ESXi host is disconnected from vCenterServer and it is in an unresponsive state
search cancel

ESXi host is disconnected from vCenterServer and it is in an unresponsive state

book

Article ID: 344759

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

Symptoms:

  • ESXi host is disconnected from vCenterSever and it gets no response.
- SSH and DCUI can connect to ESXi host
- PING response has no problem
- ESXi host reconnection fails
- No direct connection from vSphereClient to ESXi host
- Restarting the service of vpxa / hostd / rhttpproxy will not be resolved
- Restarting service.sh is not resolved
- Restarting the ESXi host will not resolve it
 
  • A situation where the Hostd service repeatedly crashes from the log of the ESXi host.
  • In the situation before the crash occurred, "The file specified is not a virtual disk" is logged in the hostd log.
  • In the virtual machine's .vmx file, "scsix: x.fileName" is set to NULL.
  • The vmdk file exists in the directory of the virtual machine.
 
  • Virtual machine configuration information (. vmx):
　　scsix:x.deviceType = "scsi-hardDisk"
　　scsix:x.fileName = ""

hostd log：

[YYYY-MM-DDTHH:MM:SS].732Z info hostd[79E44B70] [Originator@6876 sub=DiskLib opID=62557529-1-ec11 user=vpxuser:XXXXX-VSPHERE.LOCAL\XXXXXXXXXXXXXX] DISKLIB-LIB : Failed to open '/vmfs/volumes/578xxxxx-xxxxxxxx-xxxx-0025xxxxxxx/XXXXXXX' with flags 0x110015 The file specified is not a virtual disk (15). 
[YYYY-MM-DDTHH:MM:SS].732Z info hostd[79E44B70] [Originator@6876 sub=DiskLib opID=62557529-1-ec11 user=vpxuser:XXXXX-VSPHERE.LOCAL\XXXXXXXXXXXXXX] DISKLIB-LIB_MISC : DiskLibEnumExtents: failed to open '/vmfs/volumes/578xxxxx-xxxxxxxx-xxxx-0025xxxxxxx/XXXXXXX': The file specified is not a virtual disk (15). 
[YYYY-MM-DDTHH:MM:SS].732Z error hostd[79E44B70] [Originator@6876 sub=Vmsvc.vm:/vmfs/volumes/578de40f-0d70dfc2-e746-0025b591010f/XXXXX/XXXXX.vmx opID=62557529-1-ec11 user=vpxuser:XXXXX-VSPHERE.LOCAL\XXXXXXXXXXXXXX] Failed to enumerate disk [XXXXXXX] XXXXXXX, The file specified is not a virtual disk 
[YYYY-MM-DDTHH:MM:SS].739Z warning hostd[79E44B70] [Originator@6876 sub=Default opID=62557529-1-ec11 user=vpxuser:XXXXX-VSPHERE.LOCAL\XXXXXXXXXXXXXX] FileLayoutExToFileLayout: While populating disk layoutSkipping empty diskLayout configuration for VM [XXXXXXX] XXXXXXX/XXXXXXX.vmx, key 2000 
[YYYY-MM-DDTHH:MM:SS].740Z info hostd[796CDB70] [Originator@6876 sub=Snmpsvc] VmConfigListener: vm state change received, queueing reload request

[YYYY-MM-DDTHH:MM:SS].741Z info hostd[79E44B70] [Originator@6876 sub=Libs opID=62557529-1-ec11 user=vpxuser:XXXXX-VSPHERE.LOCAL\XXXXXXXXXXXXXX] Error getting extended config name from config file.
[YYYY-MM-DDTHH:MM:SS].741Z info hostd[79E44B70] [Originator@6876 sub=Libs opID=62557529-1-ec11 user=vpxuser:XXXXX-VSPHERE.LOCAL\XXXXXXXXXXXXXX] Error getting extended config file path.
[YYYY-MM-DDTHH:MM:SS].018Z Section for VMware ESX, pid=19965294, version=6.0.0, build=6921384, option=Release
[YYYY-MM-DDTHH:MM:SS].018Z verbose hostd[FFF24B20] [Originator@6876 sub=Default] Dumping early logs:

vpxa log：
[YYYY-MM-DDTHH:MM:SS].217Z error vpxa[7DDBBB70] [Originator@6876 sub=vpxaInvtHostCnx opID=WFU-1f0c1d49] [VpxaInvtHost] Can't connect to hostd/serverd. Shutting down...
[YYYY-MM-DDTHH:MM:SS].217Z error vpxa[7DF1EB70] [Originator@6876 sub=hostdstats] [PollCurrentStats] Failed to fetch current stats.  Fault:
[YYYY-MM-DDTHH:MM:SS].217Z info vpxa[7DDBBB70] [Originator@6876 sub=Default opID=WFU-1f0c1d49] [Vpxa] Shutting down now

vmkernel log：
[YYYY-MM-DDTHH:MM:SS].746Z cpu42:35335 opID=8d848a42)User: 3820: hostd-worker: wantCoreDump:hostd-worker signal:6 exitCode:0 coredump:enabled
[YYYY-MM-DDTHH:MM:SS].883Z cpu42:35335 opID=8d848a42)UserDump: 2074: hostd-worker: Dumping cartel 34536 (from world 35335) to file /var/core/hostd-worker-zdump.000 ...
[YYYY-MM-DDTHH:MM:SS].201Z cpu12:35335 opID=8d848a42)UserDump: 2194: hostd-worker: Userworld coredump complete.

vobd log：
[YYYY-MM-DDTHH:MM:SS].201Z: [UserWorldCorrelator] 28682044986303us: [vob.uw.core.dumped] /bin/hostd(34536) /var/core/hostd-worker-zdump.000
[YYYY-MM-DDTHH:MM:SS].201Z: [UserWorldCorrelator] 28682457971729us: [esx.problem.hostd.core.dumped] /bin/hostd crashed (1 time(s) so far) and a core file may have been created at /var/core/hostd-worker-zdump.000. This may have caused connections to the host to be dropped.
[YYYY-MM-DDTHH:MM:SS].202Z: An event (esx.problem.hostd.core.dumped) could not be sent immediately to hostd; queueing for retry.

 

Environment

VMware vSphere ESXi 6.0

Cause

This is a known issue that occurs because the vmdk file name is not set to "scsix: x.fileName" of the virtual machine's .vmx file.

Resolution

Issue has been resolved in ESXi 6.0 Patch 07.


Workaround:
It is possible to avoid it by setting the vmdk file name to "scsix: x.fileName" of the virtual machine's .vmx file



Powered by Wolken Software