downtime / reboot grace period

[xs-]

Would it be possible to somehow create a downtime / reboot grace period for all triggers of a host? (yes it could be part of the trigger checks, but 1.4 does not allow cross template trigger dependency).

When a host reboots (host up/down check by ping), it's possible the services (also monitored by zabbix) take some time to start. It's quite annoying to get these semi false triggers because of this.

[cstackpole]

I believe that with a quick search on the forum you will find many who agree. It has been talked about for a while. Sometimes it is a reboot of a server and sometimes it is a reset of an application.

I know this conversation was discussed quite a bit in the 1.6 road map forum post as a "maintenance" feature that holds all alerts while activated. I do not know if it is in 1.5 yet (heve not looked for it myself).

If I get time later today I will try to find it.

Have fun!

[xs-]

Hmm, i'll try and search the forums more thoroughly next time

For the time being we thought of a theoretical workaround for which we haven't been able to find a technical solution.
Most of our services checks are done by the zabbix agent using active checks. Our theory is that if we stop the zabbix agent first on shutdown and start it last on boot, the described issue would be reduced to a minimum. The problem is that windows doesn't support boot orders for services, only dependencies (which, after some initial tries, are not usable for this).

Does anyone here know if this can be done somehow? (start zabbix last on boot and stop it first on shutdown).