I am using the last_run_metadata_path do define where the file it uses for persistence is kept. The documentation says that the default value for this is "/home/ph/.logstash_jdbc_last_run", but there is no such file or directory on my server. I believe this is causing it to not work. I read in a forum that others were having the same issue, and that trying to set it to a file in $HOME also has problems because of read/write user permissions.
As suggested in the forum, I set my path to /tmp/.jdbc_metrics-buffalonews_last_run and it is now working for me, but I am concerned because, from what I understand, the tmp directory is only temporary. Wouldn't that mean whenever tmp gets cleared that that file will be removed and that Logstash won't remember where it left off?
Ah, I might have been misdiagnosing the issue. I have one very long running import using JDBC paging so that it splits it up into queries of 100000 limits. When does Logstash generate the jdbc_last_run file? When the entire pipeline ends? Or when the last output happened?
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.