below

Not the same kind of monitor tool exactly, but, I keep finding dstat hard to leave behind. Because I can see the past there! So many of these monitors have one or two or three over time graphs, but most of the information is ephemeral, only shows right now. But I really want to see network use, disk use, paging, context switching/interrupts over time! There is also Below. Which has a much more htop/btop like interface than dstat. Below records system info over time, and allows time travel! However, it's not as convenient as dstat, not at a glance, as one has to to scrub through time. But it is pretty impressive system monitoring, great for what it is! It's per-process pressure metrics are also utterly unbeatable. The way it rolls up cgroups is also stellar. https://github.com/facebookincubator/below

Interesting post. We wrote this “below” utility [1] that monitor system metrics similar to atop. We want the ability to collect all metrics into a single object, pass it around and visualize it elsewhere. Naturally we need some way to query into fields or even nested-struct fields. For example, to get the file cache usage of a particular process, we need to go through sample->processes->pid->memory->file cache. To do it ergonomically and also type-safely, we end up using proc macro to generate enums that represent field paths of the structs and then use them to query values of non-struct (leaf) fields. I always wonder if there are simpler ways or existing proc macro derives to safe us the efforts. Maybe I do need to look into serde internals for some inspirations. [1] https://github.com/facebookincubator/below/blob/main/below/b...

Was using Prometheus for a while but have kind of fallen off the bandwagon & need to get back on! I also hear good things about Facebook's Below: > A time traveling resource monitor for modern Linux systems https://github.com/facebookincubator/below Right off the bat, in their comparison (https://github.com/facebookincubator/below/blob/main/docs/co...), they call htop a Wonderful terminal interface. Kind. But the list of limitations is severe. No view over time, no cgroup support, and very limited stats available.

IO wait is counted against load by linux. So high IO pressure, i.e. found in `/proc/pressure` will cause high load. When considering reclaim (dropped file pages which then need IO to read the page on demand again) it makes sense because high IO wait will delay reads (and writes) slowing everything down. I'm liking this project https://github.com/facebookincubator/below It's packaged in Fedora.

I created something similar for our project “below” (https://github.com/facebookincubator/below/blob/main/below/b...). The program collects system resource metrics into a data structure and we need to display the fields with different styles and formats. In order to decouple the data structure from rendering, Queriable (Keyable) and FieldId (combine KeyPath + mirror struct into enum) are used. I will definitely like to checkout the KeyPath implementation as it seems more general.

I have been using glances for a terminal sysmon but I don't like that it eats so much RAM. Can someone please recommend a system monitor that is easy to comprehend and less resource hungry? I am also curious about below [0] since it came up recently. [0] https://github.com/facebookincubator/below