Creating a New User in Cassandra Database

Introduction...

Hi All,

Recently I was working on user authentication of DSE(DataStax Enterprise[1]) Cassandra and I noticed that any username and password allowed me to access the cassandra cluster. So I looked into the docs and saw that there are several authenticators[2].

By default the used authenticator is the "org.apache.cassandra.auth.AllowAllAuthorizer" authenticator which allowed access with any username and password. So to create a user and grant permissions, you can follow these steps...

Steps...

Shutdown the Cassandra cluster(can use kill -9) if it is already running and the modify the cassandra.yaml file. If you are using DSE cassandra, it will reside in "<DSE_HOME>/resources/cassandra/conf/cassandra.yaml". Else if you are using Apache Cassandra, it will reside in "<APACHE_CASSANDRA_HOME>/conf/cassandra.yaml".

Modify the file in a way that "org.apache.cassandra.auth.AllowAllAuthorizer" authenticator is commented and add "org.apache.cassandra.auth.PasswordAuthenticator" authenticator[3].

#authenticator: AllowAllAuthenticator
authenticator: PasswordAuthenticator

By default, Cassandra has a user with username as "cassandra" and password as "cassandra" which can be used to create a new user using "cqlsh" tool[4]. Login using the "cqlsh" tool by executing the following command.

./cqlsh localhost -u cassandra -p cassandra

After the cqlsh console is opened, to get the current list of users run the following command.

list users;

To create a new user, run the following command. Here the username is "myUserName" and the password is "myPassword".

CREATE USER myUserName WITH PASSWORD 'myPassword' SUPERUSER;

The above user will get created with superuser privileges. The newly created user should be there when running the "list users;" command again.

Goodluck!!!

References...

[1] - http://www.datastax.com/products/products-index
[2] - http://docs.datastax.com/en/cassandra/1.2/cassandra/security/secure_config_native_authorize_t.html
[3] - http://www.datastax.com/dev/blog/a-quick-tour-of-internal-authentication-and-authorization-security-in-datastax-enterprise-and-apache-cassandra
[4] - http://docs.datastax.com/en/cassandra/1.2/cassandra/security/security_config_native_authenticate_t.html

View Hazelcast Management Center for WSO2 Message Broker 3.0.0.

WSO2 products allows to use multiple instances to form a cluster and divide up the work to enhance the performance and reliability of their intended processes[1]. WSO2 Message Broker 3.0.0 uses Hazelcast[2] to communicate within the instances of a cluster and is heavily depended on on it when it comes message delivering scenarios and etc. The following post is on how to access the hazelcast management center[3] of the Message Broker(s). The hazelcast management center helps to monitor the overall state of the cluster.

Lets see how we can make it happen. The following prerequisites are needed :

• A WSO2 Message Broker cluster(clustering enabled).
• Tomcat to host hazelcast management center.

First we have to modify the axis2.xml file of all the instances. Access the "axis2.xml" file in the following path.

<WSO2_PRODUCT>/repository/conf/axis2/axis2.xml

At the bottom of the file in the "clustering" section, the following commented line can be seen.

<!--<parameter name="mgtCenterURL">http://localhost:8081/mancenter/</parameter>-->

Uncomment the line and modify the port to 8080. This is the default port in tomcat which we will get back later.

<parameter name="mgtCenterURL">http://localhost:8080/mancenter/</parameter>

Startup the Message Broker instance(s) afterwards.

Now lets host the hazelcast management center.

To do this we need to get the "mancenter-3.2.6.war" file from the hazelcast official site. The version of the war file should be the exact hazelcast version that is being used in WSO2 Message Broker. I.E current version is 3.2.6

Download Hazelcast - The war file is included.

After downloading, place the "mancenter-3.2.6.war" file in the "<TOMCAT_HOME>/webapps" folder and rename the war file to "mancenter.war". Start up the tomcat server using "./startup.sh" in the "bin" folder of tomcat.

Then go to the following URL :
http://localhost:8080/mancenter/

Login to the hazelcast management center using username and password as "admin" and "admin". Once logged in, you should see the hazelcast management center.

Following are several images of the hazelcast management center.

This mentioned solution should also work with other WSO2 products which uses hazelcast for clustering. But I have not tested them out. I do not guarantee this solution will work or held responsible.

[1] - https://docs.wso2.com/display/CLUSTER420/Overview
[2] - http://hazelcast.org/
[3] - http://docs.hazelcast.org/docs/latest/manual/html/managementcenter.html

Thanks,
Hemika

Unable to mount ExFat external HDD (' stderr: ERROR:._.com.apple.timemachine.donotpresent' has invalid checksum)

Hi All,

I have an external hard drive(Western Digital My Passport Ultra 1TB) which I keep all my movies, work stuff and etc. Initially I formatted the drive in ExFat format in Mac OS X platform so that I can use it in Windows and Mac. Also Ubuntu seems to have a fix to read ExFat file systems[1].

But recently when I connected this external hard drive to my Ubuntu machine, it gave me this error(something similar) saying that it cannot mount the disk: 

Error mounting /dev/sdb1 at /media/rbuse/My Passport: Command-linemount -t "exfat" -o "uhelper=udisks2,nodev,nosuid,uid=1000,gid=1000,io charset=utf8,namecase=0,e rrors=remount-ro,umask=0077" "/dev/sdb1" "/media/rbuse/My Passport"' exited with non-zero exit status 1: stdout:FUSE exfat 1.0.1

stderr: ERROR:._.com.apple.timemachine.donotpresent' has invalid checksum (0x6634 != 0x4044). '

I was stunned because all my work was in there, especially my VDIs. 

So I started my research and one solution was to plug it back in to my Mac OS X machine and use the "Disk Utility" tool to repair it(Use "Repair Disk" button) [2][3]. When I tried this, the following error came up.

Error: Disk Utility can’t repair this disk. Back up as many of your files as possible, reformat the disk, and restore your backed-up files.

I was like "Huh?!?!?" . I continued my research and found that if I attempt several times to "Repair Disk", it will work at sometime[4]. I tried like 20+ times, it didn't work.

So I googled more and found a command that should fix the broken external hard drive [5][6][7].

sudo fsck_exfat -d <device>

To find the correct value for <device>(eg : disk1s2) use the following command[8] to list down the device IDs.

diskutil list

At the end of the fsck_exfat command, I got the following error[9]. 

fsck_exfat: Could not update main boot region: Bad file descriptor

Sigh!.

As the last attempt to fix, I used the "chkdsk" command of Windows OS[10]. And IT WORKED!!. I ran the following command. It took sometime to complete the task. So be patient.

chkdsk /f <Volume_ID>

Here the Volume_ID refers "C:" or "D:" of the corrupted hard disk. eg : 

chkdsk /f D:

Hope this article helped!,

Thanks,
Hemika

[1] - http://askubuntu.com/questions/451364/how-to-enable-exfat-in-ubuntu-14-04
[2] - https://bbs.archlinux.org/viewtopic.php?pid=1481063#p1481063
[3] - http://osxdaily.com/2014/01/27/verify-disk-command-line-mac-os-x/
[4] - http://beshoy.girgis.us/2013/11/solved-error-disk-utility-cant-repair-disk-backup-many-files-possible/
[5] - https://discussions.apple.com/message/24958301#24958301
[6] - http://craigsmith.id.au/2014/07/06/repairing-a-corrupted-mac-osx-exfat-partition/
[7] - https://developer.apple.com/library/mac/documentation/Darwin/Reference/ManPages/man8/fsck_exfat.8.html
[8] - https://discussions.apple.com/thread/4154638?start=30&tstart=0
[9] - https://discussions.apple.com/thread/4154638?start=45&tstart=0
[10] - https://www.microsoft.com/resources/documentation/windows/xp/all/proddocs/en-us/chkdsk.mspx?mfr=true

Changing Version Control in IntelliJ

IntelliJ supports different types of version controlling methods. I recently had an issue of changing the version control for a project in IntelliJ. IntelliJ was showing an SVN version control, but it was not picking up the GIT version control even if all the git files were there. The solution was to edit the ".idea/vcs.xml" file.

The content was as follows.

<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
    <component name="VcsDirectoryMappings">
        <mapping directory="$PROJECT_DIR$" vcs="svn" />
    </component>
</project>

I modified it to the following.

<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
    <component name="VcsDirectoryMappings">
        <mapping directory="$PROJECT_DIR$" vcs="Git" />
    </component>
</project>

Then I reopened the project in IntelliJ and was able to see the GIT sub menu in the VCS menu.

The content in the vcs.xml file might be different depending on the project.

Recently one of my colleagues was having the same issue. In his vcs.xml file, There were several mappings all directed to "svn". We changed to "Git" and reopened the project. An error notification started appearing at the top right corner mentioning a version control error and there was a link as "configuration". Once clicking the link it showed the "Version Control" of the project. We removed the broken mappings and added a new mapping with GIT. We checked the vcs.xml file afterwards, it was as following.

<?xml version="1.0" encoding="UTF-8"?>
<project version="4">
    <component name="VcsDirectoryMappings">
        <mapping directory="" vcs="Git" />
    </component>
</project>

The Git menu also came up in the VCS menu in IntelliJ.

Please note that I haven't tested with other version control methods in IntelliJ. This is just sharing hick-ups I had during coding. :)

Open Sublime in Terminal for Mac OSX

Even though we install Sublime 2 or 3 using a DMG file in Mac OSX, we cannot open Sublime using the terminal. To fix this, run the following...

For Sublime Text 2

cd /usr/local/bin
ln -s "/Applications/Sublime Text 2.app/Contents/SharedSupport/bin/subl" subl

For Sublime Text 3

cd /usr/local/bin
ln -s "/Applications/Sublime Text.app/Contents/SharedSupport/bin/subl" subl

To open Sublime in terminal, use the following command.

subl 

To see usages of subl command

subl --help

References

• https://www.sublimetext.com/docs/3/osx_command_line.html
• http://stackoverflow.com/questions/11889484/command-subl-from-terminal-dont-work
• https://bensmann.no/open-folder-in-sublime/

Creating Runnable JAR using Maven.

At times we need to create a runnable jar using maven. So that we can run it as following.

java -jar MyJar.jar

This can be achieved using 2 different ways as far as I have learnt.

1. Using "maven-shade-plugin" Plugin of Maven
2. Using "maven-jar-plugin" Plugin of Maven

Using "maven-shade-plugin" Plugin of Maven

This might be the easiest way of creating a jar file of your java program. Simply add the following plugin to pom.xml.

<project>
  ...
  <build>
    <plugins>
      ...
      <plugin>
        <groupId>org.apache.maven.plugins</groupId>
        <artifactId>maven-shade-plugin</artifactId>
        <version>2.3</version>
        <executions>
          <execution>
            <phase>package</phase>
            <goals>
              <goal>shade</goal>
            </goals>
            <configuration>
              <transformers>
                <transformer implementation="org.apache.maven.plugins.shade.resource.ManifestResourceTransformer">
                  <mainClass>com.hemika.samples.MyMainClass</mainClass>
                </transformer>
              </transformers>
            </configuration>
          </execution>
        </executions>
      </plugin>
      ...
    </plugins>
  </build>
  ...
</project>

Here we have to add the reference path of the class which contains the main method to the "mainClass" element in the xml.

Sometimes we might encounter the following exception.

Exception in thread "main" java.lang.SecurityException: Invalid signature file digest for Manifest main attributes

The reason for this is as follows...

" The reason java.lang.SecurityException is raised is because some dependency jar files are signed jar files. A jar file is signed by using jarsigner, which creates 2 additional files and places them in META-INF: a signature file, with a .SF extension, and a signature block file, with a .DSA, .RSA, or .EC extension. Since the uber-jar file is created, the signatures and integrity of signed JAR files are no longer valid. When the uber-jar file is executed, java.lang.SecurityException is thrown. " - maven shade plugin: Invalid signature file digest for Manifest main attributes

The solution is it add the following filter.

<configuration>
 <filters>
            <filter>
              <artifact>*:*</artifact>
              <excludes>
                <exclude>META-INF/*.SF</exclude>
                <exclude>META-INF/*.DSA</exclude>
                <exclude>META-INF/*.RSA</exclude>
              </excludes>
            </filter>
 </filters>
</configuration>

To build the jar, run the following command...

mvn clean install

...and thats all for making a shaded jar.

Click here to see a sample.

Using "maven-jar-plugin" Plugin of Maven

Using this way requires to use 2 maven plugins. After executing "mvn clean install", the target folder will contain the jar file and also a "lib" folder. The "lib" folder will contain all the dependencies. This way is more advantageous as the depending library files are available from outside of the jar. Later if we need to replace the dependency jar with a new version or so, this will become useful.

Add the following to the pom.xml.

<project>
  ...
  <build>
    <plugins>
      ...
      <plugin>
 <artifactId>maven-dependency-plugin</artifactId>
 <executions>
   <execution>
     <phase>install</phase>
     <goals>
              <goal>copy-dependencies</goal>
            </goals>
            <configuration>
       <outputDirectory>${project.build.directory}/lib</outputDirectory>
     </configuration>
   </execution>
 </executions>
      </plugin>
      <plugin>
 <groupId>org.apache.maven.plugins</groupId>
 <artifactId>maven-jar-plugin</artifactId>
 <configuration>
   <archive>
            <manifest>
       <addClasspath>true</addClasspath>
       <useUniqueVersions>false</useUniqueVersions>
       <classpathPrefix>lib/</classpathPrefix>
       <mainClass>com.hemika.samples.Main</mainClass>
            </manifest>
          </archive>
 </configuration>
      </plugin>
      ...
    </plugins>
  </build>
  ...
</project>

The class that has the main method needs to be added at the "mainClass" node in the xml.

To build the jar, run the following command...

mvn clean install

...and thats all for creating a runnable jar.

Click here to see a sample.

References...

• https://maven.apache.org/plugins/maven-shade-plugin/examples/executable-jar.html
• http://zhentao-li.blogspot.com/2012/06/maven-shade-plugin-invalid-signature.html
• https://maven.apache.org/plugins/maven-dependency-plugin/examples/copying-project-dependencies.htmlhttps://maven.apache.org/plugins/maven-dependency-plugin/examples/copying-project-dependencies.html
• http://www.mkyong.com/mongodb/java-mongodb-hello-world-example/

Use a Config File for Easy SSH Connections.

Problem ?

When working with Amazon EC2 instances its pretty annoying to connect to the instances using SSH. All the time we have to give the path to the pem file, the username of the EC2 instances, the public DNS for the instance. This is pretty much annoying.

$ ssh -i Key.pem ubuntu@ec2-11-11-11-11.xx-xxxx-xx.compute.amazonaws.com

But there is an easier way that you can store all those variables in a config file an use it to connect the instances from any folder.

Solution

Go to the root directory using "CD".

$ cd

Go the to ".ssh" folder. This folder might already be there.

$ cd .ssh/

Create a file named "config" using an editor. I'll be using VIM.

$ vim config

The config file should contain all our variables for the EC2 instances or where ever you are connecting to. Lets say we have the following EC2 instances.

ID	Username	Public DNS	PEM File Path
instance1	ubuntu	ec2-11-11-11-11.xx-xxxx-xx.compute.amazonaws.com	~/.ssh/pems/Key.pem
instance2	ubuntu	ec2-22-22-22-22.xx-xxxx-xx.compute.amazonaws.com	~/.ssh/pems/Key.pem
instance3	ubuntu	ec2-33-33-33-33.xx-xxxx-xx.compute.amazonaws.com	~/.ssh/pems/Key.pem

In the above table, the ID is a unique value for each instance. You can use any value for the ID. Username is the login username for the instance. Public DNS is the host name for the instance. The PEM File Path is the file path for the identity file.

Using those data we can create the config file content as follows.

Host instance1
    User ubuntu
    Hostname ec2-11-11-11-11.xx-xxxx-xx.compute.amazonaws.com
    IdentityFile ~/.ssh/pems/Key.pem

Host instance2
    User ubuntu
    Hostname ec2-22-22-22-22.xx-xxxx-xx.compute.amazonaws.com
    IdentityFile ~/.ssh/pems/Key.pem

Host instance3
    User ubuntu
    Hostname ec2-33-33-33-33.xx-xxxx-xx.compute.amazonaws.com
    IdentityFile ~/.ssh/pems/Key.pem

After saving the "config" file, you can connect to any of those instances from any folder. If you want to connect to "instance2", it will be as follows.

$ ssh instance2

Happy SSH-ing!