Since Zabbix server 3.0.0 (up to Zabbix server 3.0.2, there is no 3.0.3 yet in the official repos), if you reboot zabbix server, the following issue is observed:

Condition: zabbix-server has been running for at lease a couple of days (e.g. if you reboot it with minimal downtime - issue will not reproduce).

All Windows hosts with the following trigger will start alerting:
{Template OS Windows:system.uptime.change(0)}<0

However, all Linux hosts with the same trigger will NOT be alerting:
{Template OS Linux:system.uptime.change(0)}<0

The evaluation is somehow broken (e.g. zabbix server POSSIBLY evaluates the trigger for Windows template as OWN uptime, although the item itself is properly configured to pull data from remote Windows Agent (item in Windows Template: system.uptime). I was able to reproduce the issue several times (e.g each time I reboot zabbix-server providing it has been running for some time, e.g. NOT after a fresh reboot).

The uptime for hosts in all templates will continue to ONLY increase (seen in the graphs).

More details below:
Latest values for uptime:
2016-05-19 09:42:31
514802
2016-05-19 09:41:31
514742
2016-05-19 09:40:31
514682
2016-05-19 09:40:31
514619
2016-05-19 09:39:31
514560

Trigger history:
OK 2016-05-19 09:41:31 17m 24s 17m 24s No
PROBLEM 2016-05-19 09:40:31 18m 24s 1m No
OK 2016-05-13 10:45:31 5d 23h 13m 5d 22h 55m No