vSphere 8.0 U2 is Released: Reminder not to Update immediately

  • 22 September 2023
  • 47 comments
  • 12925 views

Userlevel 7
Badge +20

Morning everyone,

 

VMware vSphere 8.0 U2 is now available, with improvements to lifecycle management, AI workloads, and DevOps amongst others!

https://blogs.vmware.com/vsphere/2023/09/vsphere-8-update-2-now-available.html?utm_source=rss&utm_medium=rss&utm_campaign=vsphere-8-update-2-now-available
 

A quick reminder to all, Veeam Backup & Replication v12 only supports up to vSphere 8.0 U1 currently. Veeam can’t start their validation until the code is available so we’ll now see Veeam begin their testing process.

 

You can check the status of vSphere platform support here: https://www.veeam.com/kb2443

Veeam typically target vSphere platform support within 90 days of release, so with the betas of VBR 12.1 functionality appearing online, I’d guess we won’t see 8.0 U2 support until 12.1 ships unless that encounters massive delays.

 

So yes, wait for Veeam to support vSphere 8.0 U2 before you upgrade, there’s no point having all this new functionality if you can’t protect any of your workloads!


47 comments

Userlevel 7
Badge +8

Very interesting. I’ll wait a bit and add to the test environment. 

Userlevel 7
Badge +20

8.0u2 B was released on 1th of March, and the VMwares KBs were also updated, that this version is the corrected, we don’t need to wait for the 8.0u3:

https://kb.vmware.com/s/article/95965

“In order to resolve this issue, update to ESXi 8.0 Update 2b or newer.”

Yes, this release fixes that bug with CBT.

Userlevel 1

8.0u2 B was released on 1th of March, and the VMwares KBs were also updated, that this version is the corrected, we don’t need to wait for the 8.0u3:

https://kb.vmware.com/s/article/95965

“In order to resolve this issue, update to ESXi 8.0 Update 2b or newer.”

Userlevel 7
Badge +20

Please be informed, I got info from VMware support:

This issue affecting ESXi only.
80U3 version ETA on this Summer.

So you can move forward with the vCenter upgrades, and ESXi upgrades need to be planned for the summer.

Great news.   I’d advise to wait until the fall to see if there are any other issues added with this release 😉

Yes, that is always a good idea due to updates tending to break something else.  We will see how well Broadcom’s QA is.  😋

Userlevel 7
Badge +8

Please be informed, I got info from VMware support:

This issue affecting ESXi only.
80U3 version ETA on this Summer.

So you can move forward with the vCenter upgrades, and ESXi upgrades need to be planned for the summer.

Great news.   I’d advise to wait until the fall to see if there are any other issues added with this release 😉

Userlevel 1

I guess that patch hasn’t arrived yet…… Was ready to upgrade but I’ll continue to hold tight. 

Yeah, they pulled back and now it is going to be in P3 and U3 according to the KB article.

I didn’t see that, thanks for the post. I’ll consider going to U1 or staying on 7.3 a little while longer. It’s running good and I like my CBT not broken!

Yeah, U2 on vCenter is ok but go just to U1 on hosts.  You are safe with that configuration.

 

I think this combination may be an option if you are not running PROD, because in case of any problem, the VMware support will ask you to upgrade the latest version on the given main version, so you will be forced to use ESXi 8.0u2

If you are running PROD, then you need to use vCenter 8.0u2, and ESXi 7.0u3 for a supported and maintainable config.

 

Also be informed, I got info from VMware support:

This issue affecting ESXi only.
80U3 version ETA on this Summer.

So you can move forward with the vCenter upgrades, and ESXi upgrades need to be planned for the summer.

Userlevel 7
Badge +8

I guess that patch hasn’t arrived yet…… Was ready to upgrade but I’ll continue to hold tight. 

Yeah, they pulled back and now it is going to be in P3 and U3 according to the KB article.

I didn’t see that, thanks for the post. I’ll consider going to U1 or staying on 7.3 a little while longer. It’s running good and I like my CBT not broken!

Yeah, U2 on vCenter is ok but go just to U1 on hosts.  You are safe with that configuration.

My brain will suffer. haha. I like things the same. but you are right, that’s an option forsure. 

Userlevel 7
Badge +20

I guess that patch hasn’t arrived yet…… Was ready to upgrade but I’ll continue to hold tight. 

Yeah, they pulled back and now it is going to be in P3 and U3 according to the KB article.

I didn’t see that, thanks for the post. I’ll consider going to U1 or staying on 7.3 a little while longer. It’s running good and I like my CBT not broken!

Yeah, U2 on vCenter is ok but go just to U1 on hosts.  You are safe with that configuration.

Userlevel 7
Badge +8

I guess that patch hasn’t arrived yet…… Was ready to upgrade but I’ll continue to hold tight. 

Yeah, they pulled back and now it is going to be in P3 and U3 according to the KB article.

I didn’t see that, thanks for the post. I’ll consider going to U1 or staying on 7.3 a little while longer. It’s running good and I like my CBT not broken!

Userlevel 7
Badge +20

I guess that patch hasn’t arrived yet…… Was ready to upgrade but I’ll continue to hold tight. 

Yeah, they pulled back and now it is going to be in P3 and U3 according to the KB article.

Userlevel 7
Badge +8

I guess that patch hasn’t arrived yet…… Was ready to upgrade but I’ll continue to hold tight. 

Userlevel 7
Badge +20

@molnar.tamas Actually this bug has been discovered by Veeam. All details can be found in this post and the following replies:

https://forums.veeam.com/post505571.html#p505571

The patch from VMware was planned for last week so we should see it very soon.

@Chris.Childerhose It's a VMware bug which cannot be fixed by Veeam. The only workaround from Veeam's side was to reset CBT for full backups.

Yes, I get that and understand.  I am just saying that if anyone is eager to test U2 they can, and it works minus the CBT issue.  I don’t hot extend my VMs in my lab so I don’t have the bug and can use this release.

Userlevel 7
Badge +14

@molnar.tamas Actually this bug has been discovered by Veeam. All details can be found in this post and the following replies:

https://forums.veeam.com/post505571.html#p505571

The patch from VMware was planned for last week so we should see it very soon.

@Chris.Childerhose It's a VMware bug which cannot be fixed by Veeam. The only workaround from Veeam's side was to reset CBT for full backups.

Userlevel 7
Badge +20

I better read this. Getting ready to upgrade to 12.1 and 8.2

It is a bad bug but not as bad as you would think if you followed the KB to work around it.  Veeam is working on fixing this from what I understand.

you mean VMware? There is no info from Veaam about this bug.

VMware plan to correct this issue in 8.0.3, with no available ETA today.

An I think this is a really bad bug, because you will only face it, when you need restore something, Surbackup will detect is, but lot of customers don’t use that.

And I thin there should be more info and communication about this...

Yes, VMware is going to address it, but Veeam is also aware of it and looking in to it as well from what I have heard.

I agree about more communication about it and am not dismissing it being a bad bug as we are at least aware from the KB VMware published.

Userlevel 1

I better read this. Getting ready to upgrade to 12.1 and 8.2

It is a bad bug but not as bad as you would think if you followed the KB to work around it.  Veeam is working on fixing this from what I understand.

you mean VMware? There is no info from Veaam about this bug.

VMware plan to correct this issue in 8.0.3, with no available ETA today.

An I think this is a really bad bug, because you will only face it, when you need restore something, Surbackup will detect is, but lot of customers don’t use that.

And I thin there should be more info and communication about this...

Userlevel 7
Badge +8

Right, Running an active full after each hot expand might be a bit of a pain though. Especially with the size of my file servers.


I think I may wait a while. 

While that would be a pain yes, I was referring to the cold expansion where the VM is powered off then you expand the disk. 😉

That’s actually WORSE for me lol.  24/7 shop of critical systems. I get small patching windows, but when a disk fills up nobody wants to power down a server.  Especially as others are reading for it, or an another drive etc.

 

Knowing about this is half the battle though. For some it won’t be an issue at all

Userlevel 7
Badge +20

Right, Running an active full after each hot expand might be a bit of a pain though. Especially with the size of my file servers.


I think I may wait a while. 

While that would be a pain yes, I was referring to the cold expansion where the VM is powered off then you expand the disk. 😉

Userlevel 7
Badge +8

Right, Running an active full after each hot expand might be a bit of a pain though. Especially with the size of my file servers.


I think I may wait a while. 

Userlevel 7
Badge +20

I better read this. Getting ready to upgrade to 12.1 and 8.2

It is a bad bug but not as bad as you would think if you followed the KB to work around it.  Veeam is working on fixing this from what I understand.

Userlevel 7
Badge +8

I better read this. Getting ready to upgrade to 12.1 and 8.2

Userlevel 1

If upgraded, be aware of a nasty CBT bug in 8.0.2: https://kb.vmware.com/s/article/95965?lang=en_US

I did not found any Veeam KB regarding this:

“vSphere 8.0 Update 2 changed the way disks are extended. The changes were specifically intended to make certain disk hot-grow operations more efficient.

This has unintentionally resulted in incorrect tracking of changes, leading to backups not capturing the right data and, consequently, corrupt restores.

This issue will only be seen if backups are taken after the VM disk is hot extended. Simply resizing a VM with the VM powered off will not cause this.

This problem can happen with disks of all datastore types (VVOL, VMFS, NFS, vSAN). “

Userlevel 7
Badge +20

Hello
Veeam announced compatibility-level support for vSphere 8.0 U2 with the latest VBR release v12 P20230718.

https://www.veeam.com/kb2443?ck=1676356277100

This is great and before 12.1 too.

Userlevel 7
Badge +7

Hello
Veeam announced compatibility-level support for vSphere 8.0 U2 with the latest VBR release v12 P20230718.

https://www.veeam.com/kb2443?ck=1676356277100

Userlevel 7
Badge +20

That is probably due to the I/O filter driver not being compatible with U2 just yet.  I believe in 12.1 when it is released it will be so you will need to wait until then unfortunately.

Badge

CDP doesnt work. Anyone found workaround ? Lab environment.. 

 

 

[TryOverrideApiVersion] The native API version [8.0.2.0] for VC [vcenterpr.domain.corp] was changed to current max supported [8.0.1.0]
[12.10.2023 10:12:33.551] <12> Info         [Soap] QueryClusterIoFilterInfo, clusterComputeResourceRef 'domain-c5830769', vendorName 'VEE'
[12.10.2023 10:12:33.614] <12> Info         [Soap] GetHostsUnderCluster, clusterRef 'domain-c5830769'
[12.10.2023 10:12:33.614] <12> Warning      [TryOverrideApiVersion] The native API version [8.0.2.0] for VC [vcenterpr.domain.corp] was changed to current max supported [8.0.1.0]
[12.10.2023 10:12:33.817] <12> Info         [IOFilter] Discovering filters on host [esxidr.domain.corp]
[12.10.2023 10:12:33.911] <12> Info         [IOFilter] Filters on host [esxidr.domain.corp] successfully discovered: Filter not installed
[12.10.2023 10:12:33.926] <12> Info         [IOFilter] Discover results: No CDP components discovered on cluster [Dell Intel Cluster]
[12.10.2023 10:12:33.926] <12> Info         [IOFilter] Installing CDP components on cluster [Dell Intel Cluster]
[12.10.2023 10:12:33.926] <12> Info         [IOFilter] Checking I/O filter installation requirements for cluster Dell Intel Cluster
[12.10.2023 10:12:33.926] <12> Info         [IOFilter] Checking I/O filter installation requirements for host esxidr.domain.corp
[12.10.2023 10:12:33.942] <12> Info         [Soap] GetHardwareInfo, hostRef 'host-5822178'
[12.10.2023 10:12:33.989] <12> Info         [IOFilter] Host esxidr.domain.corp meets I/O filter installation requirements
[12.10.2023 10:12:34.051] <12> Info         [Soap] InstallIoFilter, vibUrl 'https://ms-veeam.domain.corp:33034/dapi/bundle/8.0.0/12.0.1008', clusterComputeResourceRef 'domain-c5830769'
[12.10.2023 10:12:36.286] <12> Info         [Soap] QueryClusterIoFilterInfo, clustSerComputeResourceRef 'domain-c5830769', vendorName 'VEE'
[12.10.2023 10:12:36.333] <12> Error        [IOFilter] Failed to log filter issues: filter not installed
[12.10.2023 10:12:36.333] <12> Error        [IOFilter] 
[12.10.2023 10:12:36.348] <12> Error        Failed to install CDP components via cluster API for Dell Intel Cluster (System.Exception)

Comment