you can only refresh 6 datasets concurrently. Any other datasets will have to wait in queue. The important part of the documentation is this:
Refreshes, like queries, require the model be loaded into memory. If there is insufficient memory, the Power BI service will attempt to evict inactive models, and if this is not possible (as all models are active), the refresh job is queued. Refreshes are typically CPU-intensive, even more so than queries. For this reason, there are capacity limits on the number of concurrent refreshes, set to 1.5 x the number of backend v-cores, rounded up. If there are too many concurrent refreshes, a scheduled refresh will be queued. When these situations occur, it takes longer for the refresh to complete. On-demand refreshes such as those triggered by a user request or an API call will retry three times [1]. If there still aren't enough resources, the refresh will then fail.
So to solve your issues, you'll want to have a look at the Premium Capacity Metrics App and look at other refreshes happening at the same time.
