Cisco SPRI 300-510 cheat sheet

OSPFv2 external metrics: type 1 adds the redistributed metric to the internal path cost; type 2 (default) uses the external metric only, with internal cost as tiebreaker.

OSPF forwarding address in type 5 LSAs is set to 0.0.0.0 when the next hop is the advertising ASBR itself or the outgoing interface is passive/suppressed; nonzero means lookup is done through that address.

OSPF NSSA: type 7 LSAs are translated to type 5 at the NSSA ABR elected translator (highest router ID by default); type 7 P-bit set on redistribution by an ASBR allows translation.

OSPF area types: stub blocks type 5 (default route injected by ABR); totally stubby also blocks type 3; NSSA blocks type 5 but allows type 7; totally NSSA blocks type 3 and 5.

IS-IS levels: L1 uses level-1 LSPs within an area, L2 uses level-2 LSPs in the backbone; an L1/L2 router sets the ATT bit in its L1 LSP to signal an L2 exit, and L1 routers install a default route toward it.

IS-IS default metric is narrow (6-bit, max 63); wide metrics (up to 2^24) are required for TE and are enabled per level with metric-style wide.

IS-IS LSP purge requires the remaining lifetime to reach zero; zero-age lifetime is flooded to purge; overload bit (OL) prevents transit traffic but still allows local prefixes.

Redistribution loop prevention: tag routes at the redistribution point and apply route-maps that deny matching tags on the opposite redistribution; never redistribute one protocol into another bidirectionally without filtering.

Seed metric is mandatory for redistributed routes: OSPF requires metric (default 20 for BGP, 20 for connected/static via default-metric); IS-IS requires metric or default-metric; EIGRP requires bandwidth/delay/MTU.

BGP path selection order (Cisco): highest weight, highest local preference, locally originated (network/aggregate/redistribute), shortest AS_PATH, lowest origin (IGP<EGP<incomplete), lowest MED, eBGP over iBGP, lowest IGP metric to next hop, oldest route, lowest router ID, lowest neighbor address.

Weight is Cisco-proprietary and local to the router; local preference is propagated within the AS via iBGP but not across eBGP.

MED is propagated to neighboring ASes only; by default it is compared only among paths from the same neighboring AS unless bgp always-compare-med is set.

iBGP split horizon: a route learned from an iBGP peer is not advertised to another iBGP peer; full mesh, route reflectors, or confederations are required.

Route reflector rules: routes from a client are reflected to clients and non-clients; routes from a non-client are reflected to clients only; cluster ID and originator ID are used for loop prevention.

Confederation: sub-AS numbers are used internally, the confederation ID is presented to external peers; AS_PATH length across the confederation is not counted for external path selection.

BGP neighbor states: Idle, Connect, Active, OpenSent, OpenConfirm, Established; stuck in Active usually means TCP port 179 unreachable or an ACL/unicast reverse-path issue.

Next-hop reachability: iBGP requires an IGP route to the next hop or next-hop-self on the advertising router; eBGP multihop requires update-source loopback and a route to the loopback.

Common troubleshooting checks: show ip bgp summary (state/PfxRcd), show ip bgp neighbors (capabilities, timers), show ip bgp (locally originated), show ip bgp (best path reason).

LDP binds a label to each IGP prefix (FEC) and distributes it to LDP peers over TCP 646; the label is locally significant and swapped hop by hop along the LSP.

LDP session requires matching LDP router ID reachability and transport address; mismatched transport addresses cause session flaps or no session.

MPLS TE: RSVP-TE (protocol 46) signals label-switched paths; explicit-path and affinity/attribute flags constrain path selection; CSPF prunes links failing constraints before running shortest path.

MPLS TE bandwidth is reserved with the setup/hold priority model; a lower numeric priority is more important, and preemption occurs when a new LSP cannot be placed.

Segment Routing: an SRGB (Segment Routing Global Block) maps prefix SIDs; index + SRGB start = global label; the SRGB should be consistent across the domain for global SIDs to be understood.

SR-MPLS adjacency SID is locally significant and represents a specific neighbor link; a prefix SID represents a node and is globally unique via index.

SR-TE policy steers traffic via a segment list (SID list); candidate paths can be explicit or computed by a path computation element (PCE/PCEP).

SRv6 uses 128-bit SIDs encoded as locator:function; an SRH (Segment Routing Header) carries the segment list and is processed by SRv6-capable nodes.

TI-LFA is an IP fast reroute mechanism that computes a backup path using a post-convergence loop-free alternate with segment lists; it is enabled per interface under the IGP.

PIM-SM builds a shared tree (*,G) rooted at the RP; sources register with the RP, and receivers join via IGMP to the last-hop router.

PIM-SSM requires IGMPv3 and a known source; no RP is used, and the (S,G) tree is built directly with PIM (S,G) joins; group range is commonly 232.0.0.0/8.

PIM-SM switchover: after traffic exceeds the SPT threshold, the last-hop router sends an (S,G) join toward the source and prunes the shared tree with a prune toward the RP.

RPF check: multicast packets must arrive on the interface that the unicast route to the source points to; failure drops the packet and increments RPF failure counters.

MSDP peers use TCP 639 to exchange source-active (SA) messages between RPs in different PIM-SM domains; it is not used for SSM.

Anycast RP: multiple RPs share the same loopback address; MSDP mesh-group or PIM anycast-RP advertises the address so sources and receivers converge on the nearest RP.

mLDP (multicast LDP) builds point-to-multipoint or multipoint-to-multipoint LSPs using LDP label mapping; used for MVPN core transport.

PIM-DM uses flood-and-prune with assert messages; PIM-SM is the default on Cisco IOS-XR for most deployments; IGMP snooping is a Layer 2 function and does not replace PIM.

NSR (non-stop routing) keeps the control plane state synchronized between active and standby RP so that a switchover does not require neighbor re-establishment.

NSF (non-stop forwarding) relies on the neighbor's helper mode: the restarting router signals graceful restart, and neighbors keep forwarding while adjacencies rebuild.

BFD: UDP port 3784 (single hop) and 4784 (multihop); detection time = negotiated interval x multiplier; a BFD session down triggers IGP/BGP teardown faster than hello timers.

BFD echo mode uses UDP 3785 and reflects packets back from the neighbor; echo mode is not supported on all platforms and is disabled by default.

IGP fast convergence: tune hello/hold timers, enable LFA or TI-LFA, and use BFD for sub-second failure detection; BGP PIC pre-installs a backup path in the FIB.

BGP PIC (prefix-independent convergence) installs a backup next hop for prefixes so that a failed primary path can be replaced without waiting for BGP reconvergence.

TI-LFA provides 50 ms protection by precomputing a post-convergence LFA path with a segment list; it is preferred over classic LFA in SR deployments.

Graceful restart timers must be longer than the expected control-plane restart; if the restart exceeds the timer, neighbors tear down adjacencies and reconverge normally.

Bidirectional forwarding detection timers: min-echo-receive, min-rx, and multiplier must match or the session uses the higher values; mismatched timers cause session down.

OSPFv2 external metrics: type 1 adds the redistributed metric to the internal path cost; type 2 (default) uses the external metric only, with internal cost as tiebreaker.

OSPF forwarding address in type 5 LSAs is set to 0.0.0.0 when the next hop is the advertising ASBR itself or the outgoing interface is passive/suppressed; nonzero means lookup is done through that address.

OSPF NSSA: type 7 LSAs are translated to type 5 at the NSSA ABR elected translator (highest router ID by default); type 7 P-bit set on redistribution by an ASBR allows translation.

OSPF area types: stub blocks type 5 (default route injected by ABR); totally stubby also blocks type 3; NSSA blocks type 5 but allows type 7; totally NSSA blocks type 3 and 5.

IS-IS levels: L1 uses level-1 LSPs within an area, L2 uses level-2 LSPs in the backbone; an L1/L2 router sets the ATT bit in its L1 LSP to signal an L2 exit, and L1 routers install a default route toward it.

IS-IS default metric is narrow (6-bit, max 63); wide metrics (up to 2^24) are required for TE and are enabled per level with metric-style wide.

IS-IS LSP purge requires the remaining lifetime to reach zero; zero-age lifetime is flooded to purge; overload bit (OL) prevents transit traffic but still allows local prefixes.

Redistribution loop prevention: tag routes at the redistribution point and apply route-maps that deny matching tags on the opposite redistribution; never redistribute one protocol into another bidirectionally without filtering.

Seed metric is mandatory for redistributed routes: OSPF requires metric (default 20 for BGP, 20 for connected/static via default-metric); IS-IS requires metric or default-metric; EIGRP requires bandwidth/delay/MTU.

BGP path selection order (Cisco): highest weight, highest local preference, locally originated (network/aggregate/redistribute), shortest AS_PATH, lowest origin (IGP<EGP<incomplete), lowest MED, eBGP over iBGP, lowest IGP metric to next hop, oldest route, lowest router ID, lowest neighbor address.

Weight is Cisco-proprietary and local to the router; local preference is propagated within the AS via iBGP but not across eBGP.

MED is propagated to neighboring ASes only; by default it is compared only among paths from the same neighboring AS unless bgp always-compare-med is set.

iBGP split horizon: a route learned from an iBGP peer is not advertised to another iBGP peer; full mesh, route reflectors, or confederations are required.

Route reflector rules: routes from a client are reflected to clients and non-clients; routes from a non-client are reflected to clients only; cluster ID and originator ID are used for loop prevention.

Confederation: sub-AS numbers are used internally, the confederation ID is presented to external peers; AS_PATH length across the confederation is not counted for external path selection.

BGP neighbor states: Idle, Connect, Active, OpenSent, OpenConfirm, Established; stuck in Active usually means TCP port 179 unreachable or an ACL/unicast reverse-path issue.

Next-hop reachability: iBGP requires an IGP route to the next hop or next-hop-self on the advertising router; eBGP multihop requires update-source loopback and a route to the loopback.

Common troubleshooting checks: show ip bgp summary (state/PfxRcd), show ip bgp neighbors (capabilities, timers), show ip bgp (locally originated), show ip bgp (best path reason).

LDP binds a label to each IGP prefix (FEC) and distributes it to LDP peers over TCP 646; the label is locally significant and swapped hop by hop along the LSP.

LDP session requires matching LDP router ID reachability and transport address; mismatched transport addresses cause session flaps or no session.

MPLS TE: RSVP-TE (protocol 46) signals label-switched paths; explicit-path and affinity/attribute flags constrain path selection; CSPF prunes links failing constraints before running shortest path.

MPLS TE bandwidth is reserved with the setup/hold priority model; a lower numeric priority is more important, and preemption occurs when a new LSP cannot be placed.

Segment Routing: an SRGB (Segment Routing Global Block) maps prefix SIDs; index + SRGB start = global label; the SRGB should be consistent across the domain for global SIDs to be understood.

SR-MPLS adjacency SID is locally significant and represents a specific neighbor link; a prefix SID represents a node and is globally unique via index.

SR-TE policy steers traffic via a segment list (SID list); candidate paths can be explicit or computed by a path computation element (PCE/PCEP).

SRv6 uses 128-bit SIDs encoded as locator:function; an SRH (Segment Routing Header) carries the segment list and is processed by SRv6-capable nodes.

TI-LFA is an IP fast reroute mechanism that computes a backup path using a post-convergence loop-free alternate with segment lists; it is enabled per interface under the IGP.

PIM-SM builds a shared tree (*,G) rooted at the RP; sources register with the RP, and receivers join via IGMP to the last-hop router.

PIM-SSM requires IGMPv3 and a known source; no RP is used, and the (S,G) tree is built directly with PIM (S,G) joins; group range is commonly 232.0.0.0/8.

PIM-SM switchover: after traffic exceeds the SPT threshold, the last-hop router sends an (S,G) join toward the source and prunes the shared tree with a prune toward the RP.

RPF check: multicast packets must arrive on the interface that the unicast route to the source points to; failure drops the packet and increments RPF failure counters.

MSDP peers use TCP 639 to exchange source-active (SA) messages between RPs in different PIM-SM domains; it is not used for SSM.

Anycast RP: multiple RPs share the same loopback address; MSDP mesh-group or PIM anycast-RP advertises the address so sources and receivers converge on the nearest RP.

mLDP (multicast LDP) builds point-to-multipoint or multipoint-to-multipoint LSPs using LDP label mapping; used for MVPN core transport.

PIM-DM uses flood-and-prune with assert messages; PIM-SM is the default on Cisco IOS-XR for most deployments; IGMP snooping is a Layer 2 function and does not replace PIM.

NSR (non-stop routing) keeps the control plane state synchronized between active and standby RP so that a switchover does not require neighbor re-establishment.

NSF (non-stop forwarding) relies on the neighbor's helper mode: the restarting router signals graceful restart, and neighbors keep forwarding while adjacencies rebuild.

BFD: UDP port 3784 (single hop) and 4784 (multihop); detection time = negotiated interval x multiplier; a BFD session down triggers IGP/BGP teardown faster than hello timers.

BFD echo mode uses UDP 3785 and reflects packets back from the neighbor; echo mode is not supported on all platforms and is disabled by default.

IGP fast convergence: tune hello/hold timers, enable LFA or TI-LFA, and use BFD for sub-second failure detection; BGP PIC pre-installs a backup path in the FIB.

BGP PIC (prefix-independent convergence) installs a backup next hop for prefixes so that a failed primary path can be replaced without waiting for BGP reconvergence.

TI-LFA provides 50 ms protection by precomputing a post-convergence LFA path with a segment list; it is preferred over classic LFA in SR deployments.

Graceful restart timers must be longer than the expected control-plane restart; if the restart exceeds the timer, neighbors tear down adjacencies and reconverge normally.

Bidirectional forwarding detection timers: min-echo-receive, min-rx, and multiplier must match or the session uses the higher values; mismatched timers cause session down.