Upgrading to Datastax Cassandra 5.0 ( Part 2 )

-
                                                                       


Follow the instructions to upgrade  DataStax Enterprise  from 4.8 to 5.0 version. First take the backup of cassandra.yaml and dse.yaml. Don't forget to signup in Datastax acadamey , which we will need during the installation process 




Step 1:  Snapshot all keyspaces as needed.
          
             nodetool snapshot <keyspace_name>

Step 2:  Backup keyspace snapshot

Step 3:  Run nodetool drain to flush the commit log of the old installation.
            
             sudo nodetool -h hostname drain

Step 4:  Stop node services
         
            sudo service dse stop
            sudo service datastax-agent stop


Step 5:  Backup configuration file.
   
            cp -r /etc/dse/dse.yaml /home
            cp -r /etc/dse/cassandra/cassandra.yaml /home
            cp -r /etc/dse/cassandra/cassandra-topology.properties /home
            cp -r /etc/dse/cassandra/cassandra-env.sh /home



Step 6:  Make sure you have latest java is installed (1.8.0_40 minimum) or install it

             sudo java -version

Step 7:  Add Datastax repository file
   
      sudo echo "deb https://login_email_Address:login_Password@debian.datastax.com/enterprise stable main" | sudo tee -a /etc/apt/sources.list.d/datastax.sources.list


Step 8:  Add the Datastax repository key
      
            sudo curl -L https://debian.datastax.com/debian/repo_key | sudo apt-key add -

Step 6: Install the package

            sudo apt-get update
            sudo apt-get install dse-full

Step 7:  Configure the new version using backup configuration file we already taken.

Step 8:  Start Node service

             sudo service dse start
             sudo service datastax-agent start

              and make sure that everything is runnig fine by checking the log.

             tail -f -n 1000 /var/log/cassandra/system.log

Step 9:  Upgrade the SSTables
        
             sudo nodetool upgradesstables

Step 10:  Check Node status
   
               sudo nodetool status

Step 11:  Repeat the upgrade on each node in the cluster

Step 12:  If your upgrade includes DSE Search nodes:

                After all nodes are updated to 5.0, comment out all shard_transport_options instances in each dse.yaml file.
                 Restart the nodes to fully shut down the old shard transport. 

Comments

Post a Comment

Popular posts from this blog

Datastax Error : Cassandra - Saved cluster name Test Cluster != configured name

-
Image
While changing existing cluster name in datastax, i got name mismatch exception in log file. Process i followed was changed the cluster name in cassandra.yaml configuration file. $ nano /etc/dse/cassandra/cassandra.yaml     cluster_name: 'Test Cluster1' Solution :  Change the cluster name  to "New Cluster Name" : $ cqlsh `hostname -I` cqlsh> UPDATE system.local SET cluster_name = 'New Cluster Name' where key='local'; cqlsh> exit; $ nodetool flush system Stop the cluster : $ sudo service dse stop;sudo service datastax-agent stop Edit the file : $ sudo vi /etc/dse/cassandra/cassandra.yaml     cluster_name: ' New Cluster Name ' Start the cluster : $ sudo service dse start;sudo service datastax-agent start Check Installation log :  $ cat /var/log/cassandra/output.log Reference  : https://support.datastax.com/hc/en-us/articles/205289825-Change-Cluster-Name-
Continue Reading »»»»

Datastax Error : Cannot start node if snitch's data center (dc1) differs from previous data center (dc2)

-
Image
This error occurs when the node try to start and see it has snitch information which differ from previous data center. In my case, i was using DseSimpleSnitch which name the datacenter based on workload type. Previously i had enabled solr in dse default configuration file ( /etc/default/dse ) and datasenter name was Solr. Now i tried to start the dse with graph enabled, which change the name to datacenter SearchGraph which is default workload type name. CassandraDaemon.java:698 - Cannot start node if snitch's data center (Solr) differs from previous data center (SearchGraph). Please fix the snitch configuration, decommission and rebootstrap this node or use the flag -Dcassandra.ignore_dc=true Solution : 1) In the last line of  cassandra-env.sh  set in the jvm opts as given below : JVM_OPTS="$JVM_OPTS -Dcassandra.ignore_dc=true" 2) If it starts successfully, execute: nodetool repair nodetool cleanup
Continue Reading »»»»

Configure Nagios plugin " check_logfiles " for scanning log file

-
I was checking for a nagios plugin which could filter the specified log file which i had mentioned and notify me whenever a warning or error appeared in log. Finally i get in hand check_logfile created by Gerhard Lausser. Check_logfiles is used to scan the lines of a file for regular expressions. Install the plugin on client system :  $ wget https://labs.consol.de/assets/downloads/nagios/check_logfiles-3.8.0.2.tar.gz $ tar xvzf check_logfiles-3.8.0.2.tar.gz $ cd check_logfiles-3.8.0.2/ $ ./configure $ make $ make install check_logfiles executable file will get installed on location "/usr/local/nagios/libexec" Configure Check Pattern : $ nano  /etc/nagios-plugins/config/check_log.cfg @searches = (   {     tag => 'Error',     logfile => '/var/logs/error.log',     criticalpatterns => 'Error',   } ); Here i am filtering pattern "Error" from log file which i had mentioned in logfile parameter section C
Continue Reading »»»»

Popular posts from this blog

Datastax Error : Cassandra - Saved cluster name Test Cluster != configured name

-
Image
While changing existing cluster name in datastax, i got name mismatch exception in log file. Process i followed was changed the cluster name in cassandra.yaml configuration file. $ nano /etc/dse/cassandra/cassandra.yaml     cluster_name: 'Test Cluster1' Solution :  Change the cluster name  to "New Cluster Name" : $ cqlsh `hostname -I` cqlsh> UPDATE system.local SET cluster_name = 'New Cluster Name' where key='local'; cqlsh> exit; $ nodetool flush system Stop the cluster : $ sudo service dse stop;sudo service datastax-agent stop Edit the file : $ sudo vi /etc/dse/cassandra/cassandra.yaml     cluster_name: ' New Cluster Name ' Start the cluster : $ sudo service dse start;sudo service datastax-agent start Check Installation log :  $ cat /var/log/cassandra/output.log Reference  : https://support.datastax.com/hc/en-us/articles/205289825-Change-Cluster-Name-
Continue Reading »»»»

Datastax Error : Cannot start node if snitch's data center (dc1) differs from previous data center (dc2)

-
Image
This error occurs when the node try to start and see it has snitch information which differ from previous data center. In my case, i was using DseSimpleSnitch which name the datacenter based on workload type. Previously i had enabled solr in dse default configuration file ( /etc/default/dse ) and datasenter name was Solr. Now i tried to start the dse with graph enabled, which change the name to datacenter SearchGraph which is default workload type name. CassandraDaemon.java:698 - Cannot start node if snitch's data center (Solr) differs from previous data center (SearchGraph). Please fix the snitch configuration, decommission and rebootstrap this node or use the flag -Dcassandra.ignore_dc=true Solution : 1) In the last line of  cassandra-env.sh  set in the jvm opts as given below : JVM_OPTS="$JVM_OPTS -Dcassandra.ignore_dc=true" 2) If it starts successfully, execute: nodetool repair nodetool cleanup
Continue Reading »»»»

Configure Nagios plugin " check_logfiles " for scanning log file

-
I was checking for a nagios plugin which could filter the specified log file which i had mentioned and notify me whenever a warning or error appeared in log. Finally i get in hand check_logfile created by Gerhard Lausser. Check_logfiles is used to scan the lines of a file for regular expressions. Install the plugin on client system :  $ wget https://labs.consol.de/assets/downloads/nagios/check_logfiles-3.8.0.2.tar.gz $ tar xvzf check_logfiles-3.8.0.2.tar.gz $ cd check_logfiles-3.8.0.2/ $ ./configure $ make $ make install check_logfiles executable file will get installed on location "/usr/local/nagios/libexec" Configure Check Pattern : $ nano  /etc/nagios-plugins/config/check_log.cfg @searches = (   {     tag => 'Error',     logfile => '/var/logs/error.log',     criticalpatterns => 'Error',   } ); Here i am filtering pattern "Error" from log file which i had mentioned in logfile parameter section C
Continue Reading »»»»