Knowledge Base

• In a DRS cluster environment, the hostd service reaches a hard limit for memory usage, which causes hostd service to restart itself.

Symptom: The hostd service restarts and temporarily disconnects from VirtualCenter. The ESX Server host stops responding before hostd service reconnects. 

• Intermittent network and intermittent SAN issues occurring at the same time might result in conditions like unreachable SAN or high latencies. This situation might cause HA to power on the same virtual machine on a second ESX Server host, and leave this virtual machine registered on two different ESX hosts. Normally if there are intermittent SAN issues, the ESX Server host can temporarily lose locks on the .vmdk files of the powered-on virtual machines. Usually, the ESX Server host times out and, after a while, reclaims the lock. The ESX Server host loses the lock only when the same virtual machine powers on, on another ESX Server host. This is an unusual occurrence. The lock is lost only when there are network issues that make HA restart virtual machines on another host. So, if there are intermittent network and intermittent SAN issues, the virtual machine (which is running on the original ESX Server host that lost the lock to the same virtual machine currently started on the second ESX Server host) stops responding and becomes inaccessible, but shows as registered in both ESX Server hosts in the VirtualCenter Server.  With this fix, the virtual machine running on the original ESX Server host displays a dialog box with a message that states the locks have been lost and the virtual machine will be shutting down automatically.

• After an IBM SAN Volume Controller (SVC) is reset, it returns a Check Condition 2/0 0x2 0x4 0xc error code while SVC synchronizes cache with the partner node. With this fix, VMkernel marks this path as offline to cause a failover and use paths of another node (a node that is not reset) to send I/O data.

• A memory corruption condition may occur in the virtual machine hardware. A malicious request sent from the guest operating system to the virtual hardware may cause the virtual hardware to write to uncontrolled physical memory. The Common Vulnerabilities and Exposures project (cve.mitre.org) has assigned the name CVE-2008-4917 to this issue.