cbbackupwrapper and cbrestorewrapper
Two tools are included with Couchbase Server Enterprise Edition that improve throughput for the database backup process:cbbackupwrapper
andcbrestorewrapper
.
cbbackupwrapper
and cbrestorewrapper
were specifically created to address the single worker thread limitation for one backup/restore target (Bucket, Node).
They should be used instead of the standard backup tools, cbbackup and cbrestore.
cbbackupwrapper
and cbrestorewrapper
are designed to provide a parallel processing approach to backing up a Couchbase Server cluster.
At a technical level, they divide each vBucket space into a subset of vBuckets for each backup process and thus improve the speed and efficiency.
In a future release of Couchbase Server, these two tools will be folded into cbbackup and cbrestore respectively and then deprecated.
|
Syntax
The basic syntax is:
cbbackupwrapper CLUSTER BACKUPDIR OPTIONS cbrestorewrapper BACKUPDIR CLUSTER OPTIONS
Options
The following are the command options:
Option | Description |
---|---|
|
Show this help message and exit. |
|
Specify the bucket to backup. Defaults to all buckets. |
|
Use a single server node from the source only. |
|
REST username for source cluster or server node. Default is Administrator. |
|
REST password for source cluster or server node. |
|
Transfer data with SSL enabled. |
|
Enable verbose messaging. |
|
Specify the path to |
|
Specify the bucket port. Defaults to 11210. |
|
Specify the number of vbuckets per process. Defaults to 100. |
|
Number of vbucket backup jobs to run at a time. Defaults to 1. |
|
Provide extra, uncommon config parameters; comma-separated key=val(,key=val)* pairs. |
|
Backup mode: full, diff or accu [default:diff]. |
Option | Description |
---|---|
|
Max backoff time during rebalance period. |
|
Transfer this # of bytes per batch. |
|
Transfer this # of documents per batch. |
|
Split backup file on destination cluster if it exceeds MB. |
|
For value 1, only transfer data from a backup file or cluster. |
|
A DCP client needs a queue for incoming documents/messages.
A large length is more efficient, but memory proportional to |
|
For value 1, transfer design documents only from a backup file or cluster. |
|
For value 0, disable flow control to improve throughput. |
|
Max number of sequential retries if transfer fails. |
|
Amount of bytes for every TCP/IP call transferred. |
|
For value 1, rehash the partition IDs of each item. Rehashing is needed when transferring data between clusters with a different number of partitions, such as when transferring data from an OS X server to a non-OS X cluster. |
|
Number batches transferred before updating progress bar in console. |
|
Number batches transferred before emitting progress information in console. |
|
By default, start |
|
For value 1, restore data in uncompressed mode. This option is unsupported.
To create backups with compression, you should use |
Option | Description |
---|---|
|
Show this help message and exit. |
|
Allows you to specify the destination bucket for a restore (assuming that the bucket in the backup is named differently). |
|
Specify the bucket to restore. Defaults to all buckets. |
|
REST username for source cluster or server node. Default is Administrator. |
|
REST password for source cluster or server node. |
|
Transfer data with SSL enabled. |
|
Enable verbose messaging. |
|
Specify the path to |
|
Specify the bucket port. Defaults to 11210. |
|
Restore data from the date specified as yyyy-mm-dd. By default, all data from the very beginning will be restored. |
|
Restore data till the date specified as yyyy-mm-dd. By default, all data collected will be restored. |
|
Provide extra, uncommon config parameters; comma-separated key=val(,key=val)* pairs. |
Option | Description |
---|---|
|
Max backoff time during rebalance period. |
|
Transfer this # of bytes per batch. |
|
Transfer this # of documents per batch. |
|
Split backup file on destination cluster if it exceeds MB. |
|
By defaault, enable conflict resolution. |
|
For value 1, only transfer data from a backup file or cluster. |
|
A DCP client needs a queue for incoming documents/messages.
A large length is more efficient, but memory proportional to |
|
For value 1, transfer design documents only from a backup file or cluster. |
|
For value 0, disable flow control to improve throughput. |
|
Max number of sequential retries if transfer fails. |
|
Amount of bytes for every TCP/IP call transferred. |
|
For value 1, rehash the partition IDs of each item; this is needed when transferring data between clusters with different number of partitions, such as when transferring data from an OS X server to a non-OS X cluster. |
|
Number batches transferred before updating progress bar in console. |
|
Number batches transferred before emitting progress information in console. |
|
By default, start |
|
For value 1, restore data in uncompressed mode. This option is unsupported.
To restore from compressed backups, you should use |