I installed a fresh copy of ESXi 5.5 to an USB stick. The server has a RAID 1 array for VMFS.
The scratch partition is set to: /vmfs/volumes/5257d01f-4dc80900-0296-002590306ec4/.locker
But after every reboot of the system in the events there is the following warning:
No scratch partition has been configured. Recommended scratch partition size is {} MiB.
Output for vim-cmd hostsvc/advopt/view ScratchConfig.ConfiguredScratchLocation:
key = "ScratchConfig.ConfiguredScratchLocation",
value = "/vmfs/volumes/5257d01f-4dc80900-0296-002590306ec4/.locker",
Am I doing something wrong? Why is the warning message coming back every time I reboot the server?
Can you check following :-
1. Confirm Datastore UUID is correct ?
2. Scratch location must be on a Fat16 or VMFS volume.
3. The location may be a directory on a shared or remote disk as long as the ESXi host is allocated its own separate directory.
4. For Serial ATA (SATA), a disk connected through supported SAS controllers or supported on-board
SATA controllers. SATA disks will be considered remote, not local. These disks will not be used as a
scratch partition by default because they are seen as remote.
Thanks for your answer.
1. Yes the datastore UUID is correct
2. It's on a VMFS volume
I have an Adaptec 2405 Raid controller. I have 2 data disks connected to the Adaptec and created a Raid 1 array.
The VMFS volume is located on the Raid 1 array and the ESXi installation in on an USB stick.
I just did a fresh install of ESXi 5.1 u1 with the same hardware as mentioned above.
Now there is NO error message regarding the scratch partition!
Maybe a bug in 5.5?
I have done some tests today.
Same hardware but installed ESXi 5.1 u1
No error message about the scratch partition.
Upgraded the 5.1 u1 to 5.5
Error message is back again...
It certainly sounds like a bug. I am trying to find any details on it but cannot find anything yet. If you have support, then log a call.
Do you have enough space? Around 6GB? so that scratch is created.
Running a ESXi 5.5 on USB too.
If the scratch partition is on the root of the volume:
key = "ScratchConfig.ConfiguredScratchLocation",
value = "/vmfs/volumes/52dca6ec-38f49282-8596-001ec9d9684f",
The error "No scratch partition has been configured. Recommended scratch partition size is {} MiB." will not appear.
If the scratch partition is point to a folder on the volume:
key = "ScratchConfig.ConfiguredScratchLocation",
value = "/vmfs/volumes/52dca6ec-38f49282-8596-001ec9d9684f/scratch"
The error will appear upon every restart "No scratch partition has been configured. Recommended scratch partition size is {} MiB." in the events.
Current volume is vmfs type
I am seeing this exact behavior as well.
This is happening to me as well on my fresh install. I should note that I am using the FREE version of VMWare vSphere Hypervisor 5.5
I'm having the same issue on a fresh install of 5.5 update 1, using VMware-VMvisor-Installer-5.5.0.update01-1623387.x86_64.iso
This is most definitely a bug.
After a lot of trial and error I have found a workaround to this bug.
"No scratch partition has been configured. Recommended scratch partition size is {} MiB"
Go to Advanced Settings and move the ScratchConfig location to the root of the datastore.
It means you'll end up with lots of new folders in the root instead of neatly contained within the .locker folder, but this is the only way to fix this bug in 5.5.0 update 1.
example:
ScratchConfig: /vmfs/volumes/datastore
Then reboot.
After the reboot browse datastore and delete the old .locker folder as it is no longer needed.
EDIT: Just noticed that lam0073 discovered the same thing.
So if you point the scratch location to the root of a datastore, that means only one host can use the datastore as the scratch location correct? So you have to have one datastore per host scratch partition??? Not ideal, but if it works, it works...
Stephen, I would use local datastores if possible.
I would use local datastores if I could. Generally, the only reason (and my reason) to configure the scratch partition in the first place is if you don't have any local datastores. We're booting from SD card, so we need an alternate location for storing log files and dump files.
Does anyone have any experience with the syslog collector? Is that another viable alternative to manually editing the scratch location?
I got a same problem last month, and asked to vm support. I got the answer from the vm support below (in my language).
----------------------------------
This warning message is able to confirm similar events in past cases, up to ESXi5.5 U1.
※ as scratch partition, the message is output when the environment in which the route of the data store is not specified.
It becomes the issue will be corrected in the Update 2 or later.