1. chassis redundancy hierarchy
2. initial RE configuration example
you can config a master-only address on fxp0
3. copy a configuration from one RE to another
4. load a software package from the other RE
5. config RE redundancy
[edit chassis redundancy]
routing-engine slot-number (master | backup | disabled);
启用GRE时,keepalive signal自动被启用,failover时间被设置为2s,M20上是4s,不能手动更改。
When a software process fails,RE can switchover to backup RE.after config failover-routing-engine at
[edit system processes
process-name]
可以通过file show /var/log/mastership查看RE切换状态
6. config CFEB redundancy on M10i
the cfeb communiticates with RE using a dedicate 100mbps FE link that transfer routing table data from RE to the forwarding table in the ASIC.
The link also used to transfer from the CFEB to the RE routing link-state updates and other packets destined fro the router that have been received through the router interface.
[edit chassis redundancy]
cfeb slot-number (always | preferred);
7. config FEB redundancy on M120
redundancy-group group-name {
feb slot-number (backup | primary);
description description;
no-auto-failover;
}
1:1-a redundancy group have primary FEB and backup FEB,backup FEB contain same state with primary FEB.
N:1-no primary FEB,forwarding state must download from RE.
group 0,自动切换不会发生,
group 1,可以切换
8. example:config FEB redundancy
9. config SFM redundancy on M40e
M40e have 2 SFM,M160 have 4 SFM
10. config SSB redundancy on M20
[edit chassis redundancy]
ssb slot-number (always | preferred);
Graceful Routing Engine Switchover System Requirements
1. when GRE switchover and aggregated Ethernet interface configured in the same system.the aggregated Ethernet interface must not configured for fast-polling LACP.
2. Vrrp changes mastership when a RE switchover occus,even GRE configured.
3. Extended DHCP local server does not supported GRE switchover.
4. GRE not support for service PIC
5. Issue the “show system switchover” on backup RE to check the GRE
Nonstop Bridging System Requirements
1. on MX series
2. support STP/RSTP/MSTP
3. config nonstop bridging
[edit chassis redundancy]
graceful-switchover
[edit protocols layer2-control]
nonstop-bridging;
4. issue the Layer 2 control protocol-related operational mode commands on backup RE
nonstop active routing
1. NSR use the same infrastructure as GRE.but NSR also save RPD info.
2. NSR does not rely on helper routers to assist the routing platform in restoring routing info.
3. nonstop active routing is a natural replacement for graceful restart.
4. NSR reature support
BFD/BGP/ISIS/LDP..and so on.
L3 vpn support doesn’t include RSVP-based tunnels,dynamic GRE tunnel,multicast ×××,or BGP flow routes.the fllowing OSPF feature and configuration statements are also not supported.
domain-id domain-id
domain-vpn-tag number
sham-link
5. When a BFD session di distributed to the PFE,bfd packets continue to be sent during RE switchover.if nondistributeed BFD sessions are to be kept alive during a switchover,you must ensure that the session failure detection time is greater than the RE switchover time.
These BFD session are not distribured to the PFE:multihop sessions,tunnel-encapsulated sessions, and sessions over aggregated Ethernet and integrated routing and bridging (IRB) interfaces
6. NSR supports layer 2 circuit and LDP-based VPLS
7. You can disable NSR for PIM only by use nonstop-routing disable under protocol pim
8. When you enable nonstop active routing, you cannot enable automatic route distinguishers for multicast ××× routing instances
Automatic route distinguishers are enabled by configuring the route-distinguisher-id statement at the [edit routing-instances instance-name] hierarchy level
9. Verifying Nonstop Active Routing Operation
1.show task replication
2.for bgp NSR,use show bgp replication
3.for bfd session,use show bfd session,the replicated flag appears when a BFD session has been replicated to the backup RE.
Graceful restart
1. Aggregate and Static Routes
graceful-restart statement at the [edit routing-options]
2. Global graceful restart config
3. BGP
If you configure Bidirectional Forwarding Detection (BFD) and graceful restart
for IS-IS, graceful restart might not work as expected
5. OSPF
If you configure BFD and graceful restart for OSPF, graceful restart might not
work as expected
6. ×××
For Layer 3 ×××s only, you must also configure graceful restart for all routing and MPLS-related protocols within a routing instance.
That means global config and instance config
7. Graceful Restart Operational Mode Commands
1. BGP
2. ISIS and ospf
Cofig debug
3. CCC and TCC
8. VRRP
1. cofig the virtual ip address to be the same as physical interface’s address,the interface becomes the master,in this case,you must config the priority to be 255 and you must config preemption by including the preempt statement.
2. Cannnot config a virtual ip address to be the same as the interface’s address for an aggregated Ethernet interface.
3. Mixed tagging supported only for IQ2 PIC,if you include the flexible-vlan-tagging for a vrrp enable interface on a PIC that does not support mixed tagging.VRRP on that interface is disabled.
4. If you enable MAC source address filtering on an interface.you must inclue the virtual MAC address in the list of source MAC address.
00:00:5e:00:01:00 through 00:00:5e:00:01:ff are reserved for VRRP
5. Vrrp authentication for IPV4 only
6. For IPV6,you must config ipv6 router advertisements for the interface under [edit protocols router-advertisement].
But also need to cofnig virtual-router-only under [edit protocols router-advertisement interface interface-name] to avoid backup vrrp response to mster received by client
7. Vrrp can track whether a route is reachable(that is, the route exists in the routing table of the routing instance included in the configuration) and dynamically chage the priority of the vrrp group.
8. config inheritance for a vrrp roup
[edit interfaces interface-name unit logical-unit-number family inet address address vrrp-group group-id]:
vrrp-inherit-from vrrp-group
both inherithing group and active group must be on the same physical interface and on the same routing instances
9. by default,vrrp logs the error,data carrier detect(DCD) configuration,and routing socket events a file in the /var/log/vpprd.
10. Configuring the Silent Period
[edit protocols vrrp]
startup-silent-period seconds;
The silent period starts when the interface state is changed from down to up. During this period, the Master Down Event is ignored. Configure the silent period interval to avoid alarms caused by the delay or interruption of the incoming VRRP advertisement packets during the interface startup phase.