Can't monitor nodes performance after upgrading CMON from 1.2.8 to 1.2.9

Comments

3 comments

  • Avatar
    Ashraf Sharif

    Hi Diptin,

    How did you perform the upgrade? Did you follow the instructions here: http://support.severalnines.com/entries/21095371-CMON-1-2-9-released-upgrade-instructions

    ClusterControl 1.2.9 requires port 9500 to be opened on ClusterControl node.

    Regards,

    Ashraf

  • Avatar
    Diptin Patel

    Hi Ashraf,


    Yes I followed the same documentation for upgrade. For applying the schema, I omitted steps to update from 1.2.6 to 1.2.8, since I was already at 1.2.8.

     

    I applied following schemas.

    mysql -f -h127.0.0.1 -ucmon -p cmon < /usr/share/cmon/cmon_db.sql

    mysql -f- h127.0.0.1 -ucmon -p cmon < /usr/share/cmon/cmon_db_mods-1.2.8-1.2.9.sql
    mysql -f -h127.0.0.1 -ucmon -p cmon < /usr/share/cmon/cmon_data.sql

     

     

    However, Alex was able to assist me with this incident. I had another ticket open with similar issue.

     

    I’m still having one small issue though. It’s a 3 node cluster, and cluster is online (with 3 nodes) and active under overview tab, but under Nodes tab I’m missing 3rd node and it’s has following message:

     

    Node is OK

    Cluster has an even number of Galera nodes. Add a garbd (Galera arbitrator), another node or remove a node to have an odd number number of nodes.

     

    Can you please assist with this? I tried added the 3rd node, but it failed.

    Thanks,

    Diptin

  • Avatar
    Ashraf Sharif

    Hi Diptin,

    What is the output of following statements on ClusterControl node?

    mysql -u root -p cmon -e 'select * from mysql_server'
    mysql -u root -p cmon -e 'select * from hosts'

    Also please also send the output of following command:

    grep mysql_server_addresses /etc/cmon.cnf

     ** If it returns nothing, replace /etc/cmon.cnf with /etc/cmon.d/cmon_<cluster_id>.cnf

    Regards,

    Ashraf

Please sign in to leave a comment.

Powered by Zendesk