1

ElasticSearch 5.6.7 throws somewhat misleading following error and refuse to start:

[2018-08-02T10:07:50,333][ERROR][o.e.b.Bootstrap          ] [esm56n3] node validation exception
[2] bootstrap checks failed
[1]: memory locking requested for elasticsearch process but memory is not locked
[2]: system call filters failed to install; check the logs and fix your configuration or disable system call filters at your own risk

What are possible causes?

Oleksandr Iegorov
  • 41
  • 4

1 Answers1

2

Responding to my own question. Assuming that all set from OS limits point of view, it could also be that your "/tmp" containing partition missing "exec" option. You have several options to correct it and start elasticsearch process here:

  1. fix your /etc/fstab to remove noexec from /tmp containing partition and run mount -oremount
  2. just a run-time fix by enabling exec option: mount -oremount,exec
  3. Add the following option to elasticsearch jvm config (jvm.options) to point to partition with exec capabilities: -Djava.io.tmpdir=/path/to/temp/dir

Then go ahead and restart your elasticsearch systemctl daemon (you do not run it on CentOS < 7, do you? :-) More details here.

Oleksandr Iegorov
  • 41
  • 4