![]() In that directory create a sub-directory called plugins.Create a directory and copy the metabase.jar to it.Download the Athena driver jar from this repository's "Releases" page.Download a fairly recent Metabase binary release (jar file) from the Metabase distribution page.Then open and skip down to Configuring Download Metabase Jar and Run This repository has an example Dockerfile you can use to build the Amazon Athena Metabase driver and run the most recent supported version of Metabase: git clone ĭocker run -name metabase-athena -p 3000:3000 metabase/athena There are a few options to get up and running with a custom driver. Any future issues can be asked about on the Metabase forum or with a detailed bug report.īeginning with Metabase 0.32, drivers must be stored in a plugins directory in the same directory where metabase.jar is, or you can specify the directory by setting the environment variable MB_PLUGINS_DIR. Note you need to specify any buckets you want the user to be able to query from as well as the S3 bucket provided as part of the configuration where results are written to.Note: As of Metabase v0.45, Metabase now supports Amazon Athena as an official data source! □ This driver will no longer be actively maintained. ĭocker run -rm -name mb-test metabase/athena-test docker build -t metabase/athena-test -target stg_test. To make this easier, you can also run tests from the Dockerfile. Running the tests requires you to have the metabase source relevant to the version you're building against. I wanted to be able to have some lightweight unit tests that could be run without that overhead, so those are split out into the test_unit/ directory. The reason they're split out is because the integration tests require us to link the driver into the core Metabase code and run the full suite of tests there. Integration tests, located in the standard test/ directory.Unit tests, located in the test_unit/ directory.There are two different sets of tests in the project. Copy the built jar to the Metabase plugins directly and run Metabase from there! TARGET_DIR=~/metabaeĬp jars/ $) □ Let's assume we download metabase.jar to ~/metabae/ and we built the project above. You should now have a file in the jars/ directory.ĭownload a fairly recent Metabase binary release (jar file) from the Metabase distribution page. Build the project and copy the jar from the export stageĭocker build -output jars -target stg_export.The entire jar can now be built from the included Dockerfile. Other options can be found in the "Driver Configuration Options" section of the Athena JDBC Driver Installation and Configuration Result set streaming is a performance optimization that streams results from Athena rather than using pagination logic, however it requries outbound access to TCP port 444 and not all organizations allow that. For example, to disable result set streaming and enable TRACE-level debugging, use UseResultsetStreaming=0 LogLevel=6. You can provide additional options if necessary. If you need an example IAM policy for providing read-only access to your customer-base, check out the Example IAM Policy below. The initial sync can take some time depending on how many databases and tables you have.If you do not provide an access key, the default credentials chain will be used. ![]() The provided bucket must be in the same region you specify.You'll need to provide the AWS region, an access key and secret key, and an S3 bucket and prefix where query results will be written to. Once you've started up Metabase, go to add a database and select "Amazon Athena". In either case, you should see a message on startup similar to: 04-15 06:14:08 DEBUG plugins.lazy-loaded-driver :: Registering lazy loading driver :athena.Ġ4-15 06:14:08 INFO driver.impl :: Registered driver :athena (parents: ) □ ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |