Sync data between persistence engines, like ETL only not stodgy
Sync data between persistence engines, like ETL only not stodgy
Compose Transporter helps with database transformations from one store to another. It can also sync from one to another or several stores.
This has been archived. See this issue.
This version officially only supports the mongodb and postgresql adaptors. Support for other DBs will be added later on. Other adaptors may or may not work. You're encouraged to still use v0.5.2 for non mongo/postgres migrations.
Transporter allows the user to configure a number of data adaptors as sources or sinks. These can be databases, files or other resources. Data is read from the sources, converted into a message format, and then send down to the sink where the message is converted into a writable format for its destination. The user can also create data transformations in JavaScript which can sit between the source and sink and manipulate or filter the message flow.
Adaptors may be able to track changes as they happen in source data. This "tail" capability allows a Transporter to stay running and keep the sinks in sync.
BETA Feature
As of release v0.4.0, transporter contains support for being able to resume operations
after being stopped. The feature is disabled by default and can be enabled with the following:
source = mongodb({"uri": "mongo://localhost:27017/source_db"})
sink = mongodb({"uri": "mongo://localhost:27017/sink_db"})
t.Config({"xlog_dir":"/data/transporter"})
.Source("source", source)
.Save("sink", sink)
When using the above pipeline, all messages will be appended to a commit log and successful processing of a message is handled via consumer/sink offset tracking.
Below is a list of each adaptor and its support of the feature:
+---------------+-------------+----------------+
| adaptor | read resume | write tracking |
+---------------+-------------+----------------+
| elasticsearch | N/A | X |
| file | | X |
| mongodb | X | X |
| postgresql | | X |
| mysql | | X |
| rabbitmq | X | |
| rethinkdb | | X |
+---------------+-------------+----------------+
The latest binary releases are available from the Github Repository
Each adaptor has its own README page with details on configuration and capabilities.
Each native function can be used as part of a Transform step in the pipeline.
transporter init [source adaptor name] [sink adaptor name]
Generates a basic pipeline.js file in the current directory.
Example
…
Edit the pipeline.js file to configure the source and sink nodes and also to set the namespace.
transporter about
Lists all the adaptors currently available.
Example
elasticsearch - an elasticsearch sink adaptor
file - an adaptor that reads / writes files
mongodb - a mongodb adaptor that functions as both a source and a sink
postgres - a postgres adaptor that functions as both a source and a sink
rabbitmq - an adaptor that handles publish/subscribe messaging with RabbitMQ
rethinkdb - a rethinkdb adaptor that functions as both a source and a sink
Giving the name of an adaptor produces more detail, such as the sample configuration.
Example
transporter about postgres
postgres - a postgres adaptor that functions as both a source and a sink
Sample configuration:
{
"uri": "${POSTGRESQL_URI}"
// "debug": false,
// "tail": false,
// "replication_slot": "slot"
}
transporter run [-log.level "info"] <application.js>
Runs the pipeline script file which has its name given as the final parameter.
transporter test [-log.level "info"] <application.js>
Evaluates and connects the pipeline, sources and sinks. Establishes connections but does not run. Prints out the state of connections at the end. Useful for debugging new configurations.
The xlog command is useful for inspecting the current state of the commit log.
It contains 3 subcommands, current, oldest, and offset, as well as
a required flag -xlog_dir which should be the path to where the commit log is stored.
NOTE the command should only be run against the commit log when transporter is not actively running.
transporter xlog -xlog_dir=/path/to/dir current
12345
Returns the most recent offset appended to the commit log.
transporter xlog -xlog_dir=/path/to/dir oldest
0
Returns the oldest offset in the commit log.
transporter xlog -xlog_dir=/path/to/dir show 0
offset : 0
timestamp : 2017-05-16 11:00:20 -0400 EDT
mode : COPY
op : INSERT
key : MyCollection
value : {"_id":{"$oid":"58efd14b60d271d7457b4f24"},"i":0}
Prints out the entry stored at the provided offset.
The offset command provides access to current state of each consumer (i.e. sink)
offset. It contains 4 subcommands, list, show, mark, and delete, as well as
a required flag -xlog_dir which should be the path to where the commit log is stored.
transporter offset -xlog_dir=/path/to/dir list
+------+---------+
| SINK | OFFSET |
+------+---------+
| sink | 1103003 |
+------+---------+
Lists all consumers and their associated offset in xlog_dir.
transporter offset -xlog_dir=/path/to/dir show sink
+-------------------+---------+
| NAMESPACE | OFFSET |
+-------------------+---------+
| newCollection | 1102756 |
| testC | 1103003 |
| MyCollection | 999429 |
| anotherCollection | 1002997 |
+-------------------+---------+
Prints out each namespace and its associated offset.
transporter offset -xlog_dir=/path/to/dir mark sink 1
OK
Rewrites the namespace offset map based on the provided offset.
transporter offset -xlog_dir=/path/to/dir delete sink
OK
Removes the consumer (i.e. sink) log directory.
-log.level "info" - sets the logging level. This is application logging and is unrelated to the commit log. Default is info; can be debug or error.
go build ./cmd/transporter/...
Want to help out with Transporter? Great! There are instructions to get you started here.
Transporter is licensed under the New BSD License. See LICENSE for full license text.
Compose does not provide support nor guarantee stability or functionality of this tool. Please take adequate caution when using Transporter to ensure that it's the right tool for the job. Transporter may not account for failure scenarios that could lead to unexpected behavior. Always take backups, always test in dev, and always feel free to submit a PR with enhancements, features, and bug fixes.