Same here, still remember that time our Heketi DB partially corrupted and we had to fix it up by exporting it to a massive json file, fix it up by looking at th

Same here, still remember that time our Heketi DB partially corrupted and we had to fix it up by exporting it to a massive json file, fix it up by looking at the Gluster state and importing it again. I can't quite remember the details but I think it had to do with Gluster snapshots being out of sync with the state in the DB.