I’m completely newbie in Vmware world. Just like many others, I use Vmware vShpere in my day-to-day work, but it can actually be called a basic use of Vmware vSphere.
Yesterday I was working on a lab and suddenly power went off and came back again. I haven’t connected my ESXi host to the UPS and had to power it on manually. I tried to connect to the ESXi host by vSphere Client as usual, but I got this strange message:
Despite this issue, I still can connect to the ESXi host by SSH and it seems that I have full access to the commands. I also tried to troubleshoot what was completely unknown to me and those can be summarized as bellow:
– I tried to connect to the host using HTTP/HTTPS, but was unsuccessful.
– I tried to connect to the host from other PCs; Even a PC in the same L3 network as the host, to eliminate any possible firewall-related issue; unsuccessful.
– I reload the host several times; I restarted the services
[root@timaz-esxi:~] /etc/init.d/hostd status hostd is not running. [root@timaz-esxi:~] /etc/init.d/hostd start Ramdisk 'hostd' with estimated size of 303MB already exists hostd started. [root@timaz-esxi:~] /etc/init.d/hostd status hostd is running. [root@timaz-esxi:~] /etc/init.d/hostd status hostd is not running. [root@timaz-esxi:~] /etc/init.d/vpxa status vpxa is not running [root@timaz-esxi:~] /etc/init.d/vpxa start [root@timaz-esxi:~] /etc/init.d/vpxa status vpxa is running
As seen, “hostd” and “vpxa” services are down and despite that I started them manually, they automatically went offline after some seconds!
[root@timaz-esxi:~] /etc/init.d/vpxa status vpxa is not running [root@timaz-esxi:~] /etc/init.d/hostd status hostd is not running.
Some says that some files are probably have exceeded their size and then have prevented services from starting successfully. I checked various recommended files and they were very smaller than the max size.
– I successfully upgraded ESXi host 6.0 to 6.5 but the situation was the same, which connecting through SSH was OK but still HTTP/HTTPS didn’t work.
Anyway, whatever I did, I didn’t manage to solve the issue and I needed to do something, because I didn’t want to rebuild my VMs from the scratch.
As a last resort, I tried to do a clean install the ESXi host but preserve the existing datastore to see if that makes a sense before going crazy and format the whole disk!
For this, I created a bootable usb disk and installed the clean version of the ESXi host while preserving the existing VMFS datastores, instead of overriding.
After setup process was finished, I managed to connect to the ESXi host via HTTP/HTTPS. But what was important was my VM files which I didn’t want to lose them. To verify if my files are still there, I went to Storage page and Clicked on Register a VM.
I was able to see my existing files and folders on the disk; woohooo! I selected vmx file inside every existing folder which mapped to a different VM and clicked Finish button. My VM had restored and there was no one happier than me!
I did this to every VM (every folder belongs to a single VM) and was able to restore all of my VMs. The only thing that I did in addition to these, was creating port groups again, because new ESXi didn’t have those in its DB.
I think this was a little help for someone like me who is a beginner and don’t know any expert to ask for help.