Having kind of a strange issue. I have a few alerts in our system that will trigger more than once (sometimes 4 times) before the first triggered alert has even cleared. This is NOT a chattering alarm because the first instance of the alert will not clear. It will show the alert as active and while it is still active it will trigger a new instance of the alert. The only reason I can tell is because if I look at the alert history there will be multiple rows that show the same alert and all of the rows have a blank CLEARED_TIMESTAMP. This becomes an issue for text alerts because each new instance sends a text alert. The alerts are set to AUTO acknowledge as well.
All of the alerts in my system are still using the Legacy Alerting features. I just recently upgraded to 7.6.3 but this problem was happening on the earlier version as well (I think it was 7.5.2, not sure though). Ignition server is a Windows 7 PC.
Our users are starting to get frustrated with the multiple text alerts and I am at a loss for how to fix the issue. Appreciate any help.
Thanks,
-Will
After restarting my gateway it seems to have cleared up some of the issues but not all. We don’t get as many instances of the same alert now. However, what happens now is we will sometimes get a second instance of the alert when the alert clears. This does not always happen but it does happen on a pretty regular basis. For example, we will have an alert trigger and then 2 minutes later that same alert will trigger again AND clear both instances of the alert at the same time. So if you look at the alert history you will see two lines for the same alert with different ACTIVE_TIMESTAMPS but having the same CLEARED_TIMESTAMPS. The other strange thing is that this alert also has a Active Delay of 10 minutes so it should be impossible for the alert to become active again only 2 minutes after the first instance. Operators are starting to get fed up with the text alerts and I cannot for the life of me figure out what the problem is. Any feedback would be greatly appreciated.
Thanks,
-Will
I’ve seen similar strange happenings with alarms. Occasionally I’ll see an alarm trigger that won’t clear. Then I go look at the status of the alarm tag in the designer, and the alarm is clear according to the tag, but it still shows up as an active alarm in the client. So far, I’ve been able to clear the alarm by disabling and then re-enabling the tag. I’m running 7.6.2.
Thanks for the feedback tall745. Seems to be pretty familiar to my issue.
I’m not too worried about the alert showing up on the screens it’s more of the text alerts being sent out multiple times. When half the text alerts you receive are just repeats of text alerts that have already been sent it starts to get frustrating and users start to ignore the text messages.
I haven’t been able to find anything that would point me to the source of the problem. I’ve attached a picture of an alert history table that shows the issue. The alert that is shown in the picture is a text message alert. The red boxes in the picture show where the text alerts were sent out multiple times before the first alert even cleared. Any and all feedback would be appreciated at this point, even if it is a wild shot in the blue.
Thanks,
-Will
Did you ever figure out how to resolve your issue? I’m having a problem which seems similar to this one. I have a pair of transaction groups to log events. They both use the same tag as a trigger, with one creating the row in the events table when the tag changes to true, and the other setting the event’s end time when the tag goes false. Both transaction groups are set to only execute once while their trigger is active, but I end up with two events being created without the first one getting its end time set. I don’t think this should be possible.
I asked support about it yesterday, so maybe they will have an answer.
Hi,
Since the active timestamps are actually different, I would first suspect that something might be happening that is causing Ignition or the whole machine to restart. If Ignition crashes for some reason, the “service wrapper” automatically restarts it, so you might not know it was happening.
Could you go to the install directory, and then to “logs”, zip up the “wrapper.log*” files there, and then attach them here, or email them to support? We can take a look and see if something like that is happening.
Regards,