Uninstalling Jumbo Hotfix R81.20 Take 90

[Soroosh]

Hi mates,
Recently, on an open server gateway in a XL Cluster, I uninstalled an older version of JHF to install Take 99 because there was not enough space. I uninstalled the HF in Gaia Portal with no errors, but after rebooting, I lost all connectivity to the gateway. After investigating, I figured out that it could have also deleted FW kernels. The OS loaded, but there is no connectivity at all. I can access the shell/CLISH only through the ESX console. It seems that I now have to reinstall Gaia.
I'm wondering what I did wrong.
I've attached a screenshot of the installed and available hotfixes. I uninstalled Take 90 and 65 together.

[G_W_Albrecht]

There is nothing you can do wrong in WebGUI CPUSE that will delete the kernel ! After install/uninstall, GAiA WebGUI cert will change and has to be accepted in browser after deleting the old cert. If OS loads, kernel is present after all ! Screenshot shows JT 90 as installed... Did you try to ping from ESX console to0 8.8.8.8 to check if VMware internal routing is working ?

[Soroosh]

Actually, nothing works as it should!
The WebGUI is no longer available.
There is no connection between SMS and the gateway, and the Smartcosole shows that it is gone.
SSH is not possible either.
The boot time is too long as well.
The screenshot was taken before uninstalling and rebooting.
The gateway has no connectivity at all. Pinging 8.8.8.8 returns no reply.

[the_rock]

What does fw stat show?

[the_rock]

Can you send output of cpinfo -yfw1 from expert mode?

Andy

[Soroosh]

Here it is 🙂

[the_rock]

If you can please send output of fw stat, would help. Something clearly went very wrong here...

Andy

[Soroosh]

🙂

[the_rock]

Okay, question...can you try run fw unloadlocal on it to see if that works?

Andy

[Soroosh]

Unfortunately, this also returns an error.

[the_rock]

I hate to tell you this, but I dont believe there is an easy way to fix that issue. Only way I could think of is if you have recent backup or snapshot saved locally, then you can try restore it, but if not, contact TAC and see if they can offer anything, otherwise, seems like reinstall might be the only logical option.

Andy

[PhoneBoy]

Observe the disk space requirements in the release notes: https://sc1.checkpoint.com/documents/R81.20/WebAdminGuides/EN/CP_R81.20_RN/Content/Topics-RN/Open-Se... 

When installed, your VM disk should have enough free (unallocated) space to handle 2-3 snapshots of your root partition.
Otherwise, upgrades and patches may not apply (either at all or correctly).
How much disk space did you allocate for the VM?

[Soroosh]

Thanks for the notice and link! It has 110 GB. Another admin installed and configured everything before me. I would have just updated it to JHF T99, but there was not enough space for that, so I decided to uninstall the old JHFs. That went badly wrong!
It seems that, due to insufficient free space, the older JHFs/GAIA were not installed correctly, yet the Gateway was working without any problems.

[PhoneBoy]

110GB might be enough to get the image installed, but it's nowhere near enough for operational usage (including upgrades).
I usually use somewhere between 150-200GB for my VMs. 

[the_rock]

That seems like enough space, I did it before in the lab even with 50G.

Andy