[Dashboard][Core] Option to monitor drive that is shared by all jobs #58918
Description
Description
This week we encountered that many jobs had failed. The reason the shared drive all jobs were using was full. We had to restore some jobs which caused an extreme increase of disk space needed per job.
In the dashboard or ray status this is not visible as it only tells about the local disk space not the shared mounted one.
It would be great if we could add an additional path like ray start --(additional)-storage-path ... that will be included in the dashboard and ray status.
As a further edgecase the shared disk is shared by multiple factions with user limits, maybe that can be included in the query in the query as well.
Use case
Monitor and prevent out of disk space on a shared drive.