There is a closed topic regarding the same from 2 years ago ...
But the .rb code has been updated on GitHub.
What is the latest solution to using the gzip-lines codec?
I have installed the codec and am running the following conf ...
input {
file {
type => "gzip"
path => "/home/aptusdb/data/list.txt" #path to list.txt containing path/filename of all .gz files
start_position => "beginning"
sincedb_path => "gzip"
codec => "gzip_lines" #gzip_lines codec
}
}
output {
elasticsearch {
hosts => "localhost"
index => "trial1"
document_type => "sold_cars2"
}
stdout {}
}
But I am getting the following error ...
[2018-10-05T15:33:45,695][ERROR][filewatch.tailmode.handlers.grow] read_to_eof: general error reading /home/aptusdb/data/list.txt {"error"=>"java.lang.IllegalArgumentException: Object: data/sample/day=2018-09-23/hour=23/sample_data.txt.gz is not a legal argument to this wrapper, cause it doesn't respond to \"read\".", "backtrace"=>["org.jruby.util.IOInputStream.<init>(IOInputStream.java:64)", "org.jruby.ext.zlib.JZlibRubyGzipReader.initialize(JZlibRubyGzipReader.java:115)", "org.jruby.ext.zlib.JZlibRubyGzipReader.initialize19(JZlibRubyGzipReader.java:145)", "org.jruby.ext.zlib.JZlibRubyGzipReader$INVOKER$i$0$1$initialize19.call(JZlibRubyGzipReader$INVOKER$i$0$1$initialize19.gen)"]}