InfoScale™ 9.0 Replication Administrator's Guide - AIX
- Section I. Getting started with Volume Replicator
- Introducing Volume Replicator
- Understanding how Volume Replicator works
- How VVR uses kernel buffers for replication
- Replication in a shared disk group environment
- Using SmartTier with VVR
- Understanding the VVR snapshot feature
- About VVR compression
- Planning and configuring replication
- Before you begin configuring
- Choosing the mode of volume replication
- Planning the network
- Sizing the SRL
- Understanding replication settings for a Secondary
- Configuring VVR in a VCS environment
- Using the primary-elect feature to choose the primary site after a site disaster or network disruption
- Requirements for configuring VVR in a VCS environment
- Example setting up VVR in a VCS environment
- Configuring the agents for a bunker replication configuration
- Section II. Setting up and administering VVR
- Setting up replication
- Creating a Replicated Data Set
- Creating a Primary RVG of an RDS
- Adding a Secondary to an RDS
- Changing the replication settings for a Secondary
- Synchronizing the Secondary and starting replication
- Starting replication when the data volumes are zero initialized
- Displaying configuration information
- Displaying RVG and RDS information
- Displaying information about data volumes and volume sets
- Displaying information about Secondaries
- Displaying statistics with the vrstat display commands
- Collecting consolidated statistics of the VVR components
- Displaying network performance data
- Administering Volume Replicator
- Administering data volumes
- Associating a volume to a Replicated Data Set
- Associating a volume set to an RDS
- Associating a Data Change Map to a data volume as a log plex
- Resizing a data volume in a Replicated Data Set
- Administering the SRL
- Incrementally synchronizing the Secondary after SRL overflow
- Administering replication
- Administering the Replicated Data Set
- Administering Storage Checkpoints
- Creating RVG snapshots
- Using the instant snapshot feature
- About instant full snapshots
- Preparing the volumes prior to using the instant snapshot feature
- Creating instant full snapshots
- About instant space-optimized snapshots
- Creating instant space-optimized snapshots
- About instant plex-breakoff snapshots
- Administering snapshots
- Using the traditional snapshot feature
- Using Veritas Volume Manager FastResync
- Verifying the DR readiness of a VVR setup
- Backing up the Secondary
- Administering data volumes
- Using VVR for off-host processing
- Transferring the Primary role
- Migrating the Primary
- About taking over from an original Primary
- Failing back to the original Primary
- Choosing the Primary site after a site disaster or network disruption
- Troubleshooting the primary-elect feature
- Replication using a bunker site
- Introduction to replication using a bunker site
- Setting up replication using a bunker site
- Using a bunker for disaster recovery
- Replication using a bunker site in a VCS environment
- Configuring and administering VVR using System Management Interface Tool
- Accessing Volume Replicator interface in SMIT
- Setting up a simple Volume Replicator configuration using SMIT
- Displaying configuration information using SMIT
- Administering Volume Replicator using SMIT
- Taking instant snapshot of data volumes of an RVG using SMIT
- Associating a volume to a Replicated Data Set using SMIT
- Transferring the Primary role using SMIT
- Troubleshooting VVR
- Recovery from configuration errors
- Errors during an RLINK attach
- Errors during modification of an RVG
- Recovery on the Primary or Secondary
- Recovering from Primary data volume error
- Primary SRL volume error cleanup and restart
- Primary SRL header error cleanup and recovery
- Secondary data volume error cleanup and recovery
- Tuning replication performance
- SRL layout
- Tuning Volume Replicator
- VVR buffer space
- Tuning VVR compression
- VVR buffer space
- Setting up replication
- Section III. Analyzing your environment with Volume Replicator Advisor
- Introducing Volume Replicator Advisor (VRAdvisor)
- Collecting the sample of data
- About collecting the sample of data
- Collecting the sample of data on UNIX
- Collecting the sample of data on Windows
- Analyzing the sample of data
- About analyzing the sample of data
- Analyzing the collected data
- Understanding the results of the analysis
- Viewing the analysis results
- Recalculating the analysis results
- Installing Volume Replicator Advisor (VRAdvisor)
- Section IV. VVR reference
- Appendix A. VVR command reference
- Appendix B. Using the In-band Control Messaging utility vxibc and the IBC programming API
- Using the IBC messaging command-line utility
- Examples - Off-host processing
- In-band Control Messaging API
- Appendix C. Volume Replicator object states
- Appendix D. Alternate methods for synchronizing the Secondary
- Using the full synchronization feature
- Using block-level backup and Storage Checkpoint
- Using difference-based synchronization
- Examples for setting up a simple Volume Replicator configuration
- Appendix E. Migrating VVR from IPv4 to IPv6
- Migrating VVR to support IPv6 or dual stack
- About migrating to IPv6 when VCS global clustering and VVR agents are not configured
- About migrating to IPv6 when VCS global clustering and VVR agents are configured
- About migrating to IPv6 when VCS global clustering and VVR agents are configured in the presence of a bunker
- Migrating to IPv6 when VCS global clustering and VVR agents are configured in the presence of a bunker
- Appendix F. Sample main.cf files
RV_IBC_SEND
This ioctl can only be issued against the Primary RVG with a valid key obtained from the RV_IBC_REGISTER ioctl. The ioctl inserts an IBC message into the data update stream of one or all RLINKs attached to the RVG.
If it is desired that an IBC message be inserted at an exact location in the update stream, such as a point of application-level consistency, then there must be no concurrent write activity to the RVG when the RV_IBC_SEND ioctl is issued. Note that writes made to the block device interface of a data volume may be cached, so a disk sync must be done before issuing the ioctl. If there are active writes to the RVG when the ioctl is issued, the insertion point of the IBC message in the RLINK update data stream is arbitrary in relation to that activity.
The ioctl returns using the same semantics as a data write to the RVG; it returns when the IBC message has been committed to the SRL and has also been transferred to all synchronous-mode replicas attached to the RVG.
The ioctl argument structure for the RV_IBC_SEND command is:
struct ibc_send_args { /* IOCTL_STRUCT */ vx_u32_t ibc_magic; vx_u32_t ibc_version; ibc_appid_t application_id; char replica[NAME_SZ]; int flags; int freeze_timeout; caddr_t msg_buf; int msg_len; };
Argument ibc_magic is used to verify whether the ioctl structure is a valid 4.0 structure. It should be set to NM_IBC_MAGIC.
Argument ibc_version specifies the current IBC version. It should be set to NM_IBC_VERSION.
Argument application_id is the key returned by the RV_IBC_REGISTER ioctl. A registration must be done before the RV_IBC_SEND ioctl can be used.
Argument replica specifies the name of the RLINK to which the IBC message is to be send. The null string specifies a broadcast to all RLINKs currently attached to the Primary RVG.
Argument flags set to IBC_FREEZE causes the secondary replication to freeze for the time-out period specified in freeze_timeout. If replication is not desired to be frozen, then flags should be set to 0.
Argument freeze_timeout specifies a time-out value in seconds between delivery of an IBC message on the Secondary and execution of an RV_IBC_UNFREEZE ioctl against the Secondary RVG. When the time-out expires, replication at the Secondary continues. A time-out value of zero is used to specify no time-out.
Argument msg_buf is a pointer to a buffer containing an IBC message. The content of an IBC message is user-defined and has no restriction except size.
Argument msg_len is the length, in bytes, of the IBC message and can be no greater than 128k bytes.
On failure, possible values returned in errno are:
EIBC_NO_RLINK | No RLINK or specified RLINK exists |
EIO I/O | I/O error while logging the IBC message |
EIBC_MSG_LENGTH | Message is greater than maximum allowable length (128K) |