Network O&M and Troubleshooting · Topic 3 of 3

Network Troubleshooting Process and Common Cases

Learn the network troubleshooting process, methods and VRP tools for H12-811, plus common cases: interface down, VLAN, IP, routing, OSPF and loop faults.

5 min
Includes a diagram

In a nutshell

Process: collect information, analyze, locate, rectify, verify, document.
Methods: bottom-up, top-down, divide and conquer, segmented, replacement, comparison.
VRP uses tracert, not traceroute, and display, not show.

What it is

Network troubleshooting is the structured work of finding why a network service fails and restoring it. A fixed process and a few methods stop you from changing things at random.

On Huawei devices the main tools are ping, tracert, display commands, logs and alarms. Most HCIA cases are simple: an interface down, a wrong VLAN, a missing route or an OSPF neighbor that will not come up.

How it works

The troubleshooting process

Troubleshooting process: collect information, analyze possible causes, locate the fault, rectify it after a backup, verify, and document; if the fix fails, roll back and return to analysis.

Step	What you do
Collect information	Symptoms, scope (one user or many), when it started, recent changes, topology, logs and alarms
Analyze	List the possible causes that fit the symptoms
Locate the fault	Test the causes in order until one is confirmed
Rectify	Fix the cause; back up the configuration first and have a rollback plan
Verify	Check the original service works and nothing else broke
Document	Record the fault, cause and fix for future cases

If the fix does not work or makes things worse, roll back and return to analysis.

Troubleshooting methods

Method	How it works
Layered, bottom-up	Start at the physical layer and move up: cable, interface, VLAN, IP, routing, application
Layered, top-down	Start at the application and move down, useful when lower layers are clearly fine
Divide and conquer	Start in the middle, for example with `ping`, and go up or down depending on the result
Segmented	Split the path into segments (access, aggregation, core, egress) and test each
Replacement	Swap a suspect cable, module or device with a known good one
Comparison	Compare with a working device or a saved good configuration

VRP tools

Command	Use
`ping 10.1.1.1`	Reachability; `-a` sets the source address, `-c` the count, `-s` the packet size
`tracert 10.1.1.1`	Shows each hop, to find where packets stop
`display interface brief`	PHY and Protocol state of all interfaces
`display vlan`, `display port vlan`	VLAN membership and port link types
`display ip routing-table`	Whether a route exists and its protocol and next hop
`display ospf peer brief`	OSPF neighbor states
`display stp brief`	STP port roles and states
`display logbuffer`, `display trapbuffer`	Device logs and alarms
`display current-configuration`	The running configuration

<Huawei> display interface brief
Interface                   PHY   Protocol  InUti OutUti   inErrors  outErrors
GigabitEthernet0/0/1        up    up           0%     0%          0          0
GigabitEthernet0/0/2        *down down         0%     0%          0          0

*down means the interface is administratively down: someone entered shutdown. Plain down in the PHY column points to a physical problem.

Common network troubleshooting cases

Symptom	Likely causes	Check and fix
Interface PHY `*down`	Interface shut down	`undo shutdown` in the interface view
Interface PHY `down`	Cable, optical module, peer down, rate or duplex mismatch	Check cabling and modules, align speed and duplex settings
PCs in the same VLAN cannot ping across switches	Access port in the wrong VLAN, trunk not passing the VLAN	`display port vlan`; fix with `port default vlan` or `port trunk allow-pass vlan`
One host loses connectivity at random	IP address conflict	Logs and `display arp`; give the host a unique address
Ping to a remote subnet fails	Missing route or missing return route	`display ip routing-table` on each hop; add a static route or fix routing
OSPF neighbor never reaches Full	Area ID, hello or dead interval, authentication, or mask mismatch, or duplicate router ID	`display ospf peer brief`, compare both ends' interface settings
Network slow, CPU high, MAC addresses flapping	Layer 2 loop, STP disabled	`display stp brief`, enable STP, remove the loop link

OSPF defaults on broadcast and P2P networks: hello 10 seconds, dead 40 seconds. A mismatch on either stops the neighbor forming.

A worked example

Users on VLAN 20 at SW2 cannot reach the server on VLAN 20 at SW1. Both access ports are correct in display port vlan. On the link between SW1 and SW2, display port vlan shows the trunk passes only VLAN 1 and VLAN 10.

The engineer runs port trunk allow-pass vlan 20 on both ends of the trunk. A ping from a VLAN 20 PC to the server now succeeds, and the cause and fix are recorded.

How the exam asks about this

"In display interface brief, the PHY state of an interface is *down. What is the cause?" → The interface was shut down with the shutdown command.
"Which troubleshooting method starts by testing with ping and then moves to the upper or lower layers?" → Divide and conquer.
"Two directly connected routers do not form an OSPF neighbor relationship. Which settings could be mismatched?" → The hello or dead interval, the area ID or the authentication settings; each makes the routers discard the other's hello packets.

What to remember for the exam

Process: collect information, analyze, locate, rectify, verify, document.
Methods: bottom-up, top-down, divide and conquer, segmented, replacement, comparison.
VRP uses tracert, not traceroute, and display, not show.
*down is administratively down; plain down is a physical or link problem.
OSPF neighbors fail on mismatched area, timers, authentication or mask, or a duplicate router ID.

Watch out

Back up the configuration before you change anything during troubleshooting, and keep a rollback plan. A fix applied without one can turn a small fault into an outage.

Common mistakes

Changing configuration before collecting facts. Random changes hide the real cause and can add new faults. Collect information and check recent changes first.

Forgetting the return path. A ping needs a route back as well as a route there. A missing return route on the far device looks exactly like a missing forward route.

Checking only the access port. In a multi-switch VLAN fault, the trunk links in between must also allow the VLAN.

Check yourself

What does ping -a do on a VRP device?
It sets the source IP address of the ping packets, useful for testing a specific return route.
Which command shows OSPF neighbor states?
display ospf peer brief.
What symptoms suggest a Layer 2 loop?
Very high CPU usage, a slow network, broadcast storms and MAC addresses flapping between ports, often because STP is disabled.

Topic 3 of 3 in Network O&M and Troubleshooting

Test yourself on Network O&M and Troubleshooting

The full Huawei HCIA-Datacom H12-811 question bank has 600 questions, with timed practice exams and an explanation for every answer.

Subscribe Try Sample Exam

This study guide is independent and is not affiliated with, endorsed by, or sponsored by Huawei. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.