Lots of issues after every VBR update

[zadrian]

It was always never fun and getting weeks of headache after every Veeam Backup and Replication update. This 12.127 is no different.

1. The ISO file is so huge ? Is there a smaller set of files for specific modules ?

2. The various components are not updated.
- Need to update the repositories (by editing the repositories)
- Need to update transport modules (by editing WAN accelerator settings)
- need to update vCenter components, but at least it shows "outdated servers etc" (cannot remember the name but "outdated" was shown)

3. Some jobs fail as C++ module not installed (proxy VBR server) ??
I did run the VBR 12.127 update (9 GB ISO file) on 2 of my proxy VBR servers, updated with no error but got the veeam backup proxy is missing c++ runtime components.
I had to get info from Internet to use one of the updated VBR server to "install" updated transport module ?
https://www.veeam.com/kb2678

4. Need to update the IO filter. But the stupid thing is that need to put ALL ESXi into maintenance mode at the same time ?
This is plain stupid as this is literally shutting down ALL the VMs ?
I have split the several ESXi into different "clusters" but it needs to update ALL the ESXi in the cluster also ? else it will give errors ?
I have at least 2 ESXi in each of the 5 clusters, in VBR "VeeamCDP filter management" some clusters are showing up to date and some showing out of date but currently at least 1 ESXi in each cluster was not updated as at least 1 ESXi cannot be in maintenance mode (1 ESXi in prd whole other ESXi is standby with no powered on VM).
Can we have at least the IO filter update by ESXi host rather then "cluster" ??

5. suddenly after VBR update, there is SureBackup Job Failure: "The resource 'VeeamBackup_' is in use." ??
Worse of all the resource seems to be a datastore on the VM running VBR ?
I am still trying to figure out what is the issue......but I think I may solve #4 first as I hope updating the IO filters would solve some issues ???

[tyler.jurgens]

I haven't ran into nearly that many issues with upgrades, usually my Veeam upgrades go well.

However, with the I/O Filter, I'd really love to see Veeam integrate with vLCM to push any updates. That way you can manage the updates far more seamlessly.

[Gostev]

1. ISO file is big because VBR is an enterprise backup product with tons of capabilities. And 12.1 is a new product build meaning every component was rebuilt, so there's no possibility to create a smaller redistribution.

2. Normally you should upgrade all remote components at once with the Upgrade wizard, which is available directly in the backup console. If that does not work for you then we would be interested to take a look, as there are no other similar reports.

3. This is unexpected and should be investigated with Support. I do know that 12.1.2 had some security-related changes around VDDK redistribution (it was decoupled from transport service) and this change may be running into something unexpected in your environment. I vaguely remember hearing something about possible issues with installations where customers made some manual VDDK and/or C++ module swaps (could have been done long time ago) or something along these lines. But if you are facing a known issue, then support would know how to solve it already because 12.1.2 has been around for a while now.

4. Maintenance mode a VMware requirement, there's no other supported way to update the IO filter. However, of course you should not need to put all ESXi into maintenance mode at once, normally it should be a rolling upgrade. @veremin does this ring you any bells, all ESXi shut down at once??

5. This must be environment-specific issue as I haven't seen similar reports. Feel free to contact support if you can't figure this one out.

Finally, I would like to kindly ask you follow the forum rules displayed when you click New Topic with your future contributions to this community, as you did brake a few with the post above

[veremin]

4. Yes, by default, the process should be a rolling update - VMware updates one host, then moves on to the next, and so on. We don't have any control over this process - we send a command like "update IO filter" to the VMware infrastructure side and then wait for it to report that everything has been successfully completed. Unfortunately, it doesn't always handle its tasks properly, resulting in a significant number of artifacts during the upgrade (including putting all hosts into maintenance mode simultaneously instead of sequentially).

To understand what specifically occurred in your case, it would be necessary to open a support case.

Thank you.

[lampshade]

Issue #4 really burns me - this is why I update the VIB manually. I place each host in maintenance mode one at a time, pull the VIB file from veeam install folder, then run the VIB update command in the host CLI to update the VIB. once all hosts are updated, rescan in Veeam and its good to go. I just don't trust it to let it do its own thing.