Navigating through "storage > persistentmemory" on host UI client may crash the hostd service if the DIMM goes into fatal error
search cancel

Navigating through "storage > persistentmemory" on host UI client may crash the hostd service if the DIMM goes into fatal error

book

Article ID: 321013

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

Symptoms:
When a DIMM fails with a fatal error, and a user logs into the DCUI to access the datastore/persistent memory in the Graphical User Interface (GUI), you see these symptoms:
  • The GUI hangs.
  • The hostd user mode process terminates with a core dump.
  • You see a backtrace similar to:

    2019-03-29T15:41:11.071Z panic hostd[1001394896] [Originator@6876 sub=Default opID=11715a96 user=root]
    -->
    --> Panic: NOT_REACHED bora/vim/hostd/hostsvc/provider/vmkernel/VmkNvdimmSystemProviderImpl.cpp:513
    -->
    --> Backtrace:
    --> [backtrace begin] product: VMware ESX, version: 6.7.0, build: build-10341481, tag: hostd, cpu: x86_64, os: esx, buildType: beta
    --> backtrace[00] libvmacore.so[0x003BA4D0]: Vmacore::System::Stacktrace::CaptureFullWork(unsigned int)
    --> backtrace[01] libvmacore.so[0x001DBE92]: Vmacore::System::SystemFactory::CreateBacktrace(Vmacore::Ref<Vmacore::System::Backtrace>&)
    --> backtrace[02] libvmacore.so[0x003B70DA]
    --> backtrace[03] libvmacore.so[0x003B71B1]: Vmacore::PanicExit(char const*)
    --> backtrace[04] libvmacore.so[0x001A16AC]
    --> backtrace[05] libvmacore.so[0x001A175C]
    --> backtrace[06] hostd[0x00A0F9BE]
    --> backtrace[07] hostd[0x00A1032B]
    --> backtrace[08] hostd[0x0085B918]
    --> backtrace[09] libvim-types.so[0x0156F1B4]
    --> backtrace[10] libvmomi.so[0x00102492]
    --> backtrace[11] libvmomi.so[0x001030B6]: Vmomi::PropertyProviderMixin::_RetrieveContentsEx(Vmomi::PropertySet const&, bool, bool, Vmacore::Authorize::AuthEntity*, bool, Vmacore::Ref<Vmomi::Core::PropertyCollector::ObjectContent>&)
    --> backtrace[12] libvmomi.so[0x001FE30A]
    --> backtrace[13] libvmomi.so[0x00203F70]
    --> backtrace[14] libvmomi.so[0x00204914]
    --> backtrace[15] libvmomi.so[0x001FEB7A]
    --> backtrace[16] libvmomi.so[0x000E80BD]
    --> backtrace[17] libvmomi.so[0x000E84CD]
    --> backtrace[18] libvmomi.so[0x00111468]
    --> backtrace[19] libvmomi.so[0x002451DC]
    --> backtrace[20] hostd[0x00DE571F]
    --> backtrace[21] libvmacore.so[0x002CA6CC]
    --> backtrace[22] libvmacore.so[0x002CADB9]
    --> backtrace[23] libvmacore.so[0x003C326E]
    --> backtrace[24] libpthread.so.0[0x00007D3B]
    --> backtrace[25] libc.so.6[0x000E9B1D]
    --> [backtrace end]


    Note: The preceding log excerpts are only examples. Date, time, and environmental variables may vary depending on your environment.


Environment

VMware vSphere ESXi 6.7

Cause

This issue occurs due to a faulty DIMM or hardware which causes the fatal error.

Resolution

This issue is resolved in VMware vCenter Server 6.7 Update 3, available at VMware Downloads.

Since hostd is restarted automatically, the HTML client and other hostd dependent operations continue to work.

Workaround:
To work around this issue if you do not want to upgrade, do not access the faulty persistent memory area in the GUI until the faulty DIMM is replaced.