Monitoring Replication Delays and RPO (CLI)

Language:

With asynchronous replication, a time delay occurs when writing data from the source to its replica on the target. A warning and error alert can be set to notify the administrator when a replication delay is approaching or exceeds the replication point objective (RPO). These alerts prompt the administrator to check for networking problems, application problems, and to examine the health of the storage appliance using analytics.

A replication delay alert can be set when creating or editing a replication action.

To set replication delay alerts:

Navigate to the project or share, and enter action:

host_source:shares New_Project replication> action

Display the properties by entering get.

host_source:shares New_Project action (uncommitted)> get
                 origin_pkg_id =
                        target = replication-target
                          pool = demo_pool
                       enabled = true
                    continuous = false
                 include_snaps = true
   retain_user_snaps_on_target = false
                         dedup = false
  include_clone_origin_as_data = false
                 max_bandwidth = unlimited
                    bytes_sent = 0
                estimated_size = 0
           estimated_time_left = 00:00:00
            average_throughput = 0B/s
                       use_ssl = true
                   compression = on
                   export_path =
                         state = idle
             state_description = Idle (no update in progress)
                export_pending = false
                       offline = false
                   next_update = Sync now
                   replica_lag = P1H30M
        replica_data_timestamp = Wed Feb 15 2016 12:12:05 GMT+0000 (UTC)
                     last_sync = Wed Feb 15 2016 22:32:59 GMT+0000 (UTC)
                      last_try = Wed Feb 15 2016 22:32:59 GMT+0000 (UTC)
                   last_result = success
      recovery_point_objective =
     replica_lag_warning_alert =
       replica_lag_error_alert =

Set the RPO and replica lag properties for this action as shown in the following example:

host_source:shares New_Project action (uncommitted)> set recovery_point_objective=50min
                         recovery_point_objective = 50 minutes (uncommitted)
host_source:shares New_Project action (uncommitted)> set replica_lag_warning_alert=50
                             replica_lag_warning = 50% (uncommitted)
host_source:shares New_Project action (uncommitted)> set replica_lag_error_alert=180
                     replica_lag_error = 180% (uncommitted)

For a description of all properties, see Replication Action Properties.

Commit the changes for this action.

host_source:shares New_Project action (uncommitted)> commit

To view the current properties for the action, enter ls. The RPO and replica lag portion of the output is shown in the following example:

host_source:shares New_Project action (uncommitted)> ls
.
.
                   replica_lag = P1H30M
        replica_data_timestamp = Wed Feb 15 2016 12:12:05 GMT+0000 (UTC)
                     last_sync = Wed Feb 15 2016 22:32:59 GMT+0000 (UTC)
                      last_try = Wed Feb 15 2016 22:32:59 GMT+0000 (UTC)
                   last_result = success
      recovery_point_objective = 50 minutes
     replica_lag_warning_alert = 50%
       replica_lag_error_alert = 180%

Oracle^® ZFS Storage Appliance Administration Guide, Release OS8.8.0

Monitoring Replication Delays and RPO (CLI)

Related Topics