Last night, I installed the marvel plugin, after which my automation script
(1) issued 'service elasticsearch restart' within seconds of each other.
After which, in Paramedic, both nodes showed 'no data' (using Paramedic)
which was cause for a few minutes of 'whoa'.
Restarted node-1, waited for it to init, then restarted node-2 and all was
well.
Question is: is restarting ES in a cluster mode using the built-in service
facility the 'right' way? Or would it be better to halt / start using the
API? [2]
Last night, I installed the marvel plugin, after which my automation
script (1) issued 'service elasticsearch restart' within seconds of each
other.
After which, in Paramedic, both nodes showed 'no data' (using Paramedic)
which was cause for a few minutes of 'whoa'.
Restarted node-1, waited for it to init, then restarted node-2 and all was
well.
Question is: is restarting ES in a cluster mode using the built-in service
facility the 'right' way? Or would it be better to halt / start using the
API? [2]
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.