Sfcbd service restarts continuously due to lack of memory resource

Products

• VMware vSphere ESXi

Issue/Introduction

Symptoms:

• Hardware status of the ESXi host is unavailable in the vSphere Web Client and vSphere Host Client.
• The Sfcbd service keeps restarting on the ESXi host.
• You see the following message appear in DCUI occasionally.

WorldList: malloc failed when allocating allNodes list

Environment

VMware vSphere ESXi 6.0

Cause

This issue occurs due to a lack of free memory resource in sfcb resource group causes the Sfcb-watchdog script to attempt to restart the sfcb service repeatedly.

The Sfcb resource group includes the sfcb provider manager, http daamon, https daemon and sfcb-watchdog script and its child processes, such as the ps command.

Resolution

This is a known issue affecting ESXi 6.0.

To resolve the issue, use one of these options:

•  For ESXi hosts that are on a build prior to ESXi 6.0 P04

1. Connect to the ESXi host with an SSH session.
2. Run this command:

/usr/lib/vmware/rp/bin/configRP setRPMemMaxSize host/vim/vmvisor/sfcb 70

Note: This workaround does not persist between reboots.

• For ESXi hosts on ESXi 6.0 Patch 4 or later

1. Connect to the ESXi host with an SSH session.
2. Run these commands:

esxcfg-advcfg -A CIMOemSfcbRPMemMax --add-desc \'Maximum Memory for sfcb RP\' --add-default 70 --add-type int --add-min 30 --add-max 100

/etc/init.d/sfcbd-watchdog restart