I am using hd-idle (see link) to spin down my one external hard drive on my RPI server. It is not used for large parts of the day and night so it has been quite useful to set up hd-idle, which spins down the drive after an hour or so of no activity.
Now hd-idle can generate a log file where it notes down some data, e.g. when the drive was spun down, how long it was running, what time it spun down.
You can read the file to get an impression how well it works, but I’d like to see the data visualised or analysed in some way. Seeing the past month of how often per day the drive was spun down, or average length of long it was running and so on.
Searching online I couldn’t really find anything. Maybe anybody here knows more? Or what ways of recording and looking at this type of data are you using?
I don’t know how hd-idle stores its data, but sysstat and some other utilities will log I/O on a per-device data and there are utilities that can graph it.
Data is stored in a log file, which is why I wondered if someone already made a solution that just parses that and presents it as graphs. Couldn’t find that myself but seems like it does not exist.
I recommend generating some metrics from the logs and graphing them yourself.
Perhaps the free Grafana plan would have what you need to parse the log files and visualize the metrics you want.
I get your point and it was an emergency option to at some point take time and do it myself, sure. But that is not what I asked, not even sure when I have the time to do this.
There aren’t log visualizers for every artisanal log file format. But there’s a movement towards supporting JSON format logs for more services, and lots tools that can understand JSON logs making generating graphs and metrics from arbitrary logs fairly efficient.
If this tool is making the logs harder to parse by using a custom format, that’s something the tool could improve.
Some apps support both plaintext logs for humans and JSON logs for tools.
