iBGP против eBGP

Поддерживать стабильную работу сетевой инфраструктуры крупного предприятия — задача непростая. Взаимодействие промышленных и корпоративных сетевых сред на предприятиях требуют гибкого подхода к сетевым архитектурам. Именно здесь особенно важна грамотно спроектированная маршрутизация и устойчивость соединений. Правильный выбор BGP-решений, будь то внутренний iBGP или внешний eBGP, оказывает значительное влияние на производительность и отказоустойчивость сетевой инфраструктуры. В этой статье мы рассмотрим протоколы iBGP и eBGP и поговорим о том, в каких случаях будет предпочтительнее использование каждого из них. Но для начала давайте разберемся с основными понятиями. iBGP vs eBGP

https://habr.com/ru/companies/otus/articles/1001474/

#bgp #ibgp #ebgp #автономная_система #маршрутизация #маршруты #отказоустойчивость

iBGP против eBGP

Поддерживать стабильную работу сетевой инфраструктуры крупного предприятия — задача непростая. Взаимодействие промышленных и корпоративных сетевых сред на предприятиях требуют гибкого подхода к...

Хабр

I've been working on building a #VPP-based router for the last year to support a small ISP I've been partnering with. I recently began polishing it (e.g., building an image that can be installable from USB, etc.) and adding a unified configuration CLI that abstracts away the details of managing multiple VPP processes, FRR, ZFS snapshots, (eventually) immutable system, etc.

This weekend I began experimenting with adding local #LLM capabilities (using #Ollama with gpt-oss:120b running on an AMD AI MAX 395+ system with 128GB of unified RAM) and I'm really excited by the possibilities.

imp> agent [+] Connected to Ollama (gpt-oss:120b) Type your request, or 'exit' to return agent> Let's add a new loopback interface with IPv4 10.5.5.1/32 and IPv6 fc00::5:5:1/128. It can be called "service". Thinking... [Tool: add_loopback] create_lcp: True ipv4_cidr: 10.5.5.1/32 ipv6_cidr: fc00::5:5:1/128 name: service → Added loop0 (service) with 10.5.5.1/32, fc00::5:5:1/128 Thinking... The loopback interface loop0 (named “service”) has been created with: • IPv4 10.5.5.1/32 • IPv6 fc00::5:5:1/128 Let me know if you’d like any further changes! agent> exit Returning to IMP REPL... imp*> apply [+] Configuration saved to /persistent/config/router.json [+] Configuration files generated in /tmp/imp-generated-config Changes detected: Detected 1 configuration change(s): - ADD loopback loop0 [DRY-RUN] VPP Core changes (5 commands): create loopback interface instance 0 set interface state loop0 up lcp create loop0 host-if lo0 set interface ip address loop0 10.5.5.1/32 set interface ip address loop0 fc00::5:5:1/128 Apply changes live? [Y/n]: y [+] Applying changes live... Detected 1 configuration change(s): - ADD loopback loop0 Applying VPP Core changes... [+] OK (5 commands) [+] Live changes applied successfully imp>

…and watching it come up (and be automatically handled by #iBGP since I'm pinging from a host that's not directly connected).

> ping 10.5.5.1 PING 10.5.5.1 (10.5.5.1): 56 data bytes Request timeout for icmp_seq 0 Request timeout for icmp_seq 1 Request timeout for icmp_seq 2 Request timeout for icmp_seq 3 Request timeout for icmp_seq 4 Request timeout for icmp_seq 5 Request timeout for icmp_seq 6 Request timeout for icmp_seq 7 Request timeout for icmp_seq 8 Request timeout for icmp_seq 9 64 bytes from 10.5.5.1: icmp_seq=10 ttl=63 time=3.067 ms 64 bytes from 10.5.5.1: icmp_seq=11 ttl=63 time=3.132 ms 64 bytes from 10.5.5.1: icmp_seq=12 ttl=63 time=3.128 ms 64 bytes from 10.5.5.1: icmp_seq=13 ttl=63 time=4.260 ms 64 bytes from 10.5.5.1: icmp_seq=14 ttl=63 time=3.189 ms 64 bytes from 10.5.5.1: icmp_seq=15 ttl=63 time=3.992 ms
Enigmatick

Does this make sense? Like would an IPv6 Router Advertisement contain essentially other internal gateways on the layer 3 network?

I'm thinking that not all clients should have to run BGP and peer just to get the gateways to the other subnets.

Looks like there's no way to wire up FRR to the RA options in OPNsense, and I'm not sure RAs support this (though it seems logical).

#Networking #HomeLab #IPv6 #OPNsense #iBGP #BGP

After a fair bit of tinkering in between things I’ve finally gotten my BGP bits sorted properly and sending announcements up to my peer, some further filters adjusted in the peers link to the carriers and I’m now routable to the world!

#bgp #mikrotik #juniper #AS214064 #44net #ardc #ampr #routing #IPv4 #IPv6 #eBGP #iBGP #OSPF #dualstack

Confederaciones en BGP - Eduardo Collado

Para reducir el número de enlaces para el full-mesh en iBGP podemos hacer dos cosas, utilizar route reflectors o bien utilizar confederaciones. Hoy hablamos de confederaciones. Introducción a BGP y…

Eduardo Collado
I ended up just doing it the normal way of configuring everything in /etc/network/interfaces. Away with all the "magic" of dhcpcd and whatever else was trying to mess with my networking config. Done! My #RaspberryPi is now happily exchanging #OSPF and #IBGP routes with my other hosts over #WireGuard. Yay!