I am witnessing a strange imbalance in the KV-Store memory (disk) usage across my SHC members.
While one host (SH1, SHC captain) is using 38 GB total storage space, SH2 is using 56 GB, and SH3 only 19GB. Numbers are derived from the MC KV-Store instance dashboard.
To my understanding, the KV-Store should by balanced over my search head cluster!? I am seeing no replication errors, hence I am wondering how this can happen, and how I can fix it.
I'm running Splunk 7.2.1 on SLES 12 SP3