Understanding_active_and_passive_varyon_in_enhanced_concurrent_mode

An enhanced concurrent volume group can be made active on the node, or varied on, in two states: active or passive.

Note that active or passive state varyons are done automatically by HACMP upon detection of the enhanced concurrent mode volume group, based on the state of the volume group and current cluster configuration.

Important: All nodes in the cluster must be available before making any LVM changes. This ensures that all nodes have an accurate view of the state of the volume group. For more information about safely performing a forced varyon operation, and on instructions how to configure it in SMIT, see Forcing a varyon of volume groups.

• Active state varyon 
  Active state varyon behaves as ordinary varyon, and makes the logical volumes normally available.
• Passive state varyon 
  When an enhanced concurrent volume group is varied on in passive state, the LVM provides an equivalent of fencing for the volume group at the LVM level.
• Using active or passive state varyon in HACMP 
  HACMP detects when a volume group included in a shared resource group is converted to or defined as an enhanced concurrent mode volume group, and notifies the LVM which node currently owns the volume group.
• Verification checks for shared volume groups defined for auto varyon 
  Shared volume groups listed within a resource group must have the auto-varyon attribute in the AIX ODM set to No. C-SPOC does not allow you to define a volume group with auto-varyon set to yes. However, after the volume group is defined you can change the auto-varyon attribute from yes to no.
• Checking the status of a volume group 
  As with regular cluster takeover operations, you can debug and trace the cluster activity for fast disk takeover using the information logged in the hacmp.out file. You may check the status of the volume group by issuing the lsvg command.
• Avoiding a partitioned cluster 
  When configuring enhanced concurrent volume groups in shared resource groups, ensure that multiple networks exist for communication between the nodes in the cluster, to avoid cluster partitioning.

이 글은 스프링노트에서 작성되었습니다.