I'm doing a pressure test for our beats like filebeat use libbeat package,But I found github.com/elastic/beats/libbeat/reader/readfile.(*TimeoutReader).Next.func1 is continuous growing。
The presure test condition is described as follows:
i generate 100 file concurrently
and generate log line without sleep for about five seconds
It seems that the memory is constantly going up, which could be a bad indicator indeed.
From what I understand from your test there are logs coming only for the first 5 seconds and then nothing? Or you are constantly adding batches of 100 files with a specific rate?
I suggest you opening a Github issue to report this observation since it sounds valid enough. Could you please open a Github issue in order to keep track of this please? Provide all the information you posted here as well as possible steps to reproduce.
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.