To expand on my "Home-Lab" topic... What time-based logging or other system-based logging are you capturing and making available from your smart home? Are you someone that uses
nfluxDb and Grafana or some other setup? Do you bring together your smart home and broader network to provide a more comprehensive logging view? What tools do you use to achieve this?
What features or Community options do you use to monitor your HE setup?
There is a old saying: If you can't measure it, it didn't happen. I am using just regular old InfluxDB and Grafana. Actually been on the same instance of InfluxDB for 3 years now and working great. I find that most system have no issue with either the telegraf utility or platforms like Hubitat and Proxmox have their own agent.
As for uptime and system events that is a different story have gone around and around on this UptimeKuma (simple) to Zabbix (not so simple but does SNMP), I started some n8n monitors and using AI to send me notifications which is cool but darn their flows can get out of hand really fast.
For syslog collecting and firewall flow collection Graylog is good but can take a while to create the rules to make logs useable in their dashboards.
I can't say I have a particularly extensive monitoring setup, but what I do have is still using InfluxDB and Grafana.... I would like to setup something more holistic to actually monitor and alert me when things are not right... not just in HE...
In the HE space it would be the kind of things we would expect, devices that are not reporting as expected, memory and CPU utilisation, etc.
More broadly, across the network, it could be things like (Docker) containers, servers or other services not running, disk space, memory and CPU, etc...
I have dabbled in Nagios... But am considering alternatives....
This is the main thing I have monitored in the past using InfluxDB and Grafana, making use of the Hub Info driver. That said, in recent times that has been required less...
I have NodeRed connected to the event and log sockets and drop those updates into a MariaDB running on my NAS. Main purpose is to update a devices table with last update so I can monitor when a device falls off the network or need a battery replacement. I also log battery changes so I know how long they last vs replacing them prematurely.
The logs have also come in handy when troubleshooting something where I can look back to see if something happened before. Since I have 360 devices it lets me look back at debug and errors since logs on the hub donβt last more than 12 hours. This is useful if a LAN device has connection issues.
I also use InfluxDB for graphs and trends but MariaDB provides much more granular detail.
Primary uptime/fault monitoring: Uptime Kuma. Two instances with the secondary one monitoring the primary one.
Logging: NodeRED connected via MakerAPI and Hubitat Logsocket -> MySQL and InfluxDB (MySQL for logs, InfluxDB for device data) -> Grafana. I keep 4 weeks of hub logs in MySQL. Data in InfluxDB are kept longer. I have the logs from both of my hubs combined in a single data table in Grafana which helps with troubleshooting things that work across both hubs.
Everything is running under Docker on a pair of Raspberry Pi 4Bs.
I started logging data to Google sheets to display as graphs years ago when using Webcore on SmartThings, then I brought it with me to Hubitat, and I now I just use a custom app I wrote to log to data to Google sheets from my custom apps.
I've tried some of the graphing apps like Watch Tower, but in the end it is the control I have over what I send to Google sheets and the control I have to make graphs look how I want to that keeps me logging to Google.
I publish the charts in sheets, and then I can use that URL to the chart to put the graphs on my dashbaords.