Tutorials · Troubleshooting
Why Is My Network Slow? A Step-by-Step Troubleshooting Checklist
A practical checklist network admins use to find the real cause of a slow network: scope, Internet link, Wi-Fi, switch errors, loops, DNS and the application itself.
Before you start
- Admin access to the gateway, switches and Wi-Fi controller or NavCloud
- A laptop with ping, MTR and iperf3
Step by step
Define the scope
Ask who, where, since when and which applications. One user, one floor, one site or everyone? Wired, Wi-Fi or both? The answer removes half the possible causes.
Check the Internet link
Look at WAN utilisation on the gateway. A link above 80% at peak times is congested: find the top talkers (backups, updates, cloud sync) and apply QoS or a bandwidth limit.
Check for errors on switch ports
CRC errors, input errors and half-duplex links slow everything on that port. Replace the cable or patch lead and set both ends to auto-negotiation.
Look for loops and broadcast storms
Very high broadcast traffic, MAC addresses flapping between ports and CPU at 100% on switches point to a loop. Make sure RSTP is enabled and storm control is set on access ports.
Check Wi-Fi health
Look at channel utilisation, signal strength (aim for -67 dBm or better) and client count per AP. Overlapping channels and too many clients on 2.4 GHz are the usual Wi-Fi bottlenecks.
Test DNS response time
Slow DNS makes every web page feel slow. A query time above 100 ms to your own DNS server needs attention.
dig www.example.com | grep 'Query time'Measure, do not guess
Use iperf3 between two wired points to prove the LAN is fast, then from the problem area. Use MTR to remote sites to find loss on the path.
Check the application
If the network tests are clean, look at the server: CPU, disk, database and the application logs. Many 'network' problems end here.
Commands are generic, industry-standard examples. Exact syntax and menu names vary by model and firmware; check your device's manual on the Documentation page.
Pro tips
- Keep one week of interface graphs (SNMP or NavCloud) so you can see when the slowdown started.
- Fix one thing at a time and re-test, so you know what solved it.
- Document the cause in the ticket: the same problem usually comes back.
FAQ
Frequently asked questions
Should I just buy a faster Internet line?
Only after you confirm the line is the bottleneck. Many slow networks are caused by Wi-Fi design, bad cables or loops, which a bigger line will not fix.
Keep learning
Related guides
Ping, traceroute and MTR
Read more →Prevent loops with RSTP
Read more →Monitor with SNMP and syslog
Read more →How to Troubleshoot Network Problems With Ping, Traceroute and MTR
Find where a connection breaks or slows down in minutes. Use ping for reachability and loss, traceroute for the path, and MTR to see loss and latency on every hop at once.
Read more →How to Capture and Read Network Traffic With Wireshark and tcpdump
When logs are not enough, look at the packets. Capture traffic with tcpdump or Wireshark, filter it down to the problem, and spot retransmissions, DNS failures, DHCP issues and TLS errors.
Read more →How to Test Network Speed Between Two Points With iperf3
Internet speed tests do not tell you how fast your LAN, Wi-Fi or site-to-site link really is. iperf3 measures throughput, loss and jitter between two machines you control.
Read more →Stuck on a step?
Ask Navijin technical support or an authorised partner for help with your configuration.