Thank you very mutch! I would never have found this solution since im very new in the elastic ecosystem.
Querying the type from elasticdb works like a charm for now. I hope this wont slow the system too much down when we are in a stage with many millions of data.
My final config looks like this:
input {
couchdb_changes {
db => "mydbname"
host => "127.0.0.1"
port => 5984
initial_sequence => 0 #this is only required for the an initial indexing
}
}
filter {
mutate {
add_field => { "action" => "%{[@metadata][action]}" }
}
if [action] == 'delete' {
elasticsearch {
hosts => ["127.0.0.1"]
query => "_id:%{[@metadata][_id]}"
fields => ["type", "type"]
sort => ""
}
} else {
mutate {
add_field => { "type" => "%{[doc][type]}" }
}
}
}
output {
elasticsearch {
action => "%{[@metadata][action]}"
document_id => "%{[@metadata][_id]}"
host => "127.0.0.1"
index => "myindexname"
protocol => "http"
port => 9200
}
#stdout { codec => rubydebug } #enable this option for debugging purpose
}
Is there something suboptimal in the config? Something that could slow us down unnecessary?