Installing marvel
Attempting to transfer from file:/home1/isuser/Desktop/up/marvel-2.3.3.tar.gz
Transferring 1611136 bytes....................
Transfer complete
Extracting plugin archive
Extraction complete
Plugin installation was unsuccessful due to error "ENOTDIR: not a directory, unlink '/home1/isuser/Desktop/kibana-4.5.1-linux-x64/installedPlugins/.plugin.installing/archive.part'"
I tried changing the number of slashes after file as recommended by a user here but that doesn't seem to fix the issue. Does anyone have some advice?
So Kibana doesn't care about that file. Elasticsearch requires that file as part of its plugins, but Kibana uses a system more related to node.js (package.json).
i'm pretty new to elasticsearch obviously so maybe I'm missing something but after I start my node and I go to localhost:9200/_plugin/marvel shouldn't I see the UI (unless something didn't install correctly)? all I see is a blank screen
but after I start my node and I go to localhost:9200/_plugin/marvel shouldn't I see the UI (unless something didn't install correctly)? all I see is a blank screen
No worries. That was the old way in ES 1.x (and Marvel 1.x). You are expecting to see the Marvel "site plugin", which is how Marvel worked in Marvel 1.x. However, running a UI from a data store proved to be a workable, but-not-so-great idea, so we stopped doing it.
Now that the UI is a part of Kibana, you need to go to Kibana to visually use it. Assuming you have started Kibana (${KIBANA_DIR}/bin/kibana), then it starts up a node.js server listening for web requests at port 5601 by default.
Since you're running Kibana 4.5, the plugins appear under the 9 square-dots:
Click on those.
Select Marvel.
Voila, you should now be able to monitor your cluster.
That video is definitely out of date, unfortunately. As you've noticed, Elasticsearch 1.6 and 2.x are kind of different beasts. We really started to lock things down and tie things together in 2.x. Fortunately, Elasticsearch 5.x is much, much closer to ES 2.x than ES 2.x was to ES 1.6.
As for better resources, you should probably take a look at these "Elastic{content}" groups of content:
We're actually in the process of updating the Getting Started webinar to be more relevant to ES 2.x, so be on the look out for a post on that in the future.
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.