Steps to reproduce:

1. Set in zabbix_server config parameter
   CacheUpdateFrequency=60

Result:
Every minute 

slow query: 3.816762 sec, "select distinct tt.triggertagid,tt.triggerid,tt.tag,tt.value from trigger_tag tt,triggers t,hosts h,items i,functions f where t.triggerid=tt.triggerid and t.flags<>2 and h.hostid=i.hostid and i.itemid=f.itemid and f.triggerid=tt.triggerid and h.status in (0,1)"

HashAggregate (cost=268197.80..324806.64 rows=1268142 width=50) (actual time=2860.942..3699.744 rows=1147489 loops=1)

Its consuming 50MB of mem and writes a temp file to disk for ~120MB each time.

It is unlikely that this can be called a bug, but perhaps this query can be optimized.

Expected:
Fast sync