Baseline tries to update VIB which isn't installed

[clivegardner]

I am trying to remediate some HPE Proliant DL360 Gen9 hosts against the "Non-Critical Host Patches (Predefined)" baseline under the very latest vSphere 7.0. This has all gone fine with the exception of one patch, ssacli-component_4.21.7.0-7.0.0, which is the HPE Smart Array controller cli tool.

This VIB is not installed, however vCenter determines that the host is non-compliant against the baseline and this patch needs to be installed. Letting it try and do so fails.

There is however a VIB installed named hpessacli-component_5.20.8.0-7.0.0_18545638, which delivers the same functionality as far as I can tell, so it makes sense the older version wouldn't install.

Does anyone know why it is attempting to update both, despite only the hpessacli version being installed? Can anyone tell me how I can prevent the the older package being updated/installed and get the baseline back into compliance? it doesn't appear in the list of installed VIB's so I can't simply remove it that way.

Thanks

 

Clive

[AhhhOK]

Did you ever come up with a workaround for this.  I am having the same issue but with both 

Smart Storage Administrator CLI for ESXi

ssacli-component_4.21.7.0-7.0.0

and 

Smart Storage Administrator CLI for ESXi

hpessacli-component_5.20.8.0-7.0.0

saying they are missing but not being installed when Remediation is run.  Also not seeing any version of them when looking at VIBs.  

[chrisjsteinke]

Anyone figure this out?

 

I have this problem as well. But on my DL560Gen10s and my DL360Gen10s but curiously not on my DL380Gen9s or DL380Gen10 hosts. 

I have an open support case with VMWare and so far no solution. If I find a solution working with support, I can post it here. 

For me I have two SSACLI updates that are blocking the system from becoming compliant.

ssacli-component_4.21.7.0-7.0.0 
ssacli-component_5.10.45.0-7.0.0

 

I'm going to include a link to this thread to my support ticket to let them know it's not just me!

[AhhhOK]

I have not.  I had a coworker open a case with vmware to start and he will go to HPE if needed.  Once I get more info I will be happy to post it. 

[Ambrogio De Lorenzo]

Hello, any news on that.

I have same problem on all my BL460 with ESXi 7.0.3d and Non critical base line.

Regards

 Ambrogio

[chrisjsteinke]

See my follow up post at the bottom! Good Luck and I hope this works for you!

[ObjectifDubai]

Hi,

I had the same problem for a few years on HPE SYNERGY servers, two vibs caused me errors when they were installed because they were wrong, and although I was in a version higher than their version, the VUM offered them to me at installation.
I ended up on a proposal from VMWare support by resetting the update manager and the update finished offering them to me for installation. It even told me that I was up to date.

https://kb.vmware.com/s/article/2147284

 

[HostaHost]

I'm really getting sick of this sh*t with HPE and their incessant need to rename / recode existing VIBs and extensions to include their name, breaking updates in the process.  We already went through this after the Nimble acquisition where they just couldn't help themselves from renaming Nimble Connection Manager to HPE Storage Connection Manager, preventing any easy upgrades on hosts that already had NCM installed.  They didn't even bother to spend the time making it not use the same old /etc/nimble/ directory structure for its configs, just had to have that HPE name in there.  Next up they renamed ssacli to hpessacli for whatever stupid reason.  Our not even a month old 703c HPE image installs had been fine on hpessacli 5.20, but now can no longer get to current patch status with hpessacli 5.30 because the older ssacli's have begun popping into the upgrade repository.  I'm guessing this is because they gave the VIB the exact same name of "Smart Storage Administrator CLI for ESXi".

Kind of defeats the purpose of using their images when you simply can't trust they aren't going to break your ability to upgrade when using them. 

[chrisjsteinke]

I feel your pain! 

[AhhhOK]

My coworker was told by VMW support that they are fixing it in future updates.  My thought was maybe if we manually install the first package hpessacli-component_4.18.1.0-7.0.0-signed_component-15803178.zip which was included on the original 7.0 custom iso.  ( I used 7.0 U2 custom iso with hpessacli-component--5.10.45.1-7.0.0 to build my hosts ) that then they would be able to upgraded per the update manager.  I have not tried that yet.

Thoughts?

[chrisjsteinke]

I have an update on this and was able to resolve the problem. Yes, I believe a fix is coming in a later release of vSphere 7. I can't remember which one they said. 

 

The solution was to create a copy of the Non-Critical patches (Predefined) baseline and go through the copied baseline and remove the offending VIBs from the new baseline and then remove the Predefined baseline from your hosts and add the new clone (san the offending VIBs)

They sent me the following blog and doc link. 

https://kb.vmware.com/s/article/85701?lang=en_US

https://docs.vmware.com/en/VMware-vSphere/7.0/com.vmware.vsphere-lifecycle-manager.doc/GUID-8D70BB5A-B08F-41B9-80CC-95426F0AC1AD.html

I followed the above and was able to resolve the problems I was seeing. Does this work for others?

 

Good Luck and thanks again!