Troubleshooting manual failover environment installation errors

If you see errors when setting up a manual failover environment, check the following:
  • If your production server, failover server, and file server cannot communicate with each other, make sure the correct ports on your firewall are open.
  • If you notice a decrease in system or network performance after setting up a manual failover environment, your network infrastructure might not support the increase in bandwidth the manual failover environment requires. Contact your system administrator and make sure the correct file sharing technology is used. For example, instead of using NFS, it might be more efficient for your network to use a SAN. Make sure you have a dedicated network or your router or network switch is correctly configured for the increase in bandwidth.
  • If you receive any licensing errors after the install, delete the Ricoh ProcessDirector license file /aiw/aiw1/config/license/license.key. Copy scripts/failover-setup-rpd-node.sh from the installation media to /tmp and run the script. When prompted, if you accept the license agreement, make sure you type yes and press Enter. If your problem persists, contact Software Support.
  • If you receive errors when trying to mount NFS, this problem might be caused by certain versions of Linux defaulting to NFS version 4. By default, NFS version 4 does not allow shared directory ownership. The system must be set up so that the production and failover servers can change the file and directory ownership of the files on the NFS shares. If you cannot modify NFS to fix this issue, modify the directory’s ownership.

    Type chown aiw1:aiwgrp1 /aiw/aiw1 to change the ownership.

    If this does not work, edit /etc/sysconfig/autofs using a text editor and change MOUNT_NFS_DEFAULT_PROTOCOL=4 to MOUNT_NFS_DEFAULT_PROTOCOL=3. For example:

    # MOUNT_NFS_DEFAULT_PROTOCOL - specify the default protocol used by
    #                              mount.nfs(8). Since we can't identify
    #                              the default automatically we need to
    #                              set it in our configuration. This will
    #                              only make a difference for replicated
    #                              map entries as availability probing isn't
    #                              used for single host map entries.
    #
    MOUNT_NFS_DEFAULT_PROTOCOL=3

    Then edit /etc/nfsmount.conf and change Defaultvers=4 to Defaultvers=3. Then change Nfsvers=4 to Nfsvers=3.

    Update the system by restarting NFS or rebooting the servers.

  • If you switch between the production server and the failover server, and one of the servers cannot start up, then there might be locks on the file system. This is typically caused by one of the servers not being shut down correctly. To determine if this is the issue:
    1. In a command prompt, type su - aiw1 -c "db2start;db2 connect to aiwdb"

      Look in the results for a message like this: SQL1391N The database is already in use by another instance of the database manager. SQLSTATE=51023

      If you see a similar message, you must release the locks on the database.

    2. Reboot the fileserver to release the locks.