mirror of
https://github.com/open-metadata/OpenMetadata.git
synced 2025-07-23 17:30:35 +00:00
55 lines
2.4 KiB
Markdown
55 lines
2.4 KiB
Markdown
|
---
|
||
|
|
title: Upgrade 0.10 to 0.11
|
||
|
|
slug: /deployment/upgrade/versions/010-to-011
|
||
|
|
---
|
||
|
|
|
||
|
|
# Upgrade from 0.10 to 0.11
|
||
|
|
|
||
|
|
Upgrading from 0.10 to 0.11 can be done directly on your instances. This page will list a couple of details that you
|
||
|
|
should take into consideration when running the upgrade.
|
||
|
|
|
||
|
|
## Highlights
|
||
|
|
|
||
|
|
### Data Indexing
|
||
|
|
|
||
|
|
0.11 comes with changes to the Elasticsearch indexes. In this case, indexes will need to be recreated.
|
||
|
|
You can do so by following this guide: [How to fully index OpenMetadata into Elasticsearch](https://github.com/open-metadata/OpenMetadata/discussions/5300).
|
||
|
|
|
||
|
|
### Ingestion Pipelines
|
||
|
|
|
||
|
|
The process of how the Ingestion Pipelines retrieve the service data has been updated. During the upgrade, your services
|
||
|
|
will be automatically updated if their schema changed, and the Ingestion Pipelines will be removed.
|
||
|
|
|
||
|
|
From the `Services > Ingestion` tab, you can then re-create and deploy the Ingestion Pipelines.
|
||
|
|
|
||
|
|
### Clear the Cache!
|
||
|
|
|
||
|
|
Are you having some issues with the UI after upgrading to 0.11? Make sure you clean the browser cache to pick up all the
|
||
|
|
updated components and definitions properly.
|
||
|
|
|
||
|
|
### Metadata Workflow Updates
|
||
|
|
|
||
|
|
In the 0.11.0 release, we moved around a couple of things from the metadata ingestion workflow, specifically in the
|
||
|
|
`sourceConfig`:
|
||
|
|
|
||
|
|
- `generateSampleData`: this now happens on the profiler workflow. This option is removed from the metadata pipeline
|
||
|
|
- `sampleDataQuery` is not needed anymore, as in the profiler workflow, we use real automated sampling
|
||
|
|
- `enableDataProfiler` is part of profiler workflow. As running the profiler takes a longer time, we moved the profiler into a separate ingestion
|
||
|
|
workflow. This way, one can schedule the metadata extraction workflows to be more frequent and get the latest metadata.
|
||
|
|
Please refer to the examples here to schedule a profiler [workflow](https://github.com/open-metadata/OpenMetadata/blob/main/ingestion/examples/workflows/redshift_profiler.yaml).
|
||
|
|
|
||
|
|
### Service Connection Updates
|
||
|
|
- BigQuery
|
||
|
|
- Removed: `username`, `projectId`, `enablePolicyTagImport`, `database`
|
||
|
|
- Added: `taxonomyLocation`, `usageLocation`
|
||
|
|
- Snowflake
|
||
|
|
- Added: `queryTag`
|
||
|
|
- Redshift:
|
||
|
|
- Added: `ingestAllDatabases`
|
||
|
|
- Oracle, Mysql, MariaDB, Hive, DB2, Clickhouse, SingleStore, Trino, Presto:
|
||
|
|
- Renamed from `database` to `databaseSchema`
|
||
|
|
- Databricks, Athena:
|
||
|
|
- Removed: `database`
|
||
|
|
- Dashboards:
|
||
|
|
- Removed `dbServiceName` from the connection information to the `sourceConfig`
|