Skip to content

CHALLENGES.md

Latest commit

 

History

History
72 lines (48 loc) · 2.54 KB

CHALLENGES.md

File metadata and controls

72 lines (48 loc) · 2.54 KB
Raw

🧠 Challenges & Solutions

Every bump we hit while designing and verifying the network — and exactly how we engineered around it.

1. 🚧 Duplicate IPs between distribution switches (VLAN 30)

Problem: Both DIST-SW1 and DIST-SW2 wanted to own 10.10.30.2, collapsing the SVI.

Fix: Split the addresses (DIST-SW1=.2, DIST-SW2=.3) and let HSRP present a single virtual gateway (10.10.30.1) to end hosts. DHCP scope was re-aligned to that virtual IP.

2. 🚧 HSRP Active / Standby out of sync

Problem: Without priority tuning both cores could try to be Active and flip-flop.

Fix: Hard-set standby X priority 110 on DIST-SW1 and add preempt on both sides so the higher-priority switch always wins back after recovery. Verified by powering off CORE-R1 mid-call.

3. 🚧 Native VLAN mismatch on trunks

Problem: Mixing VLAN 1 and VLAN 100 as native caused 802.1Q tagging glitches and rogue broadcasts across the fabric.

Fix: Standardised on VLAN 100 as native everywhere, validated with show interface trunk on every link.

4. 🚧 Inter-VLAN routing silent failure

Problem: Forgetting ip routing or the per-VLAN SVI silently disabled all cross-VLAN traffic.

Fix: Enabled ip routing, built an SVI for every VLAN, and validated end-to-end with ping from each department.

5. 🚧 NAT on the wrong interface

Problem: Tagging the inside/outside backwards broke Internet access for every department.

Fix: Explicit ip nat inside on Gi0/0/Gi0/1 and ip nat outside on the serial link; ACL 1 permit 10.10.0.0 0.0.255.255 on the overload rule.

6. 🚧 VoIP one-way / no dial-tone

Problem: Phones wouldn't register because the Voice VLAN wasn't reachable.

Fix: Unify VLAN 150 across every switch, add HSRP for the voice gateway, and tag phones via switchport voice vlan 150 on every access port.

7. 🚧 Single Point of Failure on the distribution layer

Problem: Originally every access switch had only one uplink.

Fix: Added a second distribution switch (DIST-SW2) with parallel uplinks, enabled EtherChannel Po1 on the Gigabit uplinks, and validated failover by yanking one uplink mid-stream — no drops.

💬 Personal note from the team

"Designing this network taught us that networking is not just plugging in cables — it's the art of engineering availability and stability. Every fix here was tested live in Packet Tracer to guarantee the highest level of performance and reliability." — Alhareth Hayel & Ayman Bashi Al-Baidhani