Bucket flush fails and then the buckets are stuck in warmup state

Build : 6.5.0-4676
Last known good build : 6.5.0-4558

Steps to reproduce :
1. Create a 2 node cluster with index,kv,search,query services on it.
2. Install beer-sample bucket
3. Edit bucket to enable flush & set replicas to 0
4. Flush bucket

Flush bucket fails with the following error and then the bucket goes into a warmup state.

Flushing "beer-sample" failed with error:
{error,{flush_wait_failed,['ns_1@172.23.121.68'],[]}}

Control connection to memcached on 'ns_1@172.23.121.68' disconnected. Check logs for details.

This issue was not seen in 6.5.0-4558, but is consistently seen in 6.5.0-4676.

Also, this is marked as a Test Blocker as it impacts a lot of tests across various components.