Environment
VMware virtualisation environment
Guest OS:
Novell SUSE Linux Enterprise Server 9 Service Pack 4 (SLES9 SP4), or
Novell Linux Desktop 9 Service Pack 4 (NLD9 SP4), or
Novell Open Enterprise Server 1 (Linux based) Support Pack 2 (OES1 SP2)
Guest OS:
Novell SUSE Linux Enterprise Server 9 Service Pack 4 (SLES9 SP4), or
Novell Linux Desktop 9 Service Pack 4 (NLD9 SP4), or
Novell Open Enterprise Server 1 (Linux based) Support Pack 2 (OES1 SP2)
Situation
Changes
After a reboot, The system does not boot properly anymore. The boot process ends before the root filesystem is mounted, with messages similar to the following:
Symptoms (fresh installation)
During the "Installation Settings" phase of the installation, an error popup is presented, with an ERROR
- SLES9 Service Pack 4 was applied, or
- NLD9 SP4 was applied, or
- the kernel was updated through an online update (YaST Online Update, or Red Carpet/rug) to version 2.6.5-7.308 or newer, or
- A new installation is attempted, booting from SLES9 SP4 CD 1 or NLD9 SP4 CD1.
After a reboot, The system does not boot properly anymore. The boot process ends before the root filesystem is mounted, with messages similar to the following:
Waiting for device /dev/sda6 to appear....
[...]
No root device found; exiting to /bin/sh
sh: can't access tty; job control turned off
and a subsequent "#" prompt from which no commands (other than shell builtins) are available.No root device found; exiting to /bin/sh
sh: can't access tty; job control turned off
Symptoms (fresh installation)
During the "Installation Settings" phase of the installation, an error popup is presented, with an ERROR
No hard disks were found for the installation.
Please check your hardware!
Please check your hardware!
Resolution
Status and fix (Last updated 2008-01-24)
Update to the current version and patches of the VMware virtualisation product used. If this does not resolve the issue, contact VMware support.
For VMware ESX Server 3.0.x: apply patch ESX-1002431, "ESX Server 3.0.2, Patch ESX-1002431; Updates to VMware-esx-vmx and VMware-esx-vmkernel; Fix For Detecting LSI Logic Controller, Support for PCI-X NICs on IBM System x3655".
For VMware Server: Novell has received customer reports that this issue is fixed as of version 1.0.4.
Update to the current version and patches of the VMware virtualisation product used. If this does not resolve the issue, contact VMware support.
For VMware ESX Server 3.0.x: apply patch ESX-1002431, "ESX Server 3.0.2, Patch ESX-1002431; Updates to VMware-esx-vmx and VMware-esx-vmkernel; Fix For Detecting LSI Logic Controller, Support for PCI-X NICs on IBM System x3655".
For VMware Server: Novell has received customer reports that this issue is fixed as of version 1.0.4.
Additional Information
Root cause
This issue is due to a bug in VMware's emulation of the LSI MPT Fusion HBA causing it to report zero attached disks. This bug gets exposed by the updated driver for this HBA which is included in SLES9 SP4's kernel (which is also used in NLD9 SP4 and in online updates to OES1 SP2).
Workaround (updated system)
When updating the VMware virtualisation product is not an option, the issue can be worked around as follows:
Start the installation using SP3 CD1 rather than SP4 CD1, then apply SP4 once the installation has been completed.
Note
Although the symptoms are quite similar, this issue is different from the one documented in TID 3494481 - System fails to boot after kernel update. That issue affects physical systems however and is caused by device renaming, not by no hard disks being detected.
This issue is due to a bug in VMware's emulation of the LSI MPT Fusion HBA causing it to report zero attached disks. This bug gets exposed by the updated driver for this HBA which is included in SLES9 SP4's kernel (which is also used in NLD9 SP4 and in online updates to OES1 SP2).
Workaround (updated system)
When updating the VMware virtualisation product is not an option, the issue can be worked around as follows:
- Follow the "boot installed system" procedure documented in TID 3864925 - Troubleshooting Common Boot Issues to regain access to the installed system.
- Downgrade to the last pre-SP4 kernel, 2.6.5-7.287.3. This kernel is available for x86 (i386) as patch-11835 and for x86-64 as patch-11815.
Start the installation using SP3 CD1 rather than SP4 CD1, then apply SP4 once the installation has been completed.
Note
Although the symptoms are quite similar, this issue is different from the one documented in TID 3494481 - System fails to boot after kernel update. That issue affects physical systems however and is caused by device renaming, not by no hard disks being detected.