Hello Mark,
I am going through the same problem.
I have server where LS is running in shipper and Indexer mode with redis in between.
Shipper--> input= lumberjack and output =redis
Indexer--> input = redis and output = elasticsearch
LSF is running on client machines which talks to LS shipper.
I want LSF to talk with redis so that I can store some log file specific keys and then create those many indexes on the basis of keys (e.g keyName-YYYY-MM-DD)
But As you said LSF cant talk to redis then I have to bring shipper to client side.
So what is benefit of LSF?
How can I create multiple indexes in ES through one LS indexer running on server? sample configuration please?
In the given example, you are creating index on the basis of project name which you take from path.
I want to create indexes from values of 'Application' field which I set in logstash-forwarder on client.
For example:
in LSF config on CRM client 1 I say
{
"paths": [
"/var/log/crm/crmERROR.log"
],
"fields": {"Application":"CRM","Sub-System":"Mysystem", "type":"Error-logs "
}
in LSF config on sales client 2 I say
{
"paths": [
"/var/log/sales/SalesERROR.log"
],
"fields": {"Application":"Sales","Sub-System":"Myststem", "type":"Error-logs "
}
all those logs will be sent to redis through shipper and indexer will pick it up from redis.
Now in LS indexer I want to create indexes on the basis of this field "Application", How Can I get this value of Application field so that I can use it in output to create indexname.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.