Manual IPv6 Tunnels

Ipv6 tunnels ares used to interconnect 2 or more IPv6 network through an Ipv4 network.

A sample configuration is :

interface Tunnel 0
no ip address
ipv6 address 2001:CC1E:1:4545::5/64
ipv6 rip RIPng enable
tunnel source Loopback0
tunnel destination 150.1.4.4
tunnel mode ipv6ip | gre
end

Tunnel source and destination should be IPv4 adresses

The differents modes are

GRE : default mode (protocol 47), used tu encapsulate multiple layer 3 protocols
IPV6IP : less overhead than GRE (protocol 41) encapsulate only IPv6

OSPF Fast Hello

Normal OSPF timers could be configured defining the hello-interval and dead-interval, as hello interval is 1 sec, detect a dead neighbor within 1 second means at first loss of a hello.

ip ospf hello-interval 1-65535s
ip ospf dead-interval 1-65535s


Fast hello permit a hello-interval below 1sec. The following command means hello of 333ms and a dead-interval of 1sec :

ip ospf dead-interval minimal hello-multiplier 3

Split Horizon on Frame-Relay

Split horizon must be take into account for DV protocols :

By default split-horizon is enabled on all cisco router interface.

Except for frame-relay interface, and particulary on mutilpoint Frame-Relay interface (ex on the Hub router). It acts differently for RIP and EIGRP

RIP : by default split-horizon is disabled on FR interface

EIGRP : Split-horizon must be manually disabled on FR interface

int s0/1
no ip split-horizon eigrp 100

OSPF Network Types

A review of the different network types and behaviour of ospf :

NBMA : default for frame-relay physical and point-to-multipoint subinterface
Broadcast :default for ethernet
Point-to-Point : default for point-to-point subinterface


Remark : When using Pt-to-Mpt the next hop is the adjacent router and not the originating router as in Nbma or Broadcast networks

Tips :
Point-to-x : no DR/BDR
Non-broadcast : manual neighbor

Frame-Relay Legacy Traffic Shaping

Some recalls :

Guaranteed rate = minimum guaranteed rate by the Telco. Configured with minCIR (bps) by default is 1/2 CIR
CIR = rate provided by Telco during no congestion. Configured with CIR (bps) By default it's 2x CIR
Access rate = maximum rate of the link. Configured with Bc (bits). Bc = (Access Rate - CIR)*Tc By default Tc is 125ms or 1/8 of seconds.

Example
32kbps is guarantedd
64kbps of Cir
192kbps of Access Rate

map-class frame-relay cisco
frame-relay cir 64000
frame-relay mincir 32000
frame-relay adaptive-shaping becn
frame-relay bc 8000
frame-relay be 16000 

interface Serial0/0 
interface Serial/0/
frame-relay traffic-shaping
frame-relay class cisco

Verification commands :

show traffic-shaping

Frame-Relay DE setting

Purpose of the exercice is to set the DE bit of a frame-relay packet depending on the length. 2 solutions depending on constraints :

1) If it's for all dlci of a given interface, you could use a Service-Policy :

class-map match-all DE1024
match packet length min 1024
!
policy-map DE1024
class DE1024
set fr-de
!
interface Serial0/0
service-policy output DE1024

2) If it's for a specific dlci, you could use the de-list command :

frame-relay de-list 1 protocol ip gt 1024
!
interface Serial0/0
frame-relay de-group 1 501

To check, connect on the other side router :

R2#sh frame-relay pvc 105

PVC Statistics for interface Serial0/0 (Frame Relay DTE)

DLCI = 105, DLCI USAGE = LOCAL, PVC STATUS = STATIC, INTERFACE = Serial0/0.2

input pkts 25 output pkts 25 in bytes 13600
out bytes 13600 dropped pkts 0 in pkts dropped 0
out pkts dropped 0 out bytes dropped 0
in FECN pkts 0 in BECN pkts 0 out FECN pkts 0
out BECN pkts 0 in DE pkts 7 out DE pkts 0
out bcast pkts 0 out bcast bytes 0
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
pvc create time 00:49:33, last time pvc status changed 00:35:02

Non-Cisco SFP

Maybe you tried in vain to put non-cisco (Finisar,...) SFP inside a Catalyst.
2 Magical commands to accept it :

service unsupported-transceiver
errdisable detect cause gbic-invalid

Thats-all