After the upgrade to the latest release, receiving many emails like:

Subject: wpcaldc01: PROBLEM, Zabbix agent on wpcaldc01 is unreachable for 5 minutes
Message:
Agent ping (wpcaldc01:agent.ping): Up (1)
Original event ID: 187879
Time: 14:27:00

---------

10 seconds later would be another one:
Subject: wpcaldc01: OK, Zabbix agent on wpcaldc01 is unreachable for 5 minutes
Message:
Original event ID: 187879
Time: 14:27:00
Age: 0m

wpcaldc01 is one of our domain controllers. Agent can be on any server and version: Linux or Windows, 2.2.x or 2.4.x

Network is good, no problem with connectivity, no errors on switch port, other services work fine.

Zabbix Server data:
all cache usages - 95-98% free
gathering process busy: pinger @ 11%, poller - 2-6%, the rest - 1% and less
internal process busy: with disabled alerts (due to garbage emails), all <1%.

OS load is zero:
%Cpu(s): 0.1 us, 0.1 sy, 0.0 ni, 99.7 id, 0.1 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem: 8190512 total, 7177008 used, 1013504 free, 197452 buffers
KiB Swap: 8380412 total, 40104 used, 8340308 free. 5194992 cached Mem

When received that email, in debug log (attached):
5070:20150819:142900.906 In zbx_vc_get_value_range() itemid:50077 value_type:3 seconds:0 count:1 timestamp:1440016140
5070:20150819:142900.906 End of zbx_vc_get_value_range():SUCCEED count:1 cached:1
5070:20150819:142900.906 End of evaluate_NODATA():SUCCEED
5070:20150819:142900.906 End of evaluate_function():SUCCEED value:'0'
5070:20150819:142900.906 In evaluate_function() function:'wpcaldc01:agent.ping.nodata(5m)'
5070:20150819:142900.906 In evaluate_NODATA()
5070:20150819:142900.906 In __get_function_parameter_uint31() parameters:'5m' Nparam:1
5070:20150819:142900.906 In substitute_simple_macros() data:'5m'
5070:20150819:142900.906 __get_function_parameter_uint31() flag:0 value:300