Release date: Apr 9, 2015

Summaries and Symptoms

This patch updates the esx-base VIB to resolve the following issues:

• When you use ALT+F12 to access the VMkernel logs from DCUI and scroll through the logs rapidly using the PageUp and PageDown keys, you might encounter heartbeat warnings and purple diagnostic screen with messages similar to the following:
  
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]SP_WaitLock@vmkernel#nover+0xe5 stack: 0xnnnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]SPLockWork@vmkernel#nover+0x26 stack: 0xnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]SVGAConsoleLock@vmkernel#nover+0x19 stack: 0x30
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]SVGAConsolePuts@vmkernel#nover+0x96 stack: 0xnnnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]Term_Putl@vmkernel#nover+0x1ac stack: 0xnnnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]LogTermScroll@vmkernel#nover+0xd8 stack: 0xnnnnnnnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]LogTermExecuteVScroll@vmkernel#nover+0x28 stack: 0xnnnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]LogTermInputCallback@vmkernel#nover+0x16c stack: 0x1
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]TermReceiveInput@vmkernel#nover+0x123 stack: 0xnnnnnnnnnnnn
  nnnn-nn-nnT11:53:50.595Z cpu0:nnnnn)0xnnnnnnnnnnnn:[0xnnnnnnnnnnnn]KeyboardBH@vmkernel#nover+0x1e stack: 0xnnnnnnnnnnnn
  
  
  
• Stateless hosts might fail to apply a host profile during boot. An error message similar to the following is displayed:
  
  A specified parameter was not correct: dvsName
  
  
• After you enable Fault Tolerance on a Windows VM (Windows XP and later) configured with Virtual Hardware version 11 and vSphere 6.0 VM Tools, the virtual machine's workload throughput will drop significantly.
  
  
• When you upgrade to the ESXi 6.0 host that is configured as stateless and managed by Auto Deploy, the valid ESXi 6.0 license might not be pushed by vCenter Server before creation of management vmknic due to a race condition. As a result, the creation of management interface might fail in the hosts and display the IP address as 0.0.0.0.
  
  
• During NUMA remapping of virtual machines configured with less memory (<128 MB) and many NUMA clients, an ESXi host might fail with a purple diagnostic screen. The purple diagnostic screen might contain traces similar to the following:
  
  PCPU xx: no heartbeat (2/2 IPIs received)
  [XXXXXXX]NUMASched_RemapCandidate@vmkernel#nover+0xXX
  [XXXXXXX]VmMemRemap_SelectForNUMARemap@vmkernel#nover+0xXX
  [XXXXXXX]VmAssistantProcessTasks@vmkernel#nover+0xXX
  [XXXXXXX]CpuSched_StartWorld@vmkernel#nover+0xXX
  
  

Patch Download and Installation