Got this message after adding a new hosts. Looks at all the Advances System Settings and they are all the same as the rest of the hosts in the cluster. All host say ([] /scratch/log for Syslog.global.logDir. And the ScratchConfig.CurrentScratchLocation is set to /tmp/scratch. So not sure why the error. Anyone have a clue why I am still getting this Warning? Any help is appreciated. Thanks
Perry
If you have SD cards, like we with our Dell servers, than the msg is normal. Most likely you have supress the warning on the existing hosts some time ago. The solution would be to configure a scratch partition to your preferred Datastore (doesnt work with vSAN) and reboot the host.
Info: A long time ago there was a Problem in ESXi where the the scratch config was persistent across host reboots.
Regards,
Joerg
Do the other, existing hosts have HDD drives on which ESXi is installed?
This message shows up for systems where ESXi is installed on an SD/USB device. In such a case you'll have to configure a persistent scratch location.
André
All hosts are exactly the same. All have it installed on SD's. And all have the same settings for both those in Advanced Settings.
If you have SD cards, like we with our Dell servers, than the msg is normal. Most likely you have supress the warning on the existing hosts some time ago. The solution would be to configure a scratch partition to your preferred Datastore (doesnt work with vSAN) and reboot the host.
Info: A long time ago there was a Problem in ESXi where the the scratch config was persistent across host reboots.
Regards,
Joerg
You will see this warning in every ESXi host that is installed on SD memory. So if you don't see that on other servers (and you are sure about installation path for them, that is the SD memory) then I think you may change their /scratch/log path to another datastore with something like host profile sttings. It's not bad to check them again by running the following command on one of the ESXi that doesn't have that warning:
esxcli system syslog config get