Modifies cross data center replication (XDCR) settings
couchbase-cli setting-xdcr [--cluster <url>] [--username <user>]
[--password <password>] [--checkpoint-interval <seconds>]
[--worker-batch-size <num>] [--doc-batch-size <kilobytes>]
[--failure-restart-interval <seconds>] [--source-nozzle-per-node <num>]
[--target-nozzle-per-node <num>] [--bandwidth-usage-limit <num>]
[--enable-compression <num>] [--stats-interval <milliseconds>]
[--optimistic-replication-threshold <bytes>] [--log-level <level>]
[--max-processes <num>]
This command sets global default settings for all XDCR replications. If you only want to change the settings for a single XDCR replication see the man:couchbase-cli-xdcr-replicate[1] command.
- --checkpoint-interval <seconds>
-
The interval between checkpoints in seconds. The value of this option must be between 60 and 14,400.
- --worker-batch-size <num>
-
The worker batch size. The value of this option must be between 500 and 10,000.
- --doc-batch-size <kilobytes>
-
The document batch size in Kilobytes. The value of this option must be between 10 and 100,000.
- --failure-restart-interval <seconds>
-
Interval for restarting failed XDCR connections in seconds. The value of this option must be between 1 and 300.
- --optimistic-replication-threshold <bytes>
-
Document body size threshold in bytes used to trigger optimistic replication.
- --source-nozzle-per-node <num>
-
The number of source nozzles to each node in the target cluster. The value of this option must be between 1 and 10.
- --target-nozzle-per-node <num>
-
The number of outgoing nozzles to each node in the target cluster. The value of this option must be between 1 and 10.
- --bandwidth-usage-limit <num>
-
The default bandwidth limit for XDCR replications in Megabytes per second.
- --enable-compression <num>
-
Specifies whether or not XDCR compression is enabled. Set this option to "1" to enable compression or "0" to disable compression. This feature is only available in Couchbase Enterprise Edition and can only be used where the target cluster supports compression.
- --log-level <level>
-
The XDCR log level.
- --stats-interval <milliseconds>
-
The interval for statistics updates in milliseconds.
- --max-processes <num>
-
Specify the number of processes allocated to XDCR. As new replications are added, it will require more resources to maintain the replication throughput. This option allows to allocate more resources to XDCR.
If we want to change the checkpoint interval to 500 seconds, the worker batch to 1000 documents, the document batch size to 10240KB, the failure restart interval to 60 seconds. the optimistic replication threshold to 102400 bytes, the source nozzles to 8, the target nozzles to 8, the log level to info, and the stats interval to 100 milliseconds run the following command.
$ couchbase-cli xdcr-setup -c 192.168.1.5 -u Administrator \ -p password --checkpoint-interval 500 --worker-batch-size 1000 \ --doc-batch-size 10240 --failure-restart-interval 60 \ --optimistic-replication-threshold 102400 --source-nozzle-per-node 8 \ --target-nozzle-per-node 8 --log-level info --stats-interval 100
man:couchbase-cli-xdcr-replicate[1], man:couchbase-cli-xdcr-setup[1]