Author Archives: Costi
Author Archives: Costi
Company ABC has multiple buildings and two internet connections via 2 different ISPs. Both BR-B and BR-C receive a default route via eBGP from the ISPs and they inject it in OSPF. For some reason, when ISP-1 link goes down, entire Building-A looses the internet access ! Something must be wrong...
This post represents the solution and explanation for quiz-22. It presents how fragmented traffic is handled differently by a simple access list. It is a long read about fragmentation, Path MTU Discovery, MSS and other stuff...
Company ABC runs a static VTI-based VPN tunnel between Site-1, hosting 192.168.1.1, and Site-2, hosting 192.168.5.5. BGP is configured between the two sites, over the VTI Tunnel, making all traffic between the sites to be encrypted/protected by IPsec. The network engineer tries to configure QoS but something does not work !...
This post represents the solution and explanation for quiz-21. It is a very long post describing Pre-bestpath community, Point of Insertion, offset list and other networking hacks employed to tackle a less common problem. Make yourself a coffee and start reading...
This post represents the solution and explanation for quiz-19. As shown in the quiz, a simple change of the MSTP configuration, for example changing the vlan to instance mapping, can immediatelly create short network cuts. Read along to see full explanation.
Your company has 3 sites, each with a dedicated border router, R1, R2 and R3.
Site-1 (R1) and Site-2 (R2) have their own internet uplinks, but Site-3 (R3) connects to internet via R2. A GRE tunnel is built between R2 and R3 and applied an MTU of 1440, due to some constraints in the transit network between them. You notice that traffic between same pair of devices works for TCP 1001 but fails for TCP 1002. What's wrong ?
This post represents the solution and explanation for quiz-18.
It presents different MTU values, summarizes Path MTU Discovery and gives solutions to the quiz presented a while ago.
Hello and Welcome to my blog in 2014 !! I'll start this year by reviewing some of the most commented quizzes in the previous year. Read here to get the Top 5 most interesting quizzes in 2013.
You have just received a nice job at a big enterprise that has multiple sites connected over their own managed MPLS Core. Each site runs EIGRP as the CE - PE routing protocol. You get the task to route some traffic in a particular way, but you cannot make it. What is missing ?
This post represents the solution and explanation for quiz-17. For some temporary period of time, during network transition, your network consists both on Cisco and Juniper routers for the same role. You will see that they behave differently when it comes to advertising inactive BGP Routes.
Your company has a border router (R2) that is connected to two partner companies: Partner-DB (R1) providing database services and Partner-APP (R3) that provides different application services to your web servers in DMZ (200.200.200.0/24). You are requested to configure NAT according to some requirements.
This post represents the solution and explanation for quiz-16. Advertising inactive BGP routes can, sometimes, depend on other conditions. This article presents some scenarios with inactive BGP prefixes, suppress-inactive
and RIB-NH Matches.
As a senior network administrator, you receive complaints from server team that yesterday there were multiple short network cuts that impacted some very sensitive applications running in the data center. You investigate and find out that one of the level 1 network engineers performed some network changes. What went wrong?
This article is a continuation of previous post about RIP Auto-Summarization and it's impact on discontiguous networks in Cisco networks, but this time from Juniper's perspective. Using the default auto-summary
on Cisco devices can lead to routing loops in case of discontiguous networks, as shown in quiz 15.
This article discusses the solutions for quiz 15. Yes, I know, I know... it's about RIP ! But you need to be ready for anything when facing the challenges of a CCIE exam. Read along to review some things about auto-summarization (which is "on" by default).
Your company uses multi-vendor routing platforms (Cisco and Juniper) and has multiple sites connected via MPLS from a service provider. Each remote site has a GRE tunnel with the Headquarter (HQ) and a BGP session over this tunnel. After some security change in the network, sites that are Juniper-based behave differently than the Cisco-based ones, creating outage for the customer. What's wrong?
This article discusses the solutions for quiz 14 and it describes some scenarios to help understand why redistributing iBGP into IGP (OSPF in this case) is dangerous.
Your company decided to replace the existing Cisco devices with Juniper. The network team does that and configures the new Junipers as similar possible to the old Ciscos. But something does not work as intended. What is it ?