I have a remote outstation linked back to a central scada via a VPN and the central scada dials out sms alarms via an RV50. VPN is a bit shaky, drops out occasionally but always re-establishes itself.
Occasionally, alarms conditions occur in my remote PLC while the link is down. When the link is re-established the alarm appears in Ignition but fails to send an sms.
I have worked around this by adding an active delay to all my alarms, but surely this isn’t correct default behaviour? Am I missing something?
Try setting True the Notify Initial Event in the Alarming>General section.
The downside to this is that when you reboot Ignition it will blast out a bunch of alarms. I usually turn of my notification profiles, reboot, then after the system stabilizes I turn them back on. Luckily, Ignition is really stable and I have never had it crash and reboot on its own.
I agree with you that this should not be default behavior. It definitely used to not act this way before the new alerting system. You either choose not getting alarms, or getting blasted with false alarms. If I remember right when I was testing this, it mostly affected Digital Alarms. Analog alarms were not affected unless the value did not change.
That’s a shame. I do a lot of pumping stations in poor signal areas. They’re usually not time critical but the fact that they may NEVER receive the notification kind of undermines the whole thing.
I suppose I’ll adopt your suggestion in future. Adding five seconds to every alarm is a massive pain!
We have a similar issue; where the Alarm Active event handler is set up to trigger an email to send. Each time Ignition reboots (if the alarm is already active), the email sends again.
Is there not some sort of initialChange variable that could help here? I’ll look at the Notify Initial Events and see if that helps, but am a little confused about how that works.
We’ll try extending the ‘Continuous Event Detection Window’ and see if that helps
Any further thoughts on this issue? We have the ‘Notify Initial Events’ enabled but that causes duplicate alarms to be sent anytime we have to reboot the server. There must be a better way around this?
