Networking with Linux (Chandan Dutta Chowdhury)（Z-Library）

【One-Line Pitch】 A ground-up, hands-on tour of how Linux actually implements networking—from interfaces and routing to firewalls, VPNs, container networking, and eBPF monitoring—for engineers who want to design, configure, and troubleshoot real Linux network stacks rather than memorize theory. 【Book Arc】 - **Opening (~0%–10%)**: Frames the whole project—OSI model as the organizing spine, Linux's role as the platform for routing, load balancing, VPNs, and container networking, and a chapter-by-chapter roadmap that sets expectations for a ground-up progression. - **Early (~10%–35%)**: Builds Layer 2 and Layer 3 fundamentals—physical and virtual interfaces, bridging, VLANs, link aggregation/bonding, IP addressing and subnetting, DHCP, routing, multipath and policy-based routing, VRF, and BGP—then moves into firewalls (iptables, NAT, ipsets, the shift to nftables and firewalld) and a full troubleshooting toolkit (ping, traceroute, netstat/netcat, tcpdump, Wireshark, NetFlow). - **Middle (~35%–60%)**: Shifts from packet plumbing to services and virtualization—load balancing (including Layer 7 with HAProxy), network namespaces and isolated stacks, MACVLAN/IPVLAN, TUN/TAP, Docker and Kubernetes pod networking with CNI plugins, and traffic shaping/QoS including buffer bloat. - **Late (~60%–85%)**: Tackles tuning and secure connectivity—sysctl/proc tuning of the TCP/IP stack, TCP Fast Open, keepalive, SYN-flood hardening, HTTP/3 and QUIC, then VPNs with WireGuard and IPsec/strongSwan (PSK vs. certificates, PKI), plus overlay networking (VXLAN, GRE), Open vSwitch/SDN, and cloud VPCs on AWS, Azure, and GCP. - **Ending (~85%–100%)**: Closes with modern observability—why traditional monitoring falls short, the eBPF program architecture, XDP, bpftrace, tracepoints vs. kprobes, socket-level tracing, and Tetragon—rounding out the design-configure-troubleshoot-manage arc promised in the preface. 【Key Takeaways】 - **The OSI model is the book's scaffolding, not decoration** (Opening): each layer is revisited in Linux terms—frames and MAC at Layer 2, packets and routing at Layer 3, ports and TCP/UDP at Layer 4—so later chapters build on a shared vocabulary. - **Layer 2 is where redundancy and segmentation live** (Early): bridging, VLANs, and bonding/LACP are treated as practical tools for fault tolerance and network separation, not just concepts. - **Routing goes well beyond static tables** (Early): equal-cost multipath, policy-based routing, VRF, and BGP-based dynamic routing show how Linux can act as a serious router. - **Firewalling is taught as a transition, not a single tool** (Early): iptables remains the enterprise standard, but the book deliberately introduces nftables and firewalld as the next generation—useful for anyone maintaining legacy rules while planning migration. - **Troubleshooting is a first-class skill with a layered toolkit** (Early): connectivity tests, socket inspection, and packet capture (tcpdump/Wireshark/TShark) plus NetFlow give a repeatable diagnostic workflow. - **Namespaces are the foundation of modern virtual networking** (Middle): isolated network stacks underpin MACVLAN/IPVLAN, TUN/TAP, Docker, and Kubernetes pod networking—understanding them explains container networking rather than just configuring it. - **Bandwidth is a finite resource to be managed** (Middle): QoS and traffic shaping, queuing disciplines, rate limiting, and buffer bloat mitigation frame performance as an engineering trade-off. - **Tuning and security meet at the TCP/IP stack** (Late): sysctl/proc knobs, TCP Fast Open, keepalive, and SYN-flood hardening connect throughput goals with resilience against attack. - **Observability is evolving toward eBPF** (Ending): the book argues traditional monitoring has limits and demonstrates XDP, bpftrace, and custom eBPF programs for tracing and statistics. 【Reading Tips】 - **Deep-read Chapters 1–3 and 7** if you're new to Linux networking: OSI mapping, routing, firewalls, and namespaces are the load-bearing concepts everything else references. - **Skim the command-syntax sections** (iptables rules, tcpdump filters, ip commands) on first pass, then return with the code bundle open—this is a configure-along book, not a cover-to-cover novel. - **Treat the troubleshooting chapter as a reference** you'll revisit during real incidents; the filter-structure breakdown for tcpdump and the traceroute limitations are worth bookmarking. - **Expect the hardest going in the late chapters** (IPsec/PKI, Open vSwitch/SDN, eBPF internals)—budget extra time and pair them with the GitHub code bundle. - **Take away the mental model, not just commands**: for each topic, ask "which layer, which namespace, which tool"—that framing is what the book is really teaching. 【Coverage Limits】 This guide is built from the front matter, preface, table of contents, and early chapter excerpts; the excerpts do not cover the detailed body of later chapters (load balancing, VPN configuration steps, cloud VPC specifics, eBPF code), so chapter-level claims are drawn from the author's own roadmap rather than full chapter text.