Upgrading the container to 11.0.X.X

Deployment options: Netezza Performance Server for Cloud Pak for Data System

Learn how to upgrade the Netezza Performance Server container to 11.0.X.X.

Procedure

  1. Create a directory on the e1n1 bare metal control plane node.
    mkdir /opt/ibm/appliance/storage/ips/ips1_deployment/DIRECTORY
    Example:
    mkdir /opt/ibm/appliance/storage/ips/ips1_deployment/11.x.x.x
    In the example, the directory is named 11.x.x.x.
  2. Download container.tar.gz from Fix Central and save it in the newly created directory.
  3. Change directories to the folder in which the container is placed.
    [root@e1n1 11.x.x.x]# cd /opt/ibm/appliance/storage/ips/ips1_deployment/11.x.x.x
  4. Unpack container.tar.gz.
    [root@e1n1 11.x.x.x]# tar -zxvf container.tar.gz
    ./
    ./ipshost-image.tgz
    ./ips-deploy.sh
    ./ips_check
    ./nzconfigure.conf
    ./IBM_Netezza_Software_Support_Tools-11.0.4.1-20200216.tgz
    ./npsos.11.0.4.1.tar.gz
    ./unpack
  5. If gpfstoken is present on the system, remove the token file.
    • If you are upgrading from 11.0.6.1 or later, run:
      docker exec ipshost1 rm /nz/.gpfstoken
    • If you are upgrading from a version earlier than 11.0.6.1, run the following commands:
      1. docker exec ipshost1 bash -c "sed -i -e '/net.ipv4.conf.all.rp_filter/d' /etc/sysctl.conf"
      2. docker exec ipshost1 bash -c "sed -i -e '/net.ipv4.conf.default.rp_filter/d' /etc/sysctl.conf"
      3. docker exec ipshost1 bash -c "sed -i -e '/net.ipv4.conf.mgt1.rp_filter/d' /etc/sysctl.conf"
      4. docker exec ipshost1 bash -c "echo 'net.ipv4.conf.all.rp_filter = 0' >> /etc/sysctl.conf"
      5. docker exec ipshost1 bash -c "echo 'net.ipv4.conf.default.rp_filter = 0' >> /etc/sysctl.conf"
      6. docker exec ipshost1 bash -c "echo 'net.ipv4.conf.mgt1.rp_filter = 0' >> /etc/sysctl.conf"
  6. Ensure that the database is not running.
    nzstate
    • If the database is Stopped, you can see the following output.
      [nz@xxx-npshost ~]$ nzstate
      System state is 'Stopped'.
    • If the database is not Stopped, go to the ipshost1 container and stop the system.
      nzstop
  7. Establish whether your system is Lenovo or Dell.
    /opt/ibm/appliance/platform/xcat/scripts/xcat/display_nodes.py --vendor
    • If you have a Dell system, verify that you can access hwinventory for each of the Dell SPUs.
      /opt/dell/srvadmin/sbin/racadm -u root -p calvin -r e5n1bmc hwinventory nic
      • If ERROR: Unable to login to RAC using the specified address. does not appear, you can replace e5n1bmc with Dell SPU nodes.
      • If ERROR: Unable to login to RAC using the specified address. appears:
        1. Run the command for that SPU node and wait for 5 - 10 minutes.
          ipmitool -H e5n1bmc -U root -P calvin -I lanplus mc reset cold
        2. Replace e5n1bmc with each Dell SPU node.
        3. Verify that you can access hwinventory again.
          /opt/dell/srvadmin/sbin/racadm -u root -p calvin -r e5n1bmc hwinventory nic
  8. Upgrade the container.
    • For Lenovo, run:
      ./ips-deploy.sh -o /opt/ibm/appliance/platform/onecli -d /opt/ibm/appliance/storage/ips -i 1 -u
    • For Dell, run:
      ./ips-deploy.sh -r /opt/dell/ -d /opt/ibm/appliance/storage/ips -i 1 -u
    Where:
    -o
    The folder that contains the onecli command.
    -r
    The folder that contains the racadm's srvadmin directory.
    -d
    The Netezza Performance Server base folder.
    -i
    The Netezza Performance Server instance number.
    -u
    Used to reupgrade the container without reinitializing.
  9. Optional: If you are upgrading the container from version 11.0.6.1 or earlier to version 11.0.7.0 or later, prune logs:
    docker exec ipshost1 cp /nz/kit/sys/init/nztmpwatch.in /etc/cron.daily/nztmpwatch
    docker exec ipshost1 sed -i -e 's,@@PREFIX@@,/nz,' /etc/cron.daily/nztmpwatch
    docker exec ipshost1 sed -i -e 's,@@DATADIR@@,/nz/data.1.0,' /etc/cron.daily/nztmpwatch
    
    docker exec ipshost1 bash -c "[ -d /nz/kit.bak/log ] && tmpwatch --nodirs --mtime 360 /nz/kit.bak/log"
    docker exec ipshost1 bash -c "[ -d /nz/kit.bak/log/ldrDataRdr ] && tmpwatch --nodirs --mtime 120 /nz/kit.bak/log/ldrDataRdr"
    docker exec ipshost1 bash -c "[ -d /nz/kit.bak/log/unlDataWrtr ] && tmpwatch --nodirs --mtime 120 /nz/kit.bak/log/unlDataWrtr"

What to do next

Upgrade the software.