Open coolacid opened 10 years ago
+1.
I'm thinking of collecting example logs from
and creating a git repo with all the ones I find. It would be a good idea to congregate them in a single place.
Do you have an idea on how to make this more actionable and into bite-sized steps? I'm willing to help out muchos since I still find documentation of logstash a bit lacking. More working out of the box examples would be better.
Having a collection of recordings from asciinema could be a good idea too: https://asciinema.org/
Well, first step is to take what's here and re-classify and beautify the readmes. Add another classification of TODO where we have example logs that need either a config, example or tuturials to work with.
Some examples:
https://github.com/coolacid/GettingStartedWithELK/tree/master/Configs/Weblogs
I just need to find time to organize the whole thing. I'm pretty busy until mid Aug but should be more focused then.
I've started screwing with the whole thing.
Something that needs to be noted - some of the "configs" should be made into tutorials or examples, the configs should literally be JUST the filter commands wrapped in a type conditional.
Not sure I follow, do you have some before-and-after examples to show what you are describing?
By the way, you don't even have to copy the files into /tmp
unless you specifically want to use the File
input. You can just use exec { command => "cat weblogs.txt" interval => 60 }
as in here
@coolacid so... I don't think there's enough of a distinction between the words for Configs, Examples, Snippets, and Tutorials. In fact, all those words mean so very similar things. I think we should be more explicit. I think Examples
and Snippets
are similar to each other. It's just that Snippets cover more advanced examples.
Some other things I think should be added:
We should have a default logstash.conf. Maybe something like this is a good base:
input {
generator {
message => '127.0.0.1 - - [08/Aug/2009:00:01:01 -0000] "GET /elasticsearch/logstash/tree/82fdc11f2e35b52b05f773be52ba3c0e8072c509 HTTP/1.1" 200 1420 "https://github.com/elasticsearch/logstash/tree/82fdc11f2e35b52b05f773be52ba3c0e8072c509" "curl/7.37.1"'
type => "generated"
count => 1
}
}
filter {
grok { match => { "message" => "%{COMBINEDAPACHELOG}" } }
date { match => [ "timestamp" , "dd/MMM/yyyy:HH:mm:ss Z" ] }
}
output {
stdout { codec => rubydebug }
}
And then from there, have each example change as few things from the base logstash.conf as possible. This is to ensure that users will always have a usable example and won't require a leap of faith on understanding semantics on a particular thing on Logstash.
Makefile
probably, we should provide an easy way to download multiple versions of Logstash to test the conf files with, at least the ones listed here: https://gist.github.com/shurane/92b122ebd0a1ed665397. This is in the interest of easy access.Also, Tut**u**rial
is a typo, it should be Tut**o**rial
.
I'm totally willing to do a lot of this work, just need to make some time to work through these.
So, im on vacation. Let me revisit when I get back.
Need to reorg the whole thing.