Esxi HPE 7.0.2 (17630552) issue with USB stick

[Tony4688]

Hello everyone,

I installed Esxi HPE 7.0.2 (17630552) on USB stick of server HPE ProLiant DL380 Gen10 some months ago and it worked properly. But few days ago, suddenly it lost mount on USB stick (/vmfs/volumes/LOCKER...) and its VMs don't work properly. This is the results of "df -h" command:

[root@ESXi:~] df -h
GetUnmountedVmfsFileSystemsInt: fsUUID is null, skipping mpx.vmhba33:C0:T0:L0:7
GetUnmountedVmfsFileSystemsInt: fsUUID is null, skipping mpx.vmhba33:C0:T0:L0:7
GetUnmountedVmfsFileSystemsInt: fsUUID is null, skipping mpx.vmhba33:C0:T0:L0:7
GetUnmountedVmfsFileSystemsInt: fsUUID is null, skipping mpx.vmhba33:C0:T0:L0:7
Filesystem Size Used Available Use% Mounted on
VMFS-6 20.0T 8.0T 12.0T 40% /vmfs/volumes/DELL-LUN01
VMFS-6 12.0T 9.1T 2.9T 76% /vmfs/volumes/DELL-LUN02
vfat 4.0G 256.7M 3.7G 6% /vmfs/volumes/BOOTBANK1
vfat 4.0G 256.8M 3.7G 6% /vmfs/volumes/BOOTBANK2
vsan 13.1T 190.4G 12.9T 1% /vmfs/volumes/vsanDS

I want to backup Esxi host configuration, reinstall the latest image Esxi HPE 7.0.2 (17867351) then restore the backup configuration to save depoyment time, but backup command can not success as below:

[root@ESXi:~] vim-cmd hostsvc/firmware/backup_config
(vmodl.fault.SystemError) {
faultCause = (vmodl.MethodFault) null,
faultMessage = <unset>,
reason = "Internal error"
msg = "Received SOAP response fault from [<cs p:000000dae6875620, TCP:localhost:8307>]: backupConfiguration
A general system error occurred: Internal error"
}

I tried to backup via vSphere CLI and vSphere PowerCLI but still had the same results.

Could anyone tell me how to backup host configuration for this case?

Thanks so much!

[vXav]

I had this issue a couple weeks ago (I assume it's the same anyway) after I reinstalled my R430 on a USB stick. The productlocker symlink is missing (ls -al /) but the folder is out there in vmfs.

KB1037405 fixed it for me.

[Tony4688]

Hi vXav,

Thanks for your information, but KB1037405 does not have relation with my case, i tried its guideline but it's unable to solve my issue.

[vXav]

Right, so the symlinks are fine when you run "ls -al /" ? No red?

Just poking, people seem to have had issues with the scratch location, is it configured ok? do you see a downloads folder in there? The config backup issue looks an awful lot like this.

Also you mentioned the VMs are impacted? How so?

[Tony4688]

Hi vXav,

I understand the missing of folder "/tmp/_osdata.../downloads" can be one of the reasons when backup failed, but it's not my case.

You know, in my case the Esxi host lost the partition "/vmfs/volumes/LOCKER..." that mount to USB stick (i installed Esxi on USB stick), therefore the host lost "/scratch" location and some others too.

Almost certainly, i think i have to reinstall Esxi on this host. But i want to ask is there any solutions to backup host configuration and restore it for this case?

Many thanks!

[Tony4688]

 I 've just used vcenter to backup and restore the configuration to host, it's the last useful method.

Thank all!