeIndex and the bbolt backend. It explains how these structures support historical Watch replay, transaction comparisons, compaction, and Kubernetes resourceVersion semantics, while distinguishing etcd revisions from Raft indexes. ### Source excerpt 拆解 etcd v3.5.33 的 Revision (main, sub) 全序、keyIndex/generation 生命周期与 treeIndex 分工；简要对照 v2 平面键空间，交代 MVCC 与 Watch/compaction 的语义基础。 ## 【etcd】写入路径深读：Txn、mod revision 与 quota/alarm DevFeed: [【etcd】写入路径深读：Txn、mod revision 与 quota/alarm]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Back end](), [Raft](), [Remote Procedure Call (RPC)](), [maintenance]() Tags: [alarm](), [apiserver](), [applied-index](), [apply](), [batch](), [bbolt](), [committed-index](), [compaction](), [defrag](), [distributed](), [errcompacted](), [etcd](), [follower](), [k8s](), [kubernetes](), [mod-revision](), [nospace](), [quota](), [revision](), [txn](), [v3-5](), [v3-5-33]() ### AI overview This tutorial explains the etcd v3.5.33 write path from gRPC requests through Raft proposal, quota checks, MVCC apply, and ModRevision assignment. It also describes proposal backpressure, backend quota accounting, and how a NOSPACE alarm blocks writes while reads, deletes, and maintenance operations can continue. ### Source excerpt 走读 etcd v3.5.33 写入路径：raftRequest 背压、Txn compare/mod 与 ModRevision 分配、backend quota 与 NOSPACE alarm 如何把集群推入只读。 ## 【etcd】bbolt 后端：mmap、batch 与单写者约束 DevFeed: [【etcd】bbolt 后端：mmap、batch 与单写者约束]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Databases](), [Raft]() Tags: [apply](), [backend](), [batch](), [bbolt](), [commit](), [distributed](), [etcd](), [kubernetes](), [mmap](), [mvcc](), [raft](), [v3-5](), [v3-5-33]() ### AI overview This article explains how etcd v3.5.33 maps its MVCC storage model onto the bbolt backend. It covers bucket layout, mmap behavior, buffered batch commits, read and write transaction concurrency, and how bbolt's single-writer constraint shapes apply and linearizable reads. ### Source excerpt 钉清 etcd v3.5.33 的 bbolt 后端：bucket 布局、mmap 预映射、100ms/10000 条 batch commit、ConcurrentReadTx 与单写者如何约束 MVCC 读写并发。 ## 【etcd】WAL 与快照：预写日志、crash recovery 与 commit vs applied DevFeed: [【etcd】WAL 与快照：预写日志、crash recovery 与 commit vs applied]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Raft](), [Kubernetes]() Tags: [applied-index](), [committed-index](), [crash-recovery](), [distributed](), [etcd](), [kubernetes](), [latency](), [persistence](), [safety](), [snapshot](), [v3-5-33](), [wal]() ### AI overview This tutorial explains how etcd v3.5.33 uses WAL segments and Raft snapshots for persistence, log compaction, and crash recovery. It also describes how committed and applied indexes can diverge, causing Watch delays or extended not-ready periods even when Raft replication appears healthy. ### Source excerpt 拆解 etcd v3.5.33 的 WAL 段文件、Raft snapshot 触发与重启 recovery 顺序；钉清 committed index 与 applied index 分叉对排障与 K8s 控制面的含义。 ## 【etcd】生产全景：缺口、五轴坐标系与 16 篇路线 DevFeed: [【etcd】生产全景：缺口、五轴坐标系与 16 篇路线]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: article Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Kubernetes]() Tags: [control-plane](), [distributed](), [etcd](), [kubernetes](), [lease](), [mvcc](), [overview](), [raft](), [v3-5](), [v3-5-33](), [wal](), [watch]() ### AI overview This article positions an etcd production series by identifying gaps in existing coverage and defining five troubleshooting axes spanning Raft consensus, WAL persistence and recovery, MVCC storage and capacity, Watch behavior, and Lease handling. It anchors the series to etcd v3.5.33 and points to Kubernetes control-plane coupling and a 16-article reading route. ### Source excerpt 相对 distributed/50、39、13 钉清 etcd 生产内核缺口；定义 Raft/WAL/MVCC/Watch/Lease 五轴排障坐标系，给出 16 篇阅读路线与 K8s 控制面耦合指针。版本锚定 v3.5.33。 ## 【etcd】Lease 与 KeepAlive：TTL、checkpoint 与 Leader 切换 DevFeed: [【etcd】Lease 与 KeepAlive：TTL、checkpoint 与 Leader 切换]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Raft](), [Kubernetes](), [Remote Procedure Call (RPC)]() Tags: [api](), [checkpoint](), [client](), [concurrency](), [distributed](), [etcd](), [follower](), [k8s](), [keepalive](), [kubernetes](), [lease](), [lessor](), [primary](), [raft](), [ttl](), [v3](), [v3-5-33](), [watch]() ### AI overview This article explains etcd v3.5.33 lease behavior, focusing on the lessor's primary, demote, and promote roles, the KeepAlive renewal path that normally bypasses Raft, lease checkpointing, and TTL changes during leader transitions. It also discusses implications for Kubernetes Node Leases and distributed locks. ### Source excerpt 钉 etcd v3.5.33 lessor 的 primary/demote/promote、KeepAlive 不经 Raft 的 Renew 路径、LeaseCheckpoint 与 Promote 时 TTL 展期，以及 Leader 切换对 Node Lease 的排障落格。 ## 【etcd】运维与升级：member change、backup/restore 与 3.5-\>3.6/3.7 门 DevFeed: [【etcd】运维与升级：member change、backup/restore 与 3.5-\>3.6/3.7 门]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [upgrade](), [Kubernetes](), [migration]() Tags: [backup](), [distributed](), [etcd](), [kubernetes](), [learner](), [member-change](), [migration](), [operations](), [raft](), [recovery](), [restore](), [snapshot](), [upgrade](), [v3](), [v3-5](), [v3-5-33](), [v3-6](), [v3-7]() ### AI overview A Chinese-language operational guide to etcd v3.5.33 covering member replacement and learner promotion, live snapshot backup and restore, Kubernetes control-plane recovery, patch upgrades, and documented upgrade gates from etcd 3.5 to 3.6 and 3.7. ### Source excerpt member add/remove/replace 与 learner promote 检查单；snapshot backup/restore 及 K8s 语境 revision bump；3.5.33 补丁线与 3.6/3.7 官方升级门禁项，不超前写未钉 tag 能力。 ## 【etcd】Watch 机制：watchableStore、synced/unsynced 与 ErrCompacted DevFeed: [【etcd】Watch 机制：watchableStore、synced/unsynced 与 ErrCompacted]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Raft](), [Remote Procedure Call (RPC)](), [Kubernetes]() Tags: [apiserver](), [apply](), [compaction](), [distributed](), [errcompacted](), [etcd](), [io](), [k8s](), [kubernetes](), [lag](), [leader](), [lease](), [mvcc](), [notify](), [raft](), [range](), [restore](), [revision](), [rpc](), [v3-5](), [v3-5-33](), [watch](), [watchablestore]() ### AI overview This article examines the etcd v3.5.33 watchableStore implementation, including synced and unsynced watcher groups, victims caused by backpressure, Apply-time notification, historical synchronization, and the ErrCompacted client resynchronization boundary. It also discusses event ordering, follower watch progress, and interactions with Kubernetes apiserver watch caches. ### Source excerpt 钉 etcd v3.5.33 watchableStore 的 synced/unsynced/victims 三分法、Apply 后 notify 与历史追赶 syncWatchersLoop，以及 CompactRevision 触发 ErrCompacted 时的客户端重同步边界。 [Next page]()