SRX

last person joined: yesterday 

Ask questions and share experiences about the SRX Series, vSRX, and cSRX.
  • 1.  chassis cluster control link redundancy

    Posted 12-01-2011 09:15

    Hi,

     

    i'm running srx240's as chassis cluster in the lab.

     

    i already use ex4200s in virtual chassis, and really like the vcp - interface connection. (because they are actually redundant.)

    did i miss something on the srx or is the control link a single point of failure?

     

    btw is there any real pro's for srx chassis cluster instead of just using two single srx?

    i already found a lot of con's like missing link aggregation, missing mpls support, no upgrading for single chassis members...

     

     



  • 2.  RE: chassis cluster control link redundancy
    Best Answer

    Posted 12-01-2011 10:05
      |   view attached

    I have LACP working in a chassis cluster running under 10.4R7.5, what I have been doing when upgrading firmware was to turn one of the members off and upgrade the other, then turn the upgraded one off and turn on the other one to be upgraded back on and then turn the other back on after both have been upgraded. I'm sure this isn't supported but it has worked without issue so far. As for the single link I haven't come across anything about being able to add another link.

     

    I included my config so you can take a look at the LACP portion.

    Attachment(s)

    txt
    fwconfig.txt   10 KB 1 version


  • 3.  RE: chassis cluster control link redundancy

    Posted 12-04-2011 23:50

    Thankx so far guys,

     

    ccorkrum:

    did you test the throughput of your lacp reth0 link? 

    does the use of lacp actually increase your link speed? or is it just used for redundancy and to connect lacp-enabled switches?

     

    EDIT: the Feature Support List says LACP and Link Aggregation are not supported in chassis cluster mode for SRX240

    http://www.juniper.net/techpubs/software/junos-security/junos-security10.4/junos-srx-jseries-support-reference/jd0e3251.html

    i will actually try it out later in the day and update the post.



  • 4.  RE: chassis cluster control link redundancy

    Posted 12-05-2011 02:52

    i tried it and it actually works, until now i really trusted the Juniper Documentation... 

     

    ge-0/0/5 and 0/0/6 have redundant-parent reth1 

     

    cluster1> show interfaces reth1
    Physical interface: reth1, Enabled, Physical link is Up
    Interface index: 150, SNMP ifIndex: 526
    Link-level type: Ethernet, MTU: 1514, Speed: 2Gbps[...]

     

    maybe someone can clear that up: why does the feature list say (

    Link Aggregation Control Protocol (LACP) in chassis cluster mode

    No

    when it seems to be supported? 

     

     



  • 5.  RE: chassis cluster control link redundancy

    Posted 12-01-2011 10:22
    hi,

    there are two pro's:
    session failover
    single configuration