Upgrading NSX-T in the lab, the Host Pre-Check came back with:
Disk space requirements not met. Please ensure 180MB free space is available in tmp partition to proceed with upgrade.
On the 30th July, 2020, NSX-T 2.5.2 was released. Check out the release notes.
This is a minor update with almost 80 issues marked as fixed.
Last month VMware released updated versions of some exams, including the VCAP-NV Design exam based on NSX-T 3.0. In the past year or so I’ve been working a lot more with NSX-T, but.. I’m not a real* networking guy, so I wasn’t confident in taking NV exams.
After passing the VCP-NV on NSX-T 2.4 earlier in the year, I put off the VCAPs because they were based on NSX-T 2.4 or NSX-v 6.x, but with the updated exam out, I was running out of excuses.
You might have heard there’s a pandemic going on, so doing exams remotely is the only option here, and suited me.
Finally getting around to upgrading to ESXi 7.0, and I was stuck pretty quickly, with the error: BOOT_DISK_SIZE ERROR.
Switching between a mix of different product versions, I forget some of the specifics, and I’ve wasted too much time troubleshooting something that was working OK. Hopefully this saves you some time.
Here’s a few tips from what I do when deploying Photon OS in the lab. Keep in mind this is only for troubleshooting and testing, so don’t make the same changes to anything outside of your lab.
If you followed the previous blog NSX-T: Deploy NSX-T Manager 2.5 with OVFtool you should have NSX-T Managers deployed.
There’s a few more things we have to configure within NSX-T Manager like:
If you have the NSX-T Manager OVA on a fileshare or web server you can deploy it from the vCenter GUI, or use any other deployment tool like OVFtool.
Using OVFtool to deploy NSX-T Manager is a great way to deploy a quick and repeatable configuration. It also helps to document the configuration options.
After adding a compute manager, the Registration Status shows as
Not Registered, and Connection Status is
This is usually caused by a previously registered NSX-T Manager to vCenter, that has been deleted. If you remove the Compute Manager via the Delete option in NSX-T Manager, it also removes the registration from vCenter.
If you try and deploy vRA8 into VMware Cloud on AWS, you’ll get this error:
User has no administrative privileges
For those with NSX-v 6.4, they’d know the end of general support date was 16/01/2021, (just over a year from now). If you check the product lifecycle it now shows the end of general support is 16/01/2022.
When using the Migration Assistant to Migrate from a Windows 5.5 vCenter to 6.5 VCSA, the migration assistant was failing to run on the Windows vCenter server.
It automatically saves the logs on the desktop. When viewing the logs, this part stood out:
2019-07-22T00:31:58.193Z ERROR upgrade.states.component_states vcIntegrity:CollectRequirements: Remote script failed with an error [InternalError()] 2019-07-22T00:31:58.193Z ERROR upgrade.states.component_states vcIntegrity:CollectRequirements: failed with internal error. For details take a look at CollectRequirements_com.vmware.vcIntegrity_2019_07_22_10_31.log. 2019-07-22T00:31:58.209Z WARNING wf.wf_processor State CollectRequirements com.vmware.vcIntegrity has failed.
During an upgrade from 5.5 to 6.5, update manager started to do the upgrade, but after the host rebooted, it would boot into the ESXi 6.5 installer, but then display
About to install for 15 seconds, then flash an error message, and then go back to
About to Install. Leaving it for a few hours didn’t make any difference, and I was unable to alt-F1 and log into the console, as when the error message flashed up, it would reset the login process.
Deploying vRNI 4.1 you might get an error:
Insufficient capacity on each physical CPU.
Ovftool can be pretty awesome, but it can also be very difficult to use. I seem to have a love/hate relationship with it.
Here I will go over some common issues I have with the tool. I’ve only tested on the Windows version of ovftool.