SNMP query for hrStorage triggers mount of all autofs maps causing timeouts

Issue

• When the snmpd agent is polled over SNMP for the hrStorage table from the HOST-RESOURCES-MIB, all active autofs maps are automatically mounted.
• When there is a large number of map entries, or the map contains an entry with an NFS mountpoint on an unreachable NFS server, the SNMP query will timeout or stop responding to other queries eventually.
• The autofs mounts are never unmounted when snmpd is polled frequently enough.
• The snmpd.conf option skipNFSInHostResources true does not prevent the automatic mount or the eventual timeout.
• The timeout does not occur every time.
• Versions of net-snmp-5.7.2-32 and more recent are not affected.