学习NBU文档——02.ppt

If the system disk does not have a usable mirror or if the user chooses to skip the new split mirror functionality, the upgrade is performed on the whole boot disk group as in previous SF versions. If the user chooses to split the boot disk mirror during the upgrade, a disk group name is required as the backup disk group and the installer upgrades the system disk only after the mirror is split. After the upgrade completes, the user must decide when the split disk group is to be joined back to the upgraded boot disk group using the vxrootadm join command as displayed on the slide. Note that Symantec highly recommends not to include any data disks inside the boot disk group. If the system disk is used for some data volumes, you must ensure that all volumes are mirrored prior to starting the upgrade. The next slide includes a demonstration of the new vxrootadm integration feature during the upgrade of an SFHA cluster. 2-* SFHA 6.0 Differences: Introduction and Deployment In this demonstration, a full upgrade of a Storage Foundation High Availability 5.1 SP1 two-node cluster to Storage Foundation High Availability 6.0 is performed using the Storage Foundation 6.0 installer. The root/boot disks are encapsulated and mirrored on both nodes. Upgrade steps are performed on both systems as needed, but demonstration details are only shown for one system. ? The hastart –version command is used to confirm the VCS engine version is 5.1SP1. ? The rpm –q command is used to generate a summary listing of VRTS rpms, which includes identifying the version. ? The vxdisk command is used to confirm that the root/boot disk is encapsulated and mirrored. ? The Storage Foundation 6.0 installation files are located on an NFS share, which is navigated to. ? The contents of the Storage Foundation 6.0 installation directory are listed. ? The installer command is run with no options. ? Option G is selected to upgrade a product. ? Option 1 is selected perform a full upgrade. ? The installer is a