Restart doesn't work the first time if a stale PID exists

[andresp99999]

I am seeing an error when trying to restart elasticsearch if a PID exists. This can be reproduced with the following steps

$ sudo service elasticsearch stop
Stopping elasticsearch....................

$ ls /usr/local/var/run
search_master_1_localdomain.pid  <--- PID exists

$ sudo service elasticsearch restart
Stopping elasticsearch...PID file found, but no matching process running?
Removing PID file...   <--- PID is deleted but restart process doesn't continue.

$sudo service elasticsearch restart
No PID file found -- elasticsearch not running?
Starting Elasticsearch... [OK]   <--- The second time ES is restarted it works fine.

Ideally restart should work the first time even if a PID is present.

[martinb3]

Hello! We've re-written the cookbook using libraries that expose resources and providers, and we are planning to focus on fixing the init script as part of #321. I'm going to close this in favor of #321. Hope this helps!

[martinb3]

Hi folks -- could you try the 2.0.0_wip branch of this cookbook, and report back any issues in #321? It has the same init scripts now as the packaged version of ES; it's our goal to release a 2.0.0 version that works with ES 2.0.0. It does have some changes from the current version, but depending on how complex your resource usage is, you may or may not be affected. Thanks!