Intermittent VoIP drops and an IP conflict in an eight-seat office
An eight-seat IT-enabled services office in Ludhiana handles inbound customer calls over a VoIP softphone and uploads scanned documents to a client portal. The office has one broadband router with a 50 Mbps plan, one 8-port unmanaged switch, one network printer, and one Wi-Fi access point. Over two weeks, callers report dropped calls around 11 a.m. and 4 p.m., and scanned uploads fail with 'network timeout'. The office manager's laptop shows an IP conflict warning. A technician finds the router's DHCP pool set to 192.168.1.100–192.168.1.110, but 14 devices are connected, including staff mobile phones and a smart TV in the waiting area. One employee has plugged a personal travel router into the switch to improve Wi-Fi. The printer has a static address of 192.168.1.105. The client's portal requires a stable public IP and the office has no formal network diagram. The manager wants a low-cost fix this week.
1. Diagnose the likely cause of the IP conflict and the dropped VoIP calls, using the DHCP pool size, device count, and the printer's static address.
Read the worked answer
The DHCP pool has only 11 usable addresses (100–110), but 14 devices are connected. The printer's static address 192.168.1.105 falls inside that pool. When the router leases .105 to a laptop or phone, the printer and the laptop both claim the same address, producing the IP conflict warning. As the pool exhausts, new devices may fail to get a lease or self-assign a 169.254.x.x address, so the softphone cannot register reliably and uploads time out. The travel router is a second cause: it creates a second NAT and may run its own DHCP server, handing out addresses that conflict with the main LAN. Peak drops at 11 a.m. and 4 p.m. fit usage patterns - more staff arrive, mobiles join, and scans are uploaded at shift change - so the problem is contention, not a faulty broadband line. The unmanaged switch has no VLANs, so the smart TV and mobile traffic compete with VoIP. The absence of a diagram makes the rogue router harder to find.
2. Recommend a low-cost configuration change list for the router and switch, including DHCP range, reservations, Wi-Fi policy, and the personal travel router. Explain the order of implementation.
Read the worked answer
Order of implementation: (1) Unplug the personal travel router and tell staff that personal APs are prohibited; this removes the rogue DHCP/NAT immediately. (2) On the router, move the printer out of the DHCP pool: either set the printer to a static address below the pool (for example 192.168.1.20) or create a DHCP reservation, and reserve it so no lease is issued for that address. (3) Enlarge the pool to cover actual devices, e.g. 192.168.1.100–192.168.1.180, but exclude the printer's address; a /24 network has plenty of addresses. (4) Create a separate guest SSID for staff mobiles and the waiting-area TV, with client isolation and a bandwidth limit, so they cannot reach the office PCs or printer. (5) Enable QoS on the router for the softphone and the office PCs, giving VoIP priority over scans and video. (6) Draw a simple network diagram and label ports. Test by renewing leases on all devices, making a call, and uploading a large scan. The change is low-cost because it uses existing hardware.
3. The client requires a stable public IP. Explain whether the office should buy a static IP or use dynamic DNS, and what else could break VoIP quality even after the IP conflict is fixed.
Read the worked answer
The portal whitelists a stable public IP, so a static IP from the ISP is the safer choice; dynamic DNS can map a changing address to a name, but it adds a dependency and may still fail if the portal checks source IP. Buy the static IP if the monthly cost is acceptable and the contract depends on it. Even after the conflict is fixed, VoIP quality can still suffer from low upload bandwidth (a 50 Mbps plan often has only 5–10 Mbps upload), bufferbloat when scans upload, Wi-Fi interference and distance, jitter and latency on the broadband line, double NAT from any remaining router, and no QoS. The unmanaged switch cannot prioritise voice traffic, so a large scan can delay voice packets. Test with a ping/jitter test during peak hours, measure upload speed, connect the softphone by cable, and enable QoS or ask the ISP for a business plan with better upload and a static IP. Keep the TV and mobiles off the office SSID.