When I try to Upgrade my ESXi3.5 server (Which is Xeon E5310 on ASUS DSBV/SAS box with 5 SATA unRAID Disk on SAS controller) to ESXi4
I got Error: Unsupported Host - Unsupported Boot Disk Layout
What can I do now? Is it bug or just do not support local disk boot?
Based on the screen shot of fsc2, the ESXi installable host doesn't have a scratch partition (partition 2) as you can find in dell2.jpg. This is not a typical ESXi installable disk partition layout . Do you still remember how the host is installed?
it is just an standard installation, put the cdrom in the server and install...
in esxi, there are no realy things you can change in the setup...
If you find this information useful, please award points for "correct" or "helpful".
I think there's more to this than just the presence of a scratch partition; my ESXi 3.5 box DOES have the 4GB FAT16 partition, yet, I get the same error message as everyone else in this forum post. I've attached some read-outs from the box.
Perhaps there's a free-space consideration related to the scratch partition? I note that 1GB of the scratch space is taken up by uwswap, could this be a problem? Has anyone heard back from VMware on any progress diagnosing the issue, better yet, a patch?
there is still no update on my SR, it looks like vmware isn´t interessted to take care on this issue ....
If you find this information useful, please award points for "correct" or "helpful".
Could you post the result from 'esxcfg-scsidevs -l' on fsc box?
This is a different problem according to Filesystem Data.txt. The real problem is out of range in shell numerical comparison. Please wait for a update release.
if no one from vmware look on this, it is waste of time to post any more information here..
If you find this information useful, please award points for "correct" or "helpful".
It won't waste the time as we are looking into the issue. Your help is appreciated.
ok, on esx 3.5i, i can´t find the esxcfg-scsidevs command.
If you find this information useful, please award points for "correct" or "helpful".
here is the output of the command from an fresh esx4i installation on the fsc hardware:
~ # esxcfg-scsidevs -l
mpx.vmhba0:C0:T0:L0
Device Type: CD-ROM
Size: 0 MB
Display Name: Local HL-DT-ST CD-ROM (mpx.vmhba0:C0:T0:L0)
Plugin: NMP
Console Device: /vmfs/devices/genscsi/mpx.vmhba0:C0:T0:L0
Devfs Path: /vmfs/devices/genscsi/mpx.vmhba0:C0:T0:L0
Vendor: HL-DT-ST Model: RW/DVD GCC-T20N Revis: 1.01
SCSI Level: 5 Is Pseudo: false Status: on
Is RDM Capable: false Is Removable: true
Is Local: true
Other Names:
vml.0005000000766d686261303a303a30
naa.6003005700434d7011a3d67f05021a19
Device Type: Direct-Access
Size: 139488 MB
Display Name: Local LSI Disk (naa.6003005700434d7011a3d67f05021a19)
Plugin: NMP
Console Device: /vmfs/devices/disks/naa.6003005700434d7011a3d67f05021a19
Devfs Path: /vmfs/devices/disks/naa.6003005700434d7011a3d67f05021a19
Vendor: LSI Model: MegaRAID SAS RMB Revis: 1.12
SCSI Level: 5 Is Pseudo: false Status: on
Is RDM Capable: false Is Removable: false
Is Local: true
Other Names:
vml.02000000006003005700434d7011a3d67f05021a194d6567615241
If you find this information useful, please award points for "correct" or "helpful".
Was this fixed with a later patch?
Or is their any other method to work around it?
I got the same issue on IBM LS42 and IBM x3850 M2 servers (all of them, but not IBM HS21 i.e.)
Does anyone know if VMware is even aware of this issue ?
VMware is aware of this issue, and there are currently no solution and it is unknown if there are always an solution for this problem.
Support told me as workaround to backup the vm´s, install esxi4 new from scratch, and restore the vm´s
For this host, there is no way to upgrade and also keep vmfs volume (partition 2). If no vmfs volume has been created, vmare is going to add upgrade support in update 1 release.
So If I replace the ESXi 3.5U4 USB stick with another that has ESXi4.0 on it, will it touch my datastores? (Some conversion on first boot, etc?) I saw users complaining about disappearing datastores when they tried to upgrade their system. Should I make an offline backup of the vms then restore everything to the new system if it gets messed up somehow?
Using a Dell Poweredge 2900 system with an internal usb stick, which might have the same partition issues like the one above.
If you replace your ESXi 3.5U4 USB stick with a ESXi 4.0 stick your datastores should be untouched. However you might have to re-register all the VMs in your datastore to use them. I would still recommend an offline backup of your datastore before you perform this.
I'm having the same issue with upgrading IBM 3850M2 using ServeRAID-MR10K.
I've tired both update manager and host upgrade utility, both report the same message above
Page 30 of th VSphere upgrade guide lists only that the IBM 8K SAS is supported for the ESXI. Does this mean i have to wait until it is supported.?
Can I use ESXI4 Bootable CD perform an upgrade at each box instead?
ESXi4 does not support CD based upgrades. Is it possible to post an output of "fdisk -l" from your ESXi machine.
Thanks!
Here is the response from VMWare. This is a know issue and the only current workaround is to do a fresh clean install of ESXI 4 instead of upgrading.