A plugin for a newly developed Java/Spring Metrics application I didn’t really want to code this project but I couldn’t find a respectable alternative that would also run on any Windows machine - which is the problem and why I am not going with Graphite and statsd. This application provides multiple integration options so as to make its use under your network requirements possible. This includes a REST option that is always enabled for your use in case you want to write a small script to send the occasional metric data.
Find GraphTastic here : https://github.com/NickPadilla/GraphTastic
output {
graphtastic {
batch_number => ... # number (optional), default: 60
codec => ... # codec (optional), default: "plain"
context => ... # string (optional), default: "graphtastic"
error_file => ... # string (optional), default: ""
host => ... # string (optional), default: "127.0.0.1"
integration => ... # string, one of ["udp", "tcp", "rmi", "rest"] (optional), default: "udp"
metrics => ... # hash (optional), default: {}
port => ... # number (optional)
retries => ... # number (optional), default: 1
workers => ... # number (optional), default: 1
}
}
the number of metrics to send to GraphTastic at one time. 60 seems to be the perfect amount for UDP, with default packet size.
The codec used for output data. Output codecs are a convenient method for encoding your data before it leaves the output, without needing a separate filter in your Logstash pipeline.
if using rest as your end point you need to also provide the application url it defaults to localhost/graphtastic. You can customize the application url by changing the name of the .war file. There are other ways to change the application context, but they vary depending on the Application Server in use. Please consult your application server documentation for more on application contexts.
setting allows you to specify where we save errored transactions this makes the most sense at this point - will need to decide on how we reintegrate these error metrics NOT IMPLEMENTED!
Only handle events without any of these tags. Note this check is additional to type and tags.
host for the graphtastic server - defaults to 127.0.0.1
options are udp(fastest - default) - rmi(faster) - rest(fast) - tcp(don’t use TCP yet - some problems - errors out on linux)
metrics hash - you will provide a name for your metric and the metric data as key value pairs. so for example:
metrics => { “Response” => “%{response}” }
example for the logstash config
metrics => [ “Response”, “%{response}” ]
NOTE: you can also use the dynamic fields for the key value as well as the actual value
port for the graphtastic instance - defaults to 1199 for RMI, 1299 for TCP, 1399 for UDP, and 8080 for REST
number of attempted retry after send error - currently only way to integrate errored transactions - should try and save to a file or later consumption either by graphtastic utility or by this program after connectivity is ensured to be established.
Only handle events with all of these tags. Note that if you specify a type, the event must also match that type. Optional.
The type to act on. If a type is given, then this output will only act on messages with the same type. See any input plugin’s “type” attribute for more. Optional.
The number of workers to use for this output. Note that this setting may not be useful for all outputs.