After upgrading to vCenter Server 8.0, core files of the SNMP service fill up the /storage/core directory

Products

• VMware vCenter Server

Issue/Introduction

Symptoms:
The system dumps a core in /var/log/vmware/messages when SNMPD is started.

2023-03-29T15:59:19.442724+02:00 vcsa snmpd[88617]: P3 5: load_entmib: fetch_sysinfo failed, no entPhysicalTable entry
2023-03-29T15:59:19.442795+02:00 vcsa snmpd[88617]: P4 6: serial number longer than rfc 4133 allows, truncating
2023-03-29T15:59:19.680396+02:00 vcsa systemd[1]: snmpd.service: Main process exited, code=dumped, status=11/SEGV
2023-03-29T15:59:19.695691+02:00 vcsa vmware-snmpd: INFO: firewall close port 161 command result rc=0
2023-03-29T15:59:19.712997+02:00 vcsa systemd[1]: snmpd.service: Failed with result 'core-dump'.
2023-03-29T15:59:25.599393+02:00 vcsa systemd[1]: Stopped VMware SNMP Server Daemon.

Environment

VMware vCenter Server 8.0.x
VMware vCenter Server 7.0.x

Cause

The buffer for cpuinfo runs out and causes SNMPD to core dump.

Resolution

This issue is resolved in vCenter Server 8.0 P02.


Workaround:

1. SSH to the vCSA via root
2. Stop the snmpd service:

systemctl stop snmpd

3. Disable snmpd at start up with the following command:

systemctl disable snmpd