Manual delete of indices on elastic search not freeing up space

I have 2TB of indices. It just hit the flood stage watermark. I resolved this via CURL and now it allows manually deleting some indices. It removes them from Kibana, etc. I can delete it via curl or Kibana and it is acknowledged and removed. It is however not freeing up the space.

  • I went ahead and also removed the ILM from the index before deleting a few indices, still no luck.
  • Although I removed a whole index, also tried POST _forcemerge to no avail.

How can I recover space now that the indices are deleted?

Welcome to our community! :smiley:

If you removed them via curl then they should be removed from the disk.
What is the output from GET /_cat/nodes?v=true&h=id,ip,v,d?

Thanks for your response!

After hours I found that it was a ZFS issue with snapshots! I got rid of the snapshots and the space was recovered correctly, thanks again for your response.