logdna-agent-v2
logdna-agent-v2 copied to clipboard
logdna
Docs for Linux installation are out of date
So I had fun installing the agent in a new Linux install. This page, https://github.com/logdna/logdna-agent-v2/#installing-on-linux, has a section
Installing on Linux
Refer to the documentation for [deploying the agent on Linux](https://github.com/logdna/logdna-agent-v2/blob/master/docs/LINUX.md).
Following the link there, specifically the configure the agent section (https://github.com/logdna/logdna-agent-v2/blob/master/docs/LINUX.md#configure-the-agent), I arrived at a section that is out of date. Those steps, specifically writing etc/logdna.env and setting the environment variables inside that file as specified in step 2 did not work.
What did work is following the steps in the Configuration section of the first page (https://github.com/logdna/logdna-agent-v2/#configuration) and setting up a yaml configuration file.
Please update the documentation to prevent others from wasting time trying outdated approaches to configuring the agent. Thanks!
etc/logdna.env is specified in logdna-agent systemd service config. does your linux uses systemd?
We do not use systemd on our systems, but the problem is that the docs specifically say to follow the instructions for installation on Linux, but that's not necessarily the correct approach. It wasn't until I read the Configuration section that I tried the approach of the yaml file and found that to work.
The doc does say that agent is using systemd. https://github.com/logdna/logdna-agent-v2/blob/master/docs/LINUX.md#usage. And all steps are done in that context.
I understand that, but if you don't read the entire docs, and are coming from the link provided in my original submission, it's not clear that you should use the yaml based approach if you're not using systemd. Basically, the options to configure the agent for Linux are spread across two pages, and there's not a clear explanation of `Use .env if you're using systemd, and if you're not, try using the yaml file). Your users shouldn't have to read the entire documentation simply to configure the agent; it should be clear what approach is recommended depending on the user's situation.
@holyoaks Thanks for your feedback! We'll bring this feedback to our team and hopefully make the instruction more clear to users, so that one doesn't have to jump between multiple pages. We do have some initiative this year making the documentation better across the board. I'll create an internal JIRA to keep track.
