JN0-650 Premium File
- 107 Questions & Answers
- Last Update: Nov 8, 2026
When congestion begins, all packets stay in queue until buffer overflow, and an operator wants controlled early drops for a lower-priority TCP class. If the team bypasses or fails to implement the targeted correction (Configure a supported RED drop profile for the selected scheduler and packet-loss-priority category.), which direct operational risk or residual exposure is MOST relevant?
A tail-drop-only queue may create synchronized TCP retransmissions and large burst loss. This is the consequence linked to the observed situation: When congestion begins, all packets stay in queue until buffer overflow, and an operator wants controlled early drops for a lower-priority TCP class. It is prevented by: Configure a supported RED drop profile for the selected scheduler and packet-loss-priority category. The underlying reason is: Random early detection can drop packets probabilistically before total queue overflow.
Using identical loss treatment can discard important traffic as early as expendable packets. This risk belongs to a different operational problem: A queue should drop higher packet-loss-priority packets first during a transient congestion event. The stated condition instead exposes this direct risk: A tail-drop-only queue may create synchronized TCP retransmissions and large burst loss. The appropriate mitigation is: Configure a supported RED drop profile for the selected scheduler and packet-loss-priority category.
A correct class with excessive PLP can still suffer unnecessary early queue drops. This risk belongs to a different operational problem: Voice packets have high PLP after classification and are preferentially dropped despite a dedicated forwarding class. The stated condition instead exposes this direct risk: A tail-drop-only queue may create synchronized TCP retransmissions and large burst loss. The appropriate mitigation is: Configure a supported RED drop profile for the selected scheduler and packet-loss-priority category.
Accurate ingress marking without a suitable queue mapping still delivers the wrong forwarding behavior. This risk belongs to a different operational problem: A Junos interface maps traffic into the correct forwarding classes but the egress physical queue numbers are misassigned. The stated condition instead exposes this direct risk: A tail-drop-only queue may create synchronized TCP retransmissions and large burst loss. The appropriate mitigation is: Configure a supported RED drop profile for the selected scheduler and packet-loss-priority category.
Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. Which targeted engineering change is MOST directly justified by the observed evidence?
Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources. This change specifically addresses the reported evidence: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. The relevant protocol mechanism is: Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ. A decisive acceptance test would be: Routing-control packets and file traffic use separate output queues with intended priorities.
Correct the forwarding-class-to-output-queue associations for the supported platform and traffic policy. This is a valid solution to a different Junos situation: A Junos interface maps traffic into the correct forwarding classes but the egress physical queue numbers are misassigned. It does not directly correct the specific failure evidence here: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. The required targeted change is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.
Set an appropriate scheduler transmit-rate or shaping constraint and validate class bandwidth engineering. This is a valid solution to a different Junos situation: A high-priority queue consumes nearly the entire WAN and best-effort applications starve during sustained voice overload. It does not directly correct the specific failure evidence here: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. The required targeted change is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.
Apply an egress rewrite rule that maps internal forwarding class and PLP to the carrier's code points. This is a valid solution to a different Junos situation: An edge router reclassifies inbound DSCP values correctly, but its downstream carrier requires different DSCP bits on outgoing frames. It does not directly correct the specific failure evidence here: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. The required targeted change is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.
Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. The proposed correction is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources. Which observation MOST directly confirms this fault is resolved?
Routing-control packets and file traffic use separate output queues with intended priorities. This is the direct acceptance evidence for the observed failure: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. The mechanism being validated is: Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ.
Queue statistics show each traffic class consuming the intended output queue. That would validate a different condition: A Junos interface maps traffic into the correct forwarding classes but the egress physical queue numbers are misassigned. The decisive check for the current issue is: Routing-control packets and file traffic use separate output queues with intended priorities.
Voice retains its engineered priority while other classes obtain their planned minimum service. That would validate a different condition: A high-priority queue consumes nearly the entire WAN and best-effort applications starve during sustained voice overload. The decisive check for the current issue is: Routing-control packets and file traffic use separate output queues with intended priorities.
Outbound captures show the provider-required DSCP while the internal class decision stays unchanged. That would validate a different condition: An edge router reclassifies inbound DSCP values correctly, but its downstream carrier requires different DSCP bits on outgoing frames. The decisive check for the current issue is: Routing-control packets and file traffic use separate output queues with intended priorities.
A Junos engineer proposes this response: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources. The situation is: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. Which technical principle BEST justifies the response?
Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ. This is the controlling mechanism in the specific case: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. It predicts the acceptance result: Routing-control packets and file traffic use separate output queues with intended priorities.
Forwarding classes identify egress service queues that receive scheduled resources. This principle explains a separate situation: A Junos interface maps traffic into the correct forwarding classes but the egress physical queue numbers are misassigned. For this issue the relevant mechanism is: Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ.
Queue priority must be coupled with bounded resource allocation to avoid starvation. This principle explains a separate situation: A high-priority queue consumes nearly the entire WAN and best-effort applications starve during sustained voice overload. For this issue the relevant mechanism is: Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ.
Junos rewrite rules set outgoing header CoS bits using internal forwarding class and loss priority. This principle explains a separate situation: An edge router reclassifies inbound DSCP values correctly, but its downstream carrier requires different DSCP bits on outgoing frames. For this issue the relevant mechanism is: Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ.
Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. If the team bypasses or fails to implement the targeted correction (Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.), which direct operational risk or residual exposure is MOST relevant?
Naming two classes without separating output treatment can leave latency unchanged. This is the consequence linked to the observed situation: Two classes share one output queue even though operations expects different latency treatment for routing control and best-effort file transfers. It is prevented by: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources. The underlying reason is: Distinct forwarding class names matter only insofar as their queue mapping and scheduler differ.
Accurate ingress marking without a suitable queue mapping still delivers the wrong forwarding behavior. This risk belongs to a different operational problem: A Junos interface maps traffic into the correct forwarding classes but the egress physical queue numbers are misassigned. The stated condition instead exposes this direct risk: Naming two classes without separating output treatment can leave latency unchanged. The appropriate mitigation is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.
An unlimited priority allocation can deny useful capacity to the rest of the traffic classes. This risk belongs to a different operational problem: A high-priority queue consumes nearly the entire WAN and best-effort applications starve during sustained voice overload. The stated condition instead exposes this direct risk: Naming two classes without separating output treatment can leave latency unchanged. The appropriate mitigation is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.
Changing only an ingress classifier does not rewrite the packet headers transmitted to the peer. This risk belongs to a different operational problem: An edge router reclassifies inbound DSCP values correctly, but its downstream carrier requires different DSCP bits on outgoing frames. The stated condition instead exposes this direct risk: Naming two classes without separating output treatment can leave latency unchanged. The appropriate mitigation is: Assign supported distinct forwarding classes to appropriate queue numbers and scheduler resources.
An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. Which targeted engineering change is MOST directly justified by the observed evidence?
Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI. This change specifically addresses the reported evidence: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. The relevant protocol mechanism is: EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves. A decisive acceptance test would be: The remote leaf receives the intended Type 2 route and resolves the tenant MAC through the correct VTEP.
Verify the supported EVPN IP-prefix route model, including Type 5 where required by the design. This is a valid solution to a different Junos situation: A fabric control plane should recognize a subnet's tenant IP prefix for inter-subnet forwarding, but the team is examining only EVPN MAC route tables. It does not directly correct the specific failure evidence here: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. The required targeted change is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.
Inspect new Type 2 MAC advertisement and MAC mobility sequence handling after the move. This is a valid solution to a different Junos situation: A virtual machine moves to a different EVPN leaf, yet remote leaves continue using the stale previous VTEP for its MAC. It does not directly correct the specific failure evidence here: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. The required targeted change is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.
Verify EVPN Type 1 Ethernet autodiscovery withdrawal and affected aliasing or mass-withdraw convergence. This is a valid solution to a different Junos situation: A multihomed customer edge loses a PE uplink, yet remote ingress PEs keep forwarding toward withdrawn Ethernet segment reachability. It does not directly correct the specific failure evidence here: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. The required targeted change is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.
An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. The proposed correction is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI. Which observation MOST directly confirms this fault is resolved?
The remote leaf receives the intended Type 2 route and resolves the tenant MAC through the correct VTEP. This is the direct acceptance evidence for the observed failure: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. The mechanism being validated is: EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves.
Type 2 entries identify the host next hop while Type 3 supports BUM delivery. That would validate a different condition: A remote VXLAN leaf receives an EVPN Type 3 route for a VNI but no tenant host routes and the engineer assumes Type 3 can replace MAC reachability. The decisive check for the current issue is: The remote leaf receives the intended Type 2 route and resolves the tenant MAC through the correct VTEP.
VXLAN packets traverse the underlay and remote tenant unicast frames reach the destination. That would validate a different condition: An EVPN-VXLAN service has correct BGP Type 2 MAC routes but VXLAN data packets fail between leaves because UDP overlay transport is blocked by a security device. The decisive check for the current issue is: The remote leaf receives the intended Type 2 route and resolves the tenant MAC through the correct VTEP.
Remote PEs withdraw or reroute dependent MAC reachability promptly after the segment fails. That would validate a different condition: A multihomed customer edge loses a PE uplink, yet remote ingress PEs keep forwarding toward withdrawn Ethernet segment reachability. The decisive check for the current issue is: The remote leaf receives the intended Type 2 route and resolves the tenant MAC through the correct VTEP.
A Junos engineer proposes this response: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI. The situation is: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. Which technical principle BEST justifies the response?
EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves. This is the controlling mechanism in the specific case: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. It predicts the acceptance result: The remote leaf receives the intended Type 2 route and resolves the tenant MAC through the correct VTEP.
EVPN route targets constrain which tenant reachability a participating leaf imports. This principle explains a separate situation: A fabric leaf has a BGP EVPN session in Established but no MAC reachability for VNI 50100 while other VNIs work. For this issue the relevant mechanism is: EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves.
EVPN Type 5 can advertise IP prefixes for tenant inter-subnet routed overlay connectivity. This principle explains a separate situation: A fabric control plane should recognize a subnet's tenant IP prefix for inter-subnet forwarding, but the team is examining only EVPN MAC route tables. For this issue the relevant mechanism is: EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves.
VXLAN VNIs identify isolated logical overlay segments over a shared routed underlay. This principle explains a separate situation: Two tenants share the same physical VXLAN underlay but their Layer 2 broadcasts must remain separate despite overlapping inner VLAN numbers. For this issue the relevant mechanism is: EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves.
An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. If the team bypasses or fails to implement the targeted correction (Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.), which direct operational risk or residual exposure is MOST relevant?
Absent Type 2 reachability can force unnecessary flooding or prevent remote unicast delivery. This is the consequence linked to the observed situation: An EVPN-VXLAN leaf learns a local tenant MAC but remote VTEPs never learn that endpoint's MAC/IP binding from the control plane. It is prevented by: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI. The underlying reason is: EVPN Type 2 routes advertise MAC and optionally IP reachability information between participating leaves.
A working BGP session does not imply every VNI's EVPN routes are imported. This risk belongs to a different operational problem: A fabric leaf has a BGP EVPN session in Established but no MAC reachability for VNI 50100 while other VNIs work. The stated condition instead exposes this direct risk: Absent Type 2 reachability can force unnecessary flooding or prevent remote unicast delivery. The appropriate mitigation is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.
Expecting MAC-only routes to replace required tenant IP prefix exchange can break routed overlay traffic. This risk belongs to a different operational problem: A fabric control plane should recognize a subnet's tenant IP prefix for inter-subnet forwarding, but the team is examining only EVPN MAC route tables. The stated condition instead exposes this direct risk: Absent Type 2 reachability can force unnecessary flooding or prevent remote unicast delivery. The appropriate mitigation is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.
Reusing a tenant VNI inadvertently can merge unrelated broadcast domains. This risk belongs to a different operational problem: Two tenants share the same physical VXLAN underlay but their Layer 2 broadcasts must remain separate despite overlapping inner VLAN numbers. The stated condition instead exposes this direct risk: Absent Type 2 reachability can force unnecessary flooding or prevent remote unicast delivery. The appropriate mitigation is: Verify BGP EVPN Type 2 MAC/IP route origination and route-target import for the affected VNI.
Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. Which targeted engineering change is MOST directly justified by the observed evidence?
Check Type 3 inclusive-multicast EVPN routes and the configured BUM delivery method for that VNI. This change specifically addresses the reported evidence: Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. The relevant protocol mechanism is: EVPN Type 3 routes signal inclusive multicast membership for broadcast, unknown unicast, and multicast delivery. A decisive acceptance test would be: The intended remote VTEPs appear in the VNI's BUM replication or multicast distribution state.
Map each tenant broadcast domain to distinct VNIs and matching EVPN route-target membership. This is a valid solution to a different Junos situation: Two tenants share the same physical VXLAN underlay but their Layer 2 broadcasts must remain separate despite overlapping inner VLAN numbers. It does not directly correct the specific failure evidence here: Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. The required targeted change is: Check Type 3 inclusive-multicast EVPN routes and the configured BUM delivery method for that VNI.
Confirm that host MAC/IP advertisements use Type 2 routes in addition to Type 3 BUM membership. This is a valid solution to a different Junos situation: A remote VXLAN leaf receives an EVPN Type 3 route for a VNI but no tenant host routes and the engineer assumes Type 3 can replace MAC reachability. It does not directly correct the specific failure evidence here: Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. The required targeted change is: Check Type 3 inclusive-multicast EVPN routes and the configured BUM delivery method for that VNI.
Correct designated-forwarder election and non-DF egress suppression for that Ethernet segment and VLAN. This is a valid solution to a different Junos situation: A dual-homed customer edge receives duplicate broadcast frames when both active PEs forward remote BUM traffic toward the same segment. It does not directly correct the specific failure evidence here: Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. The required targeted change is: Check Type 3 inclusive-multicast EVPN routes and the configured BUM delivery method for that VNI.
Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. The proposed correction is: Check Type 3 inclusive-multicast EVPN routes and the configured BUM delivery method for that VNI. Which observation MOST directly confirms this fault is resolved?
The intended remote VTEPs appear in the VNI's BUM replication or multicast distribution state. This is the direct acceptance evidence for the observed failure: Two EVPN fabric leaves share a broadcast domain, but unknown-unicast and broadcast packets do not reach remote VTEPs. The mechanism being validated is: EVPN Type 3 routes signal inclusive multicast membership for broadcast, unknown unicast, and multicast delivery.
Traffic reaches only VTEPs participating in the appropriate tenant VNI. That would validate a different condition: Two tenants share the same physical VXLAN underlay but their Layer 2 broadcasts must remain separate despite overlapping inner VLAN numbers. The decisive check for the current issue is: The intended remote VTEPs appear in the VNI's BUM replication or multicast distribution state.
Type 2 entries identify the host next hop while Type 3 supports BUM delivery. That would validate a different condition: A remote VXLAN leaf receives an EVPN Type 3 route for a VNI but no tenant host routes and the engineer assumes Type 3 can replace MAC reachability. The decisive check for the current issue is: The intended remote VTEPs appear in the VNI's BUM replication or multicast distribution state.
Only the elected DF delivers remote BUM frames to the customer segment for the affected VLAN. That would validate a different condition: A dual-homed customer edge receives duplicate broadcast frames when both active PEs forward remote BUM traffic toward the same segment. The decisive check for the current issue is: The intended remote VTEPs appear in the VNI's BUM replication or multicast distribution state.
Please check your mailbox for a message from support@examlabs.com and follow the directions.