I'm a Systems Reliability and DevOps engineer for Netdata Inc. When not working, I enjoy studying linguistics and history, playing video games, and cooking all kinds of international cuisine.
DISCLAIMER: I work for the company that makes the software I'm about to suggest. Despite this, my opinions below are entirely my own and in no way representative of those of my employer.
It's dead simple to install and set up. Most cases require little to no manual configuration at all.
When you do need to configure it, the configuration is reasonably easy to understand.
It comes with it's own built-in data storage, unlike some other options which require you to provide your own.
It comes with it's own built-in web dashboard for viewing the collected data, unlike some other options which require you to provide your own.
It has very low overhead for what it does. As a sample, on my home server, I'm seeing an average of less than 1% CPU usage and less than 256MB of RAM usage to collect just short of 4000 metrics at per-second resolution. Most of the RAM usage is actually configurable in-memory caching of the data too.
It's got easy integrations for a wide variety of alerting mechanisms and easy configuration for automated alarms.
Writing new collectors is dead simple. The interface between data collectors and the main process is a reasonably easy to understand and implement plain-text protocol over stdin/stdout of the collector itself.
Cons:
Support on platforms other than Linux is somewhat limited (right now). It runs fine on FreeBSD (mostly) and macOS (with limited reporting) but not really on anything else. Windows support is planned but probably quite a ways off, and other UNIX-like platforms are pretty much not supported unless they look from userspace sufficiently like Linux/FreeBSD.
The library of data collectors is somewhat limited compared to many other offerings (though it's improving over time).
Centralized data collection functionality is somewhat limited right now. You can stream from multiple systems to one central system, but it's a pain to get aggregate data from that main system. There's a cloud-based option to remedy this in the works though.
Native binary packages are somewhat lacking on many distributions (or they're horribly out of date).
I'm a Systems Reliability and DevOps engineer for Netdata Inc. When not working, I enjoy studying linguistics and history, playing video games, and cooking all kinds of international cuisine.
Yeah, there are times that the script is a bit non-obvious. FWIW, if you're running a recent version of Ubuntu, Debian, Fedora, OpenSUSE, or CentOS (or a compatible derivative of any of them), we do have prebuilt binary packages.
I'm a Systems Reliability and DevOps engineer for Netdata Inc. When not working, I enjoy studying linguistics and history, playing video games, and cooking all kinds of international cuisine.
All of the official documentation has been moved to docs.netdata.cloud. I wasn't ever heavily involved with the documentation, but I think the gitbook was just an early attempt to modernize the documentation so that it wasn't just sitting in a GitHub Wiki.
I'm a Systems Reliability and DevOps engineer for Netdata Inc. When not working, I enjoy studying linguistics and history, playing video games, and cooking all kinds of international cuisine.
Yeah, the dashboard was actually one of the big things that got me started using Netdata in the first place, not just because it's an integral part of Netdata itself (I was used to stuff that required you to provide your own dashboard and/or charting tools), but because I find it so easy to understand. It doesn't try to throw huge numbers of options at you, it just shows you the data.
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
DISCLAIMER: I work for the company that makes the software I'm about to suggest. Despite this, my opinions below are entirely my own and in no way representative of those of my employer.
My personal favorite is Netdata.
Pros:
Cons:
thanks i will try it out ! liking the installer already :)
erm but would be nice is default option was STABLE and that NO meant NO it seems to be ignoring me saying NO and continuing to run the script
Yeah, there are times that the script is a bit non-obvious. FWIW, if you're running a recent version of Ubuntu, Debian, Fedora, OpenSUSE, or CentOS (or a compatible derivative of any of them), we do have prebuilt binary packages.
sorry but STABLE SHOULD BE DEFAULT do you need a github PR to get this fixed ?
also there are some out of date docs that dont mention you get unstable without the stable flag
EASIEST if you just change DEFAULT behaviour to STABLE
PS still loving NETDATA will install on 4 more servers later today :)
WOW debs and rpms TOO !
you need to publicise this BETTER ! but no problemo i love watching the build script building !
PS and why not add OracleLinux to your list of supported distros !
why is your gitbook archived ?
All of the official documentation has been moved to docs.netdata.cloud. I wasn't ever heavily involved with the documentation, but I think the gitbook was just an early attempt to modernize the documentation so that it wasn't just sitting in a GitHub Wiki.
cool thanks i will fork the old docs to learn about gitdocs
PS is there a link on the old repo to docs.netdata.cloud
love the graphics !
dev-to-uploads.s3.amazonaws.com/i/...
Yeah, the dashboard was actually one of the big things that got me started using Netdata in the first place, not just because it's an integral part of Netdata itself (I was used to stuff that required you to provide your own dashboard and/or charting tools), but because I find it so easy to understand. It doesn't try to throw huge numbers of options at you, it just shows you the data.