Free HCIA-Datacom V1.0 Exam Flashcards

The sender adds headers as data moves down the stack: TCP creates a segment or UDP creates a datagram, IP wraps it in a packet, and Ethernet wraps the packet in a frame before bits cross the medium. The receiver removes those headers in reverse order. This layered process lets each protocol handle a separate job.

It resolves the MAC address of its local default gateway, not the remote host. The destination IP remains the remote host's address, while the Ethernet destination changes hop by hop. ARP resolves an IPv4 next-hop address only on the local Layer 2 segment.

Favor TCP when ordered, reliable delivery and congestion control matter; TCP establishes a connection and retransmits lost data. Favor UDP when low overhead and timeliness matter more than built-in recovery, such as real-time voice or simple queries. Applications using UDP must add any required reliability themselves.

The user view, shown with an angle-bracket prompt such as <Huawei>, is mainly for monitoring and file-level operations. Enter system-view to reach the bracketed system view, such as [Huawei], where configuration begins. Use quit to move back one level and save from user view to preserve the configuration.

A /27 leaves 5 host bits, producing 32 addresses. In a conventional subnet, the network and broadcast addresses are not assigned to hosts, leaving 30 usable host addresses. The block size in the final octet is 32.

The /27 blocks advance by 32, so 77 lies in the 64–95 block. The network address is 192.168.10.64, the broadcast address is 192.168.10.95, and usable hosts run from 192.168.10.65 through 192.168.10.94.

It applies its subnet mask to both its own IPv4 address and the destination. Matching network portions mean the destination is local, so the host resolves the peer's MAC address. Different network portions mean the host resolves the default gateway's MAC address and sends the frame there.

The router first chooses the longest prefix match—the route with the most specific mask. A /24 therefore beats a matching /16, even if the /16 came from a normally preferred source. Preference and metric resolve choices among routes to the same prefix length and destination.

Preference compares routes learned from different routing sources; the lower preference is favored. Metric compares paths learned by the same routing protocol according to that protocol's rules. Longest-prefix matching still occurs before either comparison when prefixes differ.

VRP installs a direct route when an interface has an address and the interface protocol is up. The route represents the attached network and provides the foundation for reaching next hops on that link. If the interface goes down, the direct route normally disappears, which can also invalidate dependent static routes.

It creates a static route to 192.0.2.0/24 using 198.51.100.1 as the next hop. The next hop must itself be reachable through the routing table. Static routes provide deterministic forwarding but do not automatically discover an alternate path.

It is 0.0.0.0/0 and can be configured on VRP with `ip route-static 0.0.0.0 0.0.0.0 <next-hop>`. Because it is the least-specific route, it is used only when no more-specific entry matches the destination. It is common on a stub network with one exit.

A route may name a next-hop IP address without naming the final outgoing interface. The router performs another routing-table lookup for that next hop until it finds a directly reachable path and interface. If the recursive path vanishes, the original route cannot forward traffic even if its configuration remains.

They need compatible area IDs, authentication settings, hello/dead timers, and network types on the shared link. Their interfaces must also have usable Layer 3 connectivity and OSPF enabled. Checking these parameters is more useful than repeatedly restarting the protocol.

The designated router and backup designated router reduce the number of full adjacencies and link-state exchanges required on a multiaccess segment. Other routers form full adjacencies primarily with the DR and BDR rather than with every peer. This limits control-plane overhead as the segment grows.

Area 0 is the OSPF backbone. In a multi-area design, nonbackbone areas exchange inter-area routes through area border routers connected to the backbone. The hierarchy limits link-state flooding and SPF scope while preserving connectivity between areas.

Routers advertise link-state information in LSAs, build a synchronized link-state database for the area, and run the shortest path first algorithm with themselves as the root. The resulting best paths are installed in the routing table when they beat competing routes under the route-selection rules.

Verify interface and IP status, confirm OSPF is enabled on the intended interface and area, inspect neighbor state, compare adjacency parameters, then check the LSDB and routing table. This order follows the dependency chain: a route cannot appear until the link, adjacency, and topology information are healthy.

For each received frame, the switch records the source MAC address against the ingress port and VLAN. It does not learn from the destination address. Dynamic entries age out so the table can adapt when a device moves.

It sends a known unicast only through the learned destination port. It floods an unknown unicast and a broadcast through eligible ports in the same VLAN except the ingress port. VLAN membership therefore defines the Layer 2 flooding boundary.

A VLAN creates a logical Layer 2 broadcast domain independent of physical switch placement. Devices in different VLANs do not exchange Layer 2 broadcasts and require Layer 3 routing to communicate. VLANs improve segmentation, failure containment, and policy control.

Use an access port for an endpoint that belongs to one VLAN and normally sends untagged Ethernet frames. The switch associates ingress traffic with the port's VLAN and sends endpoint-facing frames untagged. This hides VLAN tagging from ordinary hosts.

A trunk carries traffic for multiple VLANs across one physical link, identifying VLANs with tags according to the port configuration. Both ends must permit the required VLANs and agree on handling of untagged traffic. A missing allowed VLAN can look like a higher-layer connectivity failure.

The port VLAN ID classifies an incoming untagged frame into a VLAN. It does not by itself define every egress tagging decision; that also depends on whether the port is access, trunk, or hybrid and how VLAN membership is configured. PVID mismatches can place untagged traffic in different VLANs at opposite ends.

They need a Layer 3 gateway for each VLAN and routing between those VLAN interfaces. Hosts use the gateway address in their own subnet; the Layer 3 device routes the packet and rewrites the Layer 2 header for the destination VLAN. A trunk alone does not provide inter-VLAN routing.

The switch with the lowest bridge ID becomes root. Bridge ID comparison considers bridge priority first and then MAC-address information as the tiebreaker. A deliberate lower priority is preferable to letting an arbitrary switch become root.

Each nonroot switch selects one root port that offers its best path toward the root bridge. Each network segment selects one designated port that offers the segment's best path toward the root. Other redundant ports are prevented from forwarding when necessary to break loops.

RSTP uses explicit port roles and rapid proposal/agreement behavior instead of relying only on legacy timer progression. An alternate port can provide a preselected backup path toward the root and transition quickly when the active path fails, provided the topology and edge-port settings are correct.

Ethernet frames have no hop limit, so broadcasts and unknown unicasts can circulate repeatedly. The result may include a broadcast storm, duplicate frames, and rapid MAC-table flapping between ports. STP or RSTP blocks redundant paths while retaining them for recovery.

An Eth-Trunk combines compatible physical links into one logical link. Traffic can be load-shared across members, and surviving members can continue carrying traffic after one link fails. The logical bundle also presents one topology object to protocols such as STP.

LACP lets peers negotiate and monitor which links actively participate in an aggregate. It can detect one-sided or incompatible configurations and retain standby members. Normally, only compatible full-duplex links with aligned service settings forward together; platform features such as mixed-rate LACP are explicit exceptions.

Stacking combines multiple physical switches into one logical management and control entity. An Eth-Trunk combines links, not device control planes. A design may use both: stacking for device-level simplification and an Eth-Trunk for link bandwidth and resiliency.

An ACL compares traffic using fields such as source, destination, and protocol. Huawei ACLs can use configuration order or automatic ordering, so verify the active match order as well as direction, attachment point, address masks, and hit counts. Merely creating an ACL does not change traffic until a feature applies it.

Authentication verifies identity, authorization determines what the verified identity may do, and accounting records activity or resource use. Keeping the functions distinct supports centralized access control and auditability. A login can authenticate successfully yet still be denied an unauthorized command.

Source NAT rewrites an inside source address, commonly so private hosts can reach an outside network. Easy IP is a Huawei source-NAT method that uses the outbound interface's current IP address, which is useful when that address is dynamically assigned. Port translation allows multiple sessions to share an address.

It publishes an inside service by mapping a reachable external address and, when configured, port to an internal server. Return traffic is translated consistently for the session. The mapping is not a substitute for an ACL or security policy; reachability and access permission must both be checked.

The client broadcasts Discover, a server responds with Offer, the client sends Request, and the server completes the lease with Acknowledgment—often remembered as DORA. The lease can also carry the mask, default gateway, DNS servers, and other options.

Initial DHCP client messages are broadcasts, and routers do not normally forward Layer 2 broadcasts. A relay receives the client message, sends it to a configured server as routable traffic, and identifies the client-side subnet so the server can select the correct address pool.

DNS resolves names, NTP synchronizes time, and HTTP transfers web resources. FTP provides a session-oriented file-transfer service, while TFTP is a simpler file-transfer protocol with fewer features. Telnet provides a plaintext remote terminal, so it should not be treated as a secure management channel.

The station is the wireless client, the access point provides radio access, and the access controller centrally manages AP behavior and WLAN services. Depending on the design, user data may be tunneled to the AC or forwarded locally, while control remains centralized.

The SSID is the human-visible WLAN service name that multiple AP radios may advertise. A BSSID identifies a particular basic service set, commonly corresponding to one AP radio's MAC identity for that WLAN. A roaming client can keep the same SSID while moving between BSSIDs.

The client first discovers candidate networks by scanning, then selects an AP, performs 802.11 authentication, and associates. Security negotiation and address acquisition follow as required by the WLAN. A visible SSID alone therefore does not prove that authentication, association, or IP connectivity will succeed.

A wireless station cannot reliably listen for a collision while transmitting and may not hear every competing station. It therefore senses the channel, waits, and uses a randomized backoff to reduce collision probability. Acknowledgments help reveal whether a frame was delivered.

CAPWAP provides tunnels between an AP and an access controller for management and control, and it can also carry user data in a tunneled-forwarding design. If an AP cannot join the AC, check basic IP reachability, controller discovery, time or certificate dependencies, and CAPWAP status before troubleshooting the SSID.

LCP establishes, tests, and negotiates the data link. Optional PAP or CHAP verifies the peer, with CHAP avoiding transmission of a reusable plaintext password. NCP then configures the carried network-layer protocol before normal payload traffic flows.

PPPoE carries PPP sessions over Ethernet, often for subscriber access. MPLS forwards traffic through a label-switched domain using labels rather than repeated destination-IP lookups at every hop. Segment Routing steers a packet through an ordered set of segment instructions and can reduce per-flow state in the core.

An SNMP manager polls agents to read managed objects in their MIBs, providing scheduled visibility and confirming device reachability. An agent sends a trap asynchronously when a notable event occurs. Polling supplies regular state; traps provide timely notification but should not be the only monitoring method.

IPv6 addresses are 128 bits and use hexadecimal notation. Link-local unicast addresses come from FE80::/10 and are used only on the local link, including important neighbor and routing-protocol exchanges. IPv6 supports unicast, multicast, and anycast but does not use broadcast.

Neighbor Discovery uses ICMPv6 messages such as router solicitation/advertisement and neighbor solicitation/advertisement. It discovers routers, resolves neighboring link-layer addresses, and checks reachability. Router advertisements can support SLAAC, while DHCPv6 may provide stateful addresses or other configuration depending on the flags and design.

SDN separates forwarding from control and makes control logically centralized through a controller with northbound and southbound interfaces. NFV runs network functions as software rather than tying them to dedicated appliances. Automation uses repeatable models, APIs, and scripts to apply and verify changes. For exam scope, stay at the outline's stated level: basic SDN and NFV concepts, automatic O&M, and Python basics.