Version 1.0.3.0 release notes (November 2017)

Connection authentication and node monitoring are among the new features introduced with this release.

What's new

Administering
  • The SMTP connection can now be authenticated.
  • A node's memory swap is now monitored, and when it is used often, a new alert is sent.
  • The LDAP platform is now monitored and can be viewed in the software inventory.
  • A node's reachability in the management network is also monitored. A new alert with the reason code of 110 has been added to report issues.

Known issues

Transaction roll back due to system error
The following message about the transaction is displayed: 
SQL1229N The current transaction has been rolled back because of a system error. SQLSTATE=40504
Workaround
Run the following commands:
apstop
apstart
Clocks on nodes lose synchronization
The clocks are out-of-sync between nodes, which might result in the ap issues command returning the following message:
SW_NEEDS_ATTENTION | 406: Time on node is not synchronized
In addition, the ap sw command reports that there is more than one device in the DOWN or UNKNOWN state.
Workaround
  • If there are external NTP servers configured in /etc/ntp.conf:
    1. Ensure that the NTP servers are reachable from each node. If not, resolve any network problems.
    2. Run the following command at least twice on each node:
      systemctl restart ntpd
  • If there are no external NTP servers configures in /etc/ntp.conf, run the following command at least twice on each node:
    systemctl restart ntpd
Attempt to upgrade to IAS version 1.0.3.0 using root user account fails with GPFS related errors
There is a defect in the process of upgrading to version 1.0.3.0 from 1.0.1.x and 1.0.2.x systems. It results in a partial upgrade, which cannot be automatically resumed. Upgrade attempts fail with an error:
error messages: Errors encountered
GpfsUpgrader.install : Fatal Problem: Failed to stop GPFS.
To fully confirm the issue, check if the apupgrade log displays the following message: 

ERROR: CmdRunnerError: Fatal Problem: Could not upgrade rpms on all nodes.

ERROR:
INFO: Failed. Failed to finish GPFS component upgrade.
error messages: Errors encountered
GpfsUpgrader.install : Fatal Problem: Could not upgrade rpms on all nodes.
ERROR: Errors encountered
ERROR: GpfsUpgrader.install : Fatal Problem: Could not upgrade rpms on all nodes.
Workaround
Note: All of the steps require root authority. To log in as the root user, contact IBM support.
  1. Stop GPFS: 
    /opt/ibm/appliance/platform/xcat/scripts/storage/stop_gpfs.py
  2. Check the version of GPFS that is installed by running the following command on all nodes: 
    rpm -qa | grep gpfs | egrep -v 'ibm-ca'

    Example output: 

    gpfs.ext-4.2.2-0.ppc64le
metagpfs-1.0.2.0-20171020181018b50.noarch
gpfs.docs-4.2.2-0.noarch
gpfs.crypto-4.2.2-0.ppc64le
gpfs.gskit-8.0.50-57.ppc64le
gpfs.base-4.2.2-0.ppc64le
gpfs.gpl-4.2.2-0.noarch
gpfs.msg.en_US-4.2.2-0.noarch
gpfs.license.adv-4.2.2-0.ppc64le
gpfs.adv-4.2.2-0.ppc64le
    All of the GPFS packages in the example node are at the 4.2.2-0 version level.
  3. Compare the versions of GPFS with the versions that are located in the /install/app_img/gpfs/ directory by running
    ls /install/app_img/gpfs/
    Example output: 
    gpfs.adv-4.2.2-0.ppc64le.rpm
gpfs.base-4.2.2-0.ppc64le.rpm
gpfs.crypto-4.2.2-0.ppc64le.rpm
gpfs.docs-4.2.2-0.noarch.rpm
gpfs.ext-4.2.2-0.ppc64le.rpm
gpfs.gpl-4.2.2-0.noarch.rpm
gpfs.gskit-8.0.50-57.ppc64le.rpm
gpfs.license.adv-4.2.2-0.ppc64le.rpm
gpfs.msg.en_US-4.2.2-0.noarch.rpm
    All of the GPFS packages in the example output are also at the 4.2.2-0 version level.
  4. If there is no discrepancy between the GPFS versions, there is no need to take any actions on this node.
    If there exists a discrepancy on a node, remove all offending GPFS related packages from the node one at a time. Then run the following command:
    rpm -e name of each GPFS package, 1 at a time
    When all of them are successfully removed, install the old GPFS packages for this node 
    rpm -i /install/app_img/gpfs/NAME.rpm
rpm -Uvh /install/app_img/gpfs_update/NAME_update.rpm
rpm -i /install/app_img/gpfs_update/metagpfs*rpm
    Example usage: 
    rpm -i /install/app_img/gpfs/gpfs.adv-4.2.2-0.ppc64le.rpm
rpm -Uvh /install/app_img/gpfs_update/gpfs.adv-4.2.2-0.ppc64le.update.rpm
  5. Start GPFS with the following command: 
    #/opt/ibm/appliance/platform/xcat/scripts/storage/start_gpfs.py
  6. Resume upgrade by rerunning the original upgrade command.