Switching

 View Only

IMPORTANT MODERATION NOTICE

This community is currently under full moderation, meaning  all posts will be reviewed before appearing in the community. Please expect a brief delay—there is no need to post multiple times. If your post is rejected, you'll receive an email outlining the reason(s). We've implemented full moderation to control spam. Thank you for your patience and participation.



  • 1.  EX series failover times

    This message was posted by a user wishing to remain anonymous
    Posted 01-28-2022 09:29
    This message was posted by a user wishing to remain anonymous

    Hello,

    I am looking to deploy a number of switches and switch stacks(VC) within our network and would like to try and find out what the expected failover times and failover options are for these switches:

    - EX2300
    - EX4300
    - EX4600

    We are using two EX4600s in VC to aggregate a mix of 2300/4300/4600s as single switches or in stacks of two, all switches will be running Layer 2 only apart from management/QoS.

    VC links will be either 10g or 1g ports, uplink ports from access to aggregation will be 2x 10g links in a LAG, I am looking for information such as:

    - If one port in the LAG fails for a single switch or VC, how much downtime if any is expected
    - If one switch in a VC fails, how much downtime is expected before the second switch takes over

    And what methods are available to reduce failover time.

    Thanks in advance,

    Chris


  • 2.  RE: EX series failover times

    Posted 02-01-2022 10:06
    *bump*

    ------------------------------
    Chris Limbert
    ------------------------------



  • 3.  RE: EX series failover times

    Posted 02-01-2022 10:21
    Hi Chris,

    without knowing the exact topology and protocols involved I can just make assumptions.

    If you run LACP over the LAG and the „active" link fails, the Backup Link needs to „converge" so I would expect 1-2 ping losses from Experience.

    Usually a VC (especially a 2-Member VC) is really really fast.
    You will most likely use 1 or 2 pings when the Backup switches to the Master state. In a 2-Member-VC you already have both devices running on the role routing-engine so failover shouldn't be a problem :) But I usually test this in a quick PoC in order to know that my topology is running the way I want it to be running.

    That being said the EX2300 has a weaker CPU than the 4300 so it might be possible to see a 3-4 ping delay if the switch also has a heavy load - but again - I can only assume than as I don't know your topology.

    ------------------------------
    Christian Scholz
    Juniper Networks Ambassador | JNCIE-SEC #374
    Mail: chs@ip4.de
    Blog: jncie.eu | Twitter: @chsjuniper | YT-Channel: netchron
    ------------------------------