We're pushing logs into an s3 bucket. We've setup the our indexer configuration and tested with a single folder within our bucket - which naming convention is the date (e.g. my-bucket-of-logs/20200804) however we want to be able to pull in logs from the last 14 days (14 folders)
You could use 14 inputs in 14 configuration files. Once a day replace the oldest configuration file with one for today. If logstash is running with -r it will reload the pipeline when it sees one of the configuration files change. It's a hack, but it might do what you want to get done.
Thanks for the reply @Badger. Unfortunately thats not a viable solution. I think best solution in this case would be to create another s3 bucket, ingest all logs there and copy over all log files into the existing buckets daily directories.
push cloudflare logs to "new-logs-bucket"
use s3-indexer.conf to point to that bucket
copy over log files into my-bucket-of-logs/{{ current_date }}
create lifecycle policy to remove logs after N amount of days from "new-logs-bucket"
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.