Check category: Availability

Check description: Verifies that all the disks in the disk group in a campus cluster have site names. Also verifies that all volumes on the disk group have the same number of plexes on each site in the campus cluster.

Check procedure:

• Verifies whether the disk group has the proper campus cluster configuration.

Check recommendation: Make sure that the site name is added to each disk in a disk group. To verify the site name, enter the following command:
# vxdisk -s list disk_name
On each site in the campus cluster, make sure that all volumes on the disk group have the same number of plexes. To verify the plex and subdisk information of a volume created on a disk group, enter the following command:
# vxprint -g disk_group.

Check category: Availability

Check description: The check fails if the volumes that have the allsites flag set do not have complete plexes allocated from all the configured sites.

Check procedure:

• Verifies that allsites volumes have complete plexes on all the configured sites.

Check recommendation:

• Make sure that the allsites volumes have a plex on the configured sites.
  
  
• Make sure that the subdisks are site-confined.

Check category: Availability

Check description: The check fails if the vxattachd daemon is not running. When the faulted disks are reconnected, this daemon automatically reattaches disks, plexes, and sites.

Check procedure:

• Verifies that the vxattachd daemon is running.

Check recommendation: Make sure that vxattachd and vxrestored are running.

Check category: Availability

Check description: Checks whether all sites have recent configuration copies and log copies, distributed in balanced manner across the sites. Each site should have at least one configuration copy and one log copy.

Check procedure:

• Verifies that each site has at least one configuration copy and one log copy.

Check recommendation: Make sure that the configuration and log copies are distributed in balanced manner across sites.

Check category: Availability

Check description: The check fails if the disks are not tagged with valid sites that are configured in the disk group.

Check procedure:

• Verifies that the disks in the disk group which have sites configured, are not untagged.

Check recommendation:

• Make sure that the disks are tagged with a valid site that is configured in the disk group.
  
  
• Make sure that the disks from an enclosure are tagged with the same site.

Check category: Availability

Check description: Checks whether the hosts are tagged with a site tag which is a configured site in all imported disk groups.
Note: This check does not apply if the host does not have an imported disk group.

Check procedure:

• Verifies that the host has been tagged with a site.

Check recommendation: Make sure that you tag the host with a valid site and that the imported disk groups have the host site as a configured site.

Check category: Availability

Check description: The check fails if the Volume Manager (VxVM) objects are not correctly configured to maintain site-consistency.

Check procedure:

• Validates that all the volumes (both data and log volumes) that have the site-consistency flag set have at least one complete plex on each of the configured sites in the disk group.

Check recommendation:

• Make sure to attach a DCO log with a version => 20 to the site-consistent volumes.
  
  
• Make sure that the site-consistent volumes have a complete plex in all the configured sites.
  
  
  
• Make sure that the state of the plexes of the site-consistent volumes match with the state of the site.

Check category: Availability

Check description: Verifies whether I/O fencing is properly configured for cluster.

Check procedure:

• Determines the fencing mode using the 'vxfenadm' command and verifies whether fencing mode is set to either SCSI-3 or SYBASE or CPS or customized.

Check recommendation: Either I/O fencing is not running on the system or fencing mode is not configured properly. It is recommended to configure I/O fencing with fencing mode configured as per product requirements when using SFCFS to avoid VxFS file system corruption.

Check category: Best practices

Check description: Checks whether the dgfailpolicy is set to 'leave' for all site-consistent disk groups with version prior to 170. This is the recommended policy for site-consistent disk groups.

Check procedure:

• Verifies that the sites configured disk groups have dgfailpolicy set to 'leave'. This is the recommended policy for campus cluster.

Check recommendation: Make sure that the dgfailpolicy of the site configured disk groups is 'leave'.

Check category: Best practices

Check description: Checks whether the diskdetachpolicy is set to 'Global' for all site-consistent disk groups. This is the recommended policy for site-consistent disk groups.

Check procedure:

• Verifies that the diskdetachpolicy for all site-consistent disk groups is set to 'Global'. This is the recommended policy for site-consistent disk groups in campus cluster.

Check recommendation: Make sure that the diskdetachpolicy for all the site-consistent disk groups is 'Global'.

Check category: Performance

Check description: Checks whether the read policy of the volumes is set to 'site read'. This is the recommended policy for campus clusters.

Check procedure:

• Verifies that the read policy of the volumes in a site-configured disk group is 'site read'.

Check recommendation: Make sure that the read policy of the volumes in the site-configured disk groups is 'site read'.