CQLSH Protocol Error proceeding from apparent version conflicts - java

I am running Cassandra 3.7 on an Ubuntu Server 16.04 LTS 32bit machine that I'm re-purposing from storage. I managed to properly configure the JVM to operate within the 32 bit environment. Nodetool status works fine. But I am having a hard time getting CQLSH to work. I end up with the following error:
('Unable to connect to any servers', {'127.0.0.1': ProtocolError("cql_version '3.4.2' is not supported by remote (w/ native protocol). Supported versions: [u'3.4.0']",)})
Obviously there is some version conflict here but I'm confused as to why and how to fix it. I had a previous 3.0 install of Cassandra that I configured cassanda-env.sh to properly start the JVM. I kept that .sh file when I installed up to 3.7. I updated cassandra.yaml with the new install.
So my question here is: does Cassandra 3.7 support CQL 3.4.2? If not then why does it appear to ship with it? If it does not, what do I need to do to downgrade to CQL 3.4.0? If it does support it and my configuration files are wrong then what's the error and how do I fix it? Or, is there another configuration that I'm completely missing here?
Here are my current configurations:
cassandra-env.sh: https://www.dropbox.com/s/dcs99hgry5behqd/cassandra-env.sh?dl=0
cassandra.yaml:https://www.dropbox.com/s/boi8bh7gin0390f/cassandra.yaml?dl=0
Any assistance greatly appreciated.
EDIT: Figured out, with the help of the first answer, that I had an older instance of the server running. I couldn't get it to start with the correct version...so then I deleted cassandra from /etc/init.d and now I can't get the debian package to automatically register the service.
EDIT2: Downloaded cassandra script from git repo and now cassandra will start again after rebooting. Now cassandra -v shows me:
3.7
Which what I wanted to see and solved my issue pertaining to the version conflict. But now unfortunately I see a new error:
Connection error: ('Unable to connect to any servers', {'127.0.0.1': TypeError('ref() does not take keyword arguments',)})
This problem is addressed in: cqlsh connection error: 'ref() does not take keyword arguments'

It looks like you're running the 3.7 cqlsh against an older server instance. First, be sure you are actually running Cassandra 3.7 (you can probably just check the jar path or look in the logs).
If you want to connect using the advertised cql_version there is a command line option:
https://github.com/apache/cassandra/blob/cassandra-3.7/bin/cqlsh.py#L222
select release_version, cql_version from system.local;
(or just look at the cqlsh welcome header) to see what these parameters are.

Related

Hadoop Installation and Running Taking Too Long in Windows 10

I successfully installed hadoop version 3.2.2 and have Java version 8 (JDK 1.8.0) and followed the tutorial here: https://www.youtube.com/watch?v=g7Qpnmi0Q-s
Here are my executions when running Hadoop in command prompt here:
The issue here is that hadoop (version 3.2.2) is taking way too long to load after successfully installing it in my local drive exactly like in the tutorial I followed above. I don't receive any errors or shutdown terminals after running start-dfs.cmd and start-yarn.cmd.
I can send how I set up my environment variables but I believe that I followed exactly that as described in the tutorial. Here is my setup:
Is there a solution to solve this issue?
I tried other StackOverflow posts but none of them specified how to tackle my issue with not loading Hadoop quickly. It actually never finishes loading even without errors or shutdowns.

Cassandra starting issue : CassandraDaemon

I am getting below error while starting my cassandra setup in my machine windows 10.
>>cassandra.bat -f
WARNING! Powershell script execution unavailable.
Please use 'powershell Set-ExecutionPolicy Unrestricted'
on this user-account to run cassandra with fully featured
functionality on this platform.
Starting with legacy startup options
Starting Cassandra Server
Error: Could not find or load main class org.apache.cassandra.service.CassandraDaemon
Find below more details :
apache-cassandra-3.11.4
java version "1.8.0_45"
Python 3.8.1
I have added CASSANDRA_HOME as well in the environment variable.
I tried all possible solution got it from google/stackoverflow but nothing worked.
Previously i was using 3.11.5 but got some error so i got to know that i should use 3.11.4 so i am using it but even this version not working. Find the link for the same.
Please help

JDBC Connector for MySQL not working in JMeter

As per the problems this person was having;
JDBC Connector not working
I'm running Debian 9.1
I installed Java via the package manager
I added Java to the path in /etc/config & added JAVA_HOME to the path in /etc/environment
I then installed jmeter via the package manager
I then downloaded the mysql-connector-java-5.1.44-bin.jar and copied this to /usr/share/jmeter/lib
But after all this, I still dont get the JDBC connection option under Add->Config Element within JMeter, nor do i get any options for JDBC within JMeter.
Can anyone help me as to what im doing wrong?
While I am new to JMeter, I have successfully run simple "Hello World" HTTP requests queried from a PHP Script & a NodeJS script so far.
Update
Added JMeter log by request
Jmeter log
I have no idea what the problem was with Jmeter on my Debian machine... (as prior to this I was trying to use it on Ubuntu Server, and it had the same issue as Debian);
But anyway, I just installed the Oracle JDK & JMeter for Windows, on my old Windows 7 laptop, dropped in the JDBC connector to the lib directory - & immediately it had the option for 'JDBC Connection', which never appeared on either the Ubuntu or Debian installs.
So if anyone else runs into this issue on Ubuntu or Debian ~ I might suggest trying a Windows 7 VM for running it.
Note here
This was the JMeter I installed on Windows (I didn't get it from Apache):
JMeter for Windows
For some reason I cannot find JMeter JDBC components either in Debian or in Ubuntu repositories, maybe it wasn't included by maintainers.
So instead of installing JMeter using package manager you can just download the latest JMeter version (which is JMeter 3.2 as of now while in repositories you have much older versions, 2.11/2.13) from the official website. JMeter is pure Java-based application so all you will need to do in order to be able to use it is to install Java 8 or above.
Alternative option would be downloading jmeter-jdbc JAR appropriate to your JMeter version and dropping it under /usr/share/jmeter/lib/ext folder (super-user privileges might be required), this way you will get JDBC test elements.

Oracle Sql developer error: could not install some modules

I have downloaded a fresh copy of sqldeveloper (sqldeveloper-4.1.3.20.78-no-jre) from the Oracle website. When I unzip it and lauch the GUI, I keep getting the following error message:
Warning - could not install some modules: Dynamic Module Config - No
module providing the capability org.netbeans.Netbinox could be found.
My system:
Windows 7 32 bit Hp Probook.
Jdk 1.6 , 1.7, 1.8 available in java/jdk
subdirectory.
Netbeans 8.0.2 installed
Finally got the way out after some dirty hours. Initially I extracted the sqldeveloper zip in a nested directory.
i.e: E://A/B/C/sqldeveloper.
starting sqldeveloper gui from the above directory threw erros.
Solution:
Go in C:\users**UserName**\AppData\Roaming : delete SQL Developer and sqldeveloper folders
Extract the Sql Developer zip in a new non nested directory. i.e: E://
Start it and should work now
Reference: https://community.oracle.com/thread/3870680?start=0&tstart=0
This is a very old post but I have had the same issue recently on version 22.2, and this might help.
WARNING: You may lose important settings with this solution.
Try deleting the system cache under C:\Users<your_user_name>\AppData\Roaming\SQL Developer\{whatever}
(or simply everything in that folder). This solved the problem for me, but obviously you do lose configuration settings, connections, and so on, ie, everything you've ever done in SQL Developer - so be careful.
In my case it was enough to change JDK pass to Java 8 in the
C:\Users\YourName\AppData\Roaming\sqldeveloper\19.2.1\product.conf file:
SetJavaHome C:\Program Files\Java\jdk1.8.0_202
I have Java 13 as default and although SQL developer complained at startup it worked OK with it until I started it and went away from the computer, so I missed the warning screen. After that it started to show the "could not install the modules" error.
For those that are working on a linux machine, for me worked just changing the jdk version that I'm currently using. I change went to the path usr/lib/jvm/ and look for the versions that my pc have. Then in ~/.sqldeveloper/"yourversion"/product.conf I changed the Java home to java-8-openjdk (It was using the 11 version)
In my case renaming/deleting only this folder was good enough:
C:\Users*username*\AppData\Roaming\SQL Developer\system22.2.0.173.2018
This way I could still import old connections after starting sqldeveloper again
Based on This I just deleted mypath\sqldeveloper\netbeans\platform\modules\org-netbeans-core-windows.jar and it works fine for me.
I found that I was getting this error after I reinstalled java while keeping an instance of sqldeveloper running.
After I ended the running instance, then I could start a new instance.
This is tricky, so when the could not install some modules board shows up, click the disable modules and continue, then immediately open the app again it should work :)))
The solution in my case was to unzip the SQL developer in a directory without special characters.
At first, I had exported it in a directory that included the hash (#) symbol.
I now have it in a nested directory again, without any special characters, and it works as expected.

RXTX gnu.io.PortInUseException: & WARNING: RXTX Version mismatch

I am trying to get GPS data from a NMEA 0183 GPS via the Java Marine API, using the RXTX API.
Running the Java Marine supplied SerialPortExample.java, I return the two errors. Perhaps they are related and so I am bundling them together into this question.
gnu.io.PortInUseException: Unknown Application
I believe this is caused by the RXTX API. A search reveals that OSX often is missing the required var/lock folder, which one would rectify by with sudo mkdir /var/lock (full details: https://groups.google.com/forum/#!topic/openhab/YUJC8DTLokc) however this does not solve my problem.
WARNING: RXTX Version mismatch
I have Jar RXTX-2.1-7 but native lib Version = RXTX-2.2pre2. I am trying to download the latest version of RXTX, but keep running into dead links (http://rxtx.qbang.org)
I am running Mavericks on MB Air.
Can anyone help?
Problem 1
gnu.io.PortInUseException: Unknown Application
I solved this by looking up the ports in use on my machine with ls /dev/tty.*
This flagged up a lot of old bluetooth devices, I cleared most of them off (Apple > System Preferences > Bluetooth) and it solved the problem
Problem 2
WARNING: RXTX Version mismatch
I took some time to delete everything associated with the RXTX install, and ensured that I was only using one version. Seems to have done the trick.
unknown port exception
This error happens while some of other programs are using the serial port .Make sure your no other applications are using serial port when your application works .check your application is also running only single copy.
RXTX version mismatch
download rxtx jar matched with your java version
I had the same bug. Make sure you terminate any other processes that could be using that connection. I was trying to run my program from the command line but the port was connected to energia's serial monitor.
RXTX: if you manually added any RXTX files delete them and run sudo apt-get install librxtx-java, this should clear up the mismatch.

Categories

Resources