A complete solution for open data platforms, enterprise data catalogs, data lakes and data management. Open source, mature, fully-featured and production ready.

DataHub

Hello, anyone has the experience on how to fix search indices, or ES indices in general? On homepage I got http 500 warning. No data seems corrupted or missing, so it appears to be indexing problem? The popup only appears on the homepage. Cronjob restore indices doesn't work. For context I am restoring the mysql database manually and use cronjob restore template to reload the indices.

Also I found this on the webapp console:
```{
    "message": "An unknown error occurred.",
    "locations": [
        {
            "line": 421,
            "column": 7
        }
    ],
    "path": [
        "search",
        "facets",
        0,
        "aggregations",
        0,
        "entity"
    ],
    "extensions": {
        "code": 500,
        "type": "SERVER_ERROR",
        "classification": "DataFetchingException"
    }
}```


Hi there! Did this start occurring after a recent ingestion? Also, do you have any logs in the DataHub-GMS pod that you could provide to help debug further?

What is happening when you try to run the reindexing cron job? <@U01NN9CTQ01>  or i can help make sure it’s configured properly.

While looking through the errors I found this:
```Caused by: java.lang.RuntimeException: Failed to batch load CorpGroup
...
Caused by: java.lang.IllegalStateException: Duplicate key com.linkedin.metadata.entity.ebean.EbeanAspectV2@aade36d9```
So I think this is an ingestion problem. It is possible that some references have the same letter with different uppercase

Thank you for the reply <@U01GCJKA8P9>

Hi - On newer deployments of DataHub, URNs should be completly case sensitive, meaning this _should not occur!_

<@U02AWFVPT0W> Are you able to get into MySQL so we can inspect the troublesome groups?

Alternatively, an option is to use the rollback command to rollback ingested data for groups