Forum

Information and discussion related to the Kognitio on Hadoop product
Contributor
Offline
User avatar
Posts: 48
Joined: Tue May 28, 2013 1:44 pm

Moving from Kognitio standalone to Kognitio on Hadoop

by anonymous2 » Fri Dec 02, 2016 1:52 pm

I'm used to working with the Kognitio standalone product (i.e. a dedicated system running Kognitio software, rather than running Kognitio as an application on Hadoop).

Now we are looking to use Kognitio on Hadoop, but I want to still be able to login and use the command-line tools I am used to, like wxprobe.

Is there a way to make kodoop look more like Kognitio standalone for me?
Reply with quote Top
Contributor
Offline
User avatar
Posts: 384
Joined: Thu May 23, 2013 4:48 pm

Re: Moving from Kognitio standalone to Kognitio on Hadoop

by markc » Fri Dec 02, 2016 1:57 pm

Assuming you have a user account on an edge node that you use for starting the Kognitio on Hadoop instance (lets call that user "kodoopuser" in this example, and say the Kognitio on Hadoop instance is called kodclust).

Now you can login to the edge node as kodoopuser, and run "kodoop msg kodclust shell" - this gives a bash prompt with access to all of the usual commands, like wxprobe. You can also use `wxlogd smd` to find the latest smd directory, etc.

Things like the log files are also stored in HDFS - if you look in the kodoopuser's HDFS home directory, they will be under .kodoop-clusters/kodclust/logs.

With Kognitio standalone you only ever have one Kognitio instance running, but with Kognitio on Hadoop (I'll call that kodoop from now on, for ease of typing), you can have multiple instances of the kodoop application on one Hadoop cluster. They can all be running concurrently too. You can look under user's home directories in HDFS (as described above) to find which clusters are being run by which users.
Reply with quote Top

Who is online

Users browsing this forum: No registered users and 1 guest

cron