Whenever elasticsearch service dies with the following lines of logs, all the jar files inside lib folder gone missing. After this, the elasticsearch services stop working and not starting.
Why do elasticsearch jars go missing suddenly??
[INFO ][o.e.c.m.MetaDataDeleteIndexService] [EIcXa59] [myindex/1ozT978aRwyiowvNPmpElQ] deleting index
[INFO ][o.e.n.Node ] [EIcXa59] stopping ...
[INFO ][o.e.x.w.WatcherService ] [EIcXa59] stopping watch service, reason [shutdown initiated]
[INFO ][o.e.x.m.p.l.CppLogMessageHandler] [EIcXa59] [controller/53328] [Main.cc@148] Ml controller exiting
[INFO ][o.e.x.m.p.NativeController] [EIcXa59] Native controller process has stopped - no new native processes can be started
[INFO ][o.e.n.Node ] [EIcXa59] stopped
[INFO ][o.e.n.Node ] [EIcXa59] closing ...
[INFO ][o.e.n.Node ] [EIcXa59] closed
We are running elasticsearch x64 service in windows 10.
It is observed from windows event logs that this issue happens randomly during system restarts.
All jars are missing in lib folder and the only empty folder "tools" is present inside lib
I found below lines in windows event viewer during system restart.
Application popup: elasticsearch-service-x64.exe - Application Error : The instruction at 0x72786bc8 referenced memory at 0x00000188. The memory could not be read.
Environment: OS - Windows 10
Elasticsearch - 6.6.2
After system restart, Elasticsearch service stops working [not starting].
And also found that the jar files in lib folder of elasticsearch are missing. No jars present. only empty tools folder is present in elasticsearch directory.
We are not sure why jars went missing after system reboot.
Below lines are from the elasticsearch.log file after system restart initiated:
[INFO ][o.e.c.m.MetaDataDeleteIndexService] [EIcXa59] [myindex/1ozT978aRwyiowvNPmpElQ] deleting index
[INFO ][o.e.n.Node ] [EIcXa59] stopping ...
[INFO ][o.e.x.w.WatcherService ] [EIcXa59] stopping watch service, reason [shutdown initiated]
[INFO ][o.e.x.m.p.l.CppLogMessageHandler] [EIcXa59] [controller/53328] [Main.cc@148] Ml controller exiting
[INFO ][o.e.x.m.p.NativeController] [EIcXa59] Native controller process has stopped - no new native processes can be started
[INFO ][o.e.n.Node ] [EIcXa59] stopped
[INFO ][o.e.n.Node ] [EIcXa59] closing ...
[INFO ][o.e.n.Node ] [EIcXa59] closed
I do not have any other information to share. Still analysing the issue.
Below lines are from the elasticsearch.log file after system restart initiated:
This line and the next ones are expected as you are shutting down the machine, the service has to stop.
[INFO ][o.e.n.Node ] [EIcXa59] stopping ...
This one is not expected:
[INFO ][o.e.c.m.MetaDataDeleteIndexService] [EIcXa59] [myindex/1ozT978aRwyiowvNPmpElQ] deleting index
Someone decided to remove myindex.
We are not sure why jars went missing after system reboot.
I have no idea. May be you have other processes running on your machine like an antivirus and this one found that the lib dir is not healthy to keep and is moving everything to another location?
But please can you answer this simple question:
What is the full path of the lib dir you are speaking about?
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.