ESXi 5.5 or 6.0 host disconnects from vCenter Server

VMware vSphere ESXi

ESXi host losing connection to vCenter Server is resolved in ESXi 5.5 Patch 08 and ESXi 6.0 Patch 03.

Symptoms:

In the /var/log/vpxa.log file, you see entries similar to:

YYYY-MM-DDTHH:MM:SS.SSSZ [2A330B70 info 'commonvpxLro' opID=778b566a-e7] [VpxLRO] -- BEGIN task-internal-402632 -- -- vpxapi.VpxaService.queryBatchPerformanceStatistics -- 05a84037-c508-e746-d9ae-f6757232865c
In the /var/log/vmkernel.log file, you see entries similar to:

YYYY-MM-DDTHH:MM:SS.SSSZ [40440B70 info 'Vimsvc.ha-eventmgr'] Event 530430 : /usr/lib/vmware/vpxa/bin/vpxa crashed (1 time(s) so far) and a core file might have been created at /var/core/vpxa-worker-zdump.000. This might have caused connections to the host to be dropped.
YYYY-MM-DDTHH:MM:SS.SSSZ cpu0:6126820)User: 2888: wantCoreDump : vpxa-worker -enabled : 1 2016-02-17T05:29:33.367Z cpu0:6126820)UserDump: 1820: Dumping cartel 6123450 (from world 6126820) to file /var/core/vpxa-worker-zdump.000 ...
YYYY-MM-DDTHH:MM:SS.SSSZ cpu0:6126820)User: 2888: wantCoreDump : vpxa-worker -enabled : 1
YYYY-MM-DDTHH:MM:SS.SSSZ cpu0:6126820)UserDump: 1820: Dumping cartel 6123450 (from world 6126820) to file /var/core/vpxa-worker-zdump.000 ...
YYYY-MM-DDTHH:MM:SS.SSSZ cpu18:6126820)UserDump: 1944: Userworld coredump complete.

VMware vSphere ESXi 6.0
VMware vSphere ESXi 5.1
VMware vSphere ESXi 5.5

This issue occurs when too many performance counters are being registered on the ESXi hosts vpxa process causing the process to run out of memory.

This is a known issue affecting ESXi 5.5 and 6.0.

This issue is resolved in ESXi 5.5 Patch 08 and ESXi 6.0 Patch 03.

For ESXi 5.5

This issue is resolved in ESXi 5.5 Patch 08. For more information, see VMware ESXi 5.5, Patch ESXi550-201608401-BG: Updates esx-base (2144361)

For ESXi 6.0

This issue is resolved in ESXi 6.0 Patch 03. For more information, see VMware ESXi 6.0, Patch ESXi-6.0.0-20160804001-standard (2145667)

To work around this issue, increase the default memory allocation to the vpxa process on the ESXi Host.

Note: This workaround may only reduce the occurrences of this issue.

Connect to the affected ESXi host with an SSH session.
Run this command to set the grpID of the vpxa process to a variable:

grpID=$(vsish -e set /sched/groupPathNameToID host vim vmvisor vpxa | cut -d' ' -f 1)
Run this command to increase the max memory allocation of the vpxa process to 400MB (default value is 304):

vsish -e set /sched/groups/$grpID/memAllocationInMB max=400 minLimit=unlimited
Verify that the max memory allocation of the vpxa process is changed:

vsish -e get /sched/groups/$grpID/memAllocationInMB

For example:
vsish -e get /sched/groups/$grpID/memAllocationInMB

sched-allocation {
min:0
max:400
shares:0
minLimit:-1
units:units: 3 -> mb
}

Note: This setting does not persist across reboot.

To be alerted when this document is updated, click the Subscribe to Article link in the Actions box..