Forum Discussion

ryang1991_17782's avatar
ryang1991_17782
Icon for Nimbostratus rankNimbostratus
Jan 19, 2015

Is upgrading a Viprion pair similar to upgrading an LTM pair?

Hey all,

 

I was wondering if the process of implementing a hotfix to a Viprion pair (Model 2100/2400) is similar to the process implementing a hotfix to an LTM pair (Model 6900). I'm assuming the process is the same if they're both active/standby devices, but I just want to make sure. Both pairs of devices are running version 10.2.4.

 

application delivery
LTM
viprion

2 Replies

Sort By
  • amolari's avatar
    amolari
    Icon for Cirrus rankCirrus
    Jan 19, 2015

    it's the same but with extra precaution:

     

    • If you have multiple blades in the chassis, that all of them do have management interface connectivity, as another blade might become the Master after reboot.
    • it takes a long time to upgrade until system is up again (up to 45 minuts if I remember well)
  • StephanManthey's avatar
    StephanManthey
    Icon for MVP rankMVP
    Jan 19, 2015

    Hi Ryan,

    on VIPRION with multiple blades an independent installation per blade is used.

    Please crosscheck by running the following:

    tmsh show sys software

    If you are installing a hotfix, it will be replicated automatically to each blade in the chassis and show up in /shared/images/ directory of each blade. The clusterd process is responsible for distribution of software.

    As soon as all necessary components are in place (for installing a hotfix to a new installation volume both base ISO image and hotfix ISO image will be required on each blade) the installation begins.

    Running the following command allows to monitor the progress on CLI:

    watch tmsh show sys software
    If the installation is finished you may want to use the 'switchboot' command to modify the default installation volume.

    Please use the 'clsh' command on CLI to reboot the chassis as it runs the command on all available blades:

    clsh reboot Or simply reboot by using the WebUI after all software installation on all volumes is finished. After reboot it will be the quorum process to determine the cluster master blade.

    To verify cluster state from CLI you can use:

    tmsh show sys cluster

    Please do not reboot the chassis or individual blades while the installation on one of the blades is still in progress.

    If the hotfix involves upgrades to BIOS or FPGAs multiple reboots (independently per blade) may happen. Never reboot a blade until BIOS of FPGA updates are finished (you will see from command line prompt). Please note, that you can expect individual console outputs per blade.

    Thanks, Stephan