A brief introduction:

I have set a maintenance period with data collection for all hosts during which I run a vacuum full on my Zabbix database. The maintenance starts 10 minutes before the vacuum and ends about 10-15 minutes after it completes. In 2.0.9, everything was OK - no alerts were generated during that period.

After upgrading to 2.2.0 however, I'm receiving alerts from every possible nodata() trigger I have set during the database vacuum. Those alerts don't seem to show up in the web interface, and only PROBLEMs are being sent - no OK alert follows them.

The server log (with DebugLevel=3 set) shows that the maintenance is turned on and off properly:

18343:20131206:113001.516 putting host [vm1] into maintenance (with data collection)
18343:20131206:113001.517 putting host [vm2] into maintenance (with data collection)
18343:20131206:113001.518 putting host [vm3] into maintenance (with data collection)
<snip>
18343:20131206:120001.279 taking host [vm1] out of maintenance
18343:20131206:120001.279 taking host [vm2] out of maintenance
18343:20131206:120001.279 taking host [vm3] out of maintenance

If there is any more information I can provide, please don't hesitate to ask.