I don't look at my logs unless I am in the process of changing / adding something. It would be useful if RM could trigger on any Error in the logs. The rule could be made to notify the user in some way of their choosing.
No such mechanism exists, nor should it. This would serve to amplify errors, and that's not desirable.
2 Likes
This what you're looking for I think. Use with caution!
2 Likes
OK Thank you.
Thanks.
I really don't get many/ any errors except when I changing or adding a special driver or app.
I'm not sure I will try it based on Bruce's warning. We'll see.
1 Like
Think about the idea that a bad driver / app could go haywire and throw lots of errors, and then you amplify those by running rules and sending notifications. Avalanche of bad stuff is the potential.
The better approach, imo, is to always keep a live Logs tab open, and simply look at it from time to time. If you see any errors, track them down, understand what causes them, and presumably fix whatever it is. I've got a driver or two that throw errors from time to time. It doesn't bother me and is basically harmless. One of these days I'll deal with it. But if I say a lot of red I'd get on it post haste. If I ever see red from a built-in app or driver, I know it has to be investigated right away.
2 Likes
I personally use NodeRed to do this. I have it setup listening to the event and log websockets and it inserts this data in a MariaDB running in my NAS. Primary use case is to monitor battery devices to let me know if the battery died or dropped off the network. I get a Pushover notification 2 times a day if a device hasnāt updated in X hours. I also query the logs table for any errors and I include the app and error count within that timeframe in that same push notification. From there I can act on it by logging into my hub to see. Some errors like Ecobee I just ignore.
2 Likes
The motivation and intent in your request is spot-on what "most of the people left to adopt HA" would want; I'm not talking all the roll-up-their-sleeves types that are already here. I'm talking EVERYBODY ELSE that is likely the bulk of the market remaining.
Actually....taking your idea further....you know that little "new message bubble" on the HE page, that's exactly where most of that market would likely want a little heads up....where HE itself is monitoring the logs and depending on the severity says...
"Hey, you might want to look at this developing situation cause it may be, or may soon be, causing a problem, ....and you might not know what to do with the info but just cut and paste it into the HE Community Form for help 
"
With all due respect to the folk that created the beast, and know everything about what's under-the-HE-hood, I think it's a step back in time to expect anyone but the folks that spend time in these forums to regularly look at the logs.
1 Like
There's a distinction to be made here between those who only use built-in apps and drivers, and those who venture into installing custom apps and drivers. I agree completely with respect to the first group, that looking at logs is something they might be directed to if something doesn't work as expected.
But, for the second group, once you are putting unknown and potentially crap software on your hub, you'd better be looking at logs, especially if things don't work as expected. If everything works great, no logs.
1 Like
Exactly why I removed the monitor I linked to above.
I see your point about the issue of multiple errors exploding.
Now first, this is not an important issue with me, just a thought. This thought came about because I can go a week or more without connecting to the Hub. So when looking for something to do I thought of an error notification rule. Honestly most of my automation involves lighting, failure to work is noticeable so this becomes my first line of notification.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.
