Forum Discussion

werner_verheyle's avatar
werner_verheyle
Icon for Altocumulus rankAltocumulus
Jan 24, 2024

ltm ve network interface driver

Hi ,

we have some older LTM VE devices running in v12.1.3.7, for support reasons we are upgrading them to v14 & later to v15. (LTM upgrade path )

They are running on vmware ESX v7 (so higher than 6.7).

When we tried to upgrade to v14.1.5.2 as a first step in going to v15, we lost all connectivity to this system. Even our mgmt interface was not responding anymore (ssh or https). When accessing over a vmware console, we could see the config was loaded and running in v14. And the interfaces were still present but not able to send any communication over any interface.

At the end we used the switchboot command for returning to old version v12.

I'm aware of several article like https://my.f5.com/manage/s/article/K74921042 concerning network driver to be updated. But when we try to perform these "tmctl -d blade tmm/device_probed" command then i'm getting an error "tmctl: tmm/device_probed: No such table". 

Can anyone tell us why we are seeing this error on tmctl command ? i know we need to upgrade , and we need to check the network driver after doing this. But if i can't check network drivers , how can i know if they need to be updated as described in the article.

any suggestion is welcome.

  • Does the BIG-IP VE license support v14? Check the range on the license. Did you re-activate license prior to creating new volume with new code? You should be within the upgrade path (https://my.f5.com/manage/s/article/K13845), however, sometimes it makes sense to a) upgrade to latest version within current major release first and b) perform a smaller step upgrade perhaps using v13 first.

    Did you perform any troubleshooting while on the v12 code? You should have been able to use the VMware console to access via virtual terminal. Did you have an F5 support case open to have someone on hand to assist with such issues? Did you try to see why the config wasnt loading --- did you try to run the hidden bigpipe command (not removed, but execution path no longer included) with the load daol params? This does a manual load of the staged new config and will output any errors, which you can then tshoot.

  • Hello,

    Adding to the above notes, you can check whether the configuration was loaded successfully when you booted to the new volume or not.

    #show sys mcp.

    and if the config failed to load, you can verify the configuration and check why it is failing.

    #load sys config verify

    You can also check the interfaces status after the upgrade, and whether the mac address mapping to the interfaces is set correctly or not.

    I faced the same issue before, and the solution was to create a new VM on the old version and perform the upgrade again. 

    I recommend to open a case with F5 support first before procedding with any aciton, so they can check the current status of the F5 instance.

    Thanks,

    Mohamed Salah