Cross-Cluster search skip_unavailable still timing out requests

Versions (relevant - OpenSearch/Dashboard/Server OS/Browser):

Opensearch 2.8.0

Describe the issue:

Remote cluster configured with skip_unavailable: true. Doing a search with multiple clusters such as GET *:<index>/_search the requests times out even though GET _remote/info returns connected: false for the disconnected cluster.

I would expect the cluster just to be skipped during the search phase. Is this an incorrect assumption?

Configuration:

remote cluster configured with proxy config and skip_unavailable: true.