Answer: A. All work together. Only GR restart mode is not supported while NSR is in effect.
Answer: A. The JUNOS NSR implementation is not designed to be hitless through a local routing process restart. GR, in contrast, does work through such a restart, albeit with control plane disruption. The remaining methods are all viable NSR failover trigger methods.
Answer: B. When all goes to plan, there should be no dataplane hits in a graceful restart. This makes B the only false answer.
Answer: D. Only D is true. Matched versions and configurations are recommended with GRES and NSR, but are not technically mandated.
Answer: C. NSR is based on protocol replication with independent message processing on both REs. Using the same messages and processing rules, the backup RE computes a shadow RIB that is expected to math that on the master RE. The RIB/FIB state is not simply replicated between the two REs.
Answer: D. The
show system switchover command is run only
on the backup RE, which is where the
ksyncd process runs and reports GRES
Answer: A. While various
replication commands can be run on both REs, the
show task replication command can only run
on the master RE and is the best gauge of overall system NSR
Answer: D. BGP keepalives are always RE based. All other session types are handled in the PFE by the PPM process.
Answer: D. According to the 11.4 documentation, RE-based BFD sessions should use at least a 2,500 ms detection timer to ...