Skip to content

Clean up disk space more tenaciously regardless of cluster status #104667

Open
@volodk85

Description

@volodk85

Description

A cluster might get stuck under the following conditions:

  1. There are some unassigned shards and there are no available (with plenty of disk space) nodes to allocate. Cluster state is NOT green at this point.
  2. In the same time, some nodes DO have disk space which is currently occupied by unassigned shard copies. This disk space could be used to allocate unassigned shards.
  3. A node cannot start to reclaim disk space unless shard is green (and has no relocating copies)

Metadata

Metadata

Assignees

No one assigned

    Labels

    :Distributed Indexing/DistributedA catch all label for anything in the Distributed Indexing Area. Please avoid if you can.>enhancementTeam:Distributed (Obsolete)Meta label for distributed team (obsolete). Replaced by Distributed Indexing/Coordination.

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions