java - Successful Oryx Install on Google Compute Engine? -

I'm trying to get Oryx up and running on Google Compute Engine. I created a new installation and installed Oryx to:

  git clone https://github.com/cloudera/oryx.git cd oryx mvn -DiskipTests install  

and saved this installation as an image on Google Compute Engine ("oryx-image").

Finding Issues with the Orics and Google File Systems () I am using hdfs: // as the default file system.

Finding issues with the default Hadoop package on Google Compute Engine (e.g., any lenient libraries required for the default Oryx configuration), I have also tried Showcount 2.4 After making the 1Turbal with affectionate, these instructions have been followed: (Side note: is the JDK version described here is sufficient for the oracles?). I then used my saved image with Orex ("orix-image"):

  ./bdutil --bucket & lt; A few-bucket & gt; --image oryx-image -n $ number \ --env_var_files hadoop2_env.sh --default_fs hdfs  

and my saved hoodop turbul:

  # File: To deploy Hadop 2.4.1: Hadiopp 2_Anv.sh HDOOP_iv.SH = "GS: // and lieutenant; some-bucket" / hadoop-2.4.1.tar.gz "  Cluster on Google Compute Engine (with default file system = hdfs: //) is still no luck. 
 
 I can successfully run test handop jobs on GCE, GCE ( See) on snap implementation I can test and test Oryx jobs from GCE  locally  from master node: / P> 
  # file: oryx.conf model.Local-data = The correct model.Local-computation = true  
 
 The only problem is to run the orix successfully, either with the data in hdfs: // or gs: // Google Compute Engine. 
 < P> I have got many different instructions for environmental variables, etc., and I do not know which are necessary, and for more problems A pioneer could be I was wondering if there is a documentation for installation / running on GCE. Perhaps someone has already gone through the process and can give instructions and / or at least confirm a successful establishment? 
 
 Instructions for establishing with HoPOP 2.4.1 GCi with Sniper (found in) were fantastic. 
 
 Thank you! 
 

I do not know That's a straightforward answer, but I can comment here at some points. I think there are so many issues getting the standard handop installation and running on GCE.

I have never run it on GCE, but it should not directly talk about whether it runs on bare metal or GCE or EC 2 it uses only Hadoop Yes, although it considers Hadoop , And HDFS (I think hard-coding HDFS: // can be removed, it is certain; I do not know that it will work with non-HDFS file systems.) So if one is by default in GCE There is a separate file system, so your best The student condition is to use HDFS.

I think I think about sniper as an essential part of the hopop installation. If you are installing Handup with a hand, then I think that you have to take some more steps. This is the reason why I recommend a free (open, open source) destro which takes care of it for you.

It should also set things like HADOOP_CONF_DIR , HM, I also think at least on the client side as an essential part of a Hadoop Setup.

Any version of Java 6 or above is OK.

Is it possible to try a distro? It can be very painful, I am sorry that I do not have any more instructions here, but it seems that GCE and lieutenant; -> Hadop Hint is more that Hadop End Lieutenant; -> Orex If the app can turn it into ways to adjust GCE, then I can do this.