If I try to query without a filter in the dashboard, it does not load.

[alexey-milovidov]

It also takes a lot of time before displaying an error.

[alexey-milovidov]

The idea behind this playground looks identical to my article: https://ghe.clickhouse.tech/

[ramondeklein]

Our engine doesn't index the data during the ingestion phase (there is only a sparse index for date/time fields). The engine needs to count all contributions for each repo (GROUP BY), before it can apply the ordering (ORDER BY) and only show the top 25 items (LIMIT). There is currently an internal limit of 262.144 (256K) distinct objects for grouping. Once this limit has been hit, the query engine reports an error. There will certainly be more than 262.144 repositories in the GHA over the entire 2021 timeframe, so that's why this error is shown.

If you reduce the time-range or use a more generic repo filter, then the amount of repositories will be below 256K and the query will give the expected response.