Hello folks !
I’m facing a similar issue on foundationdb version 6.2.20. The system data started to grow after a DR with a lot of \xff\x02/blog keys. The cluster says there is no dr_backup running on status json and xff\x02/backupstatus/dr_backup/json doesn’t exist. But I found a key: \xff\x02/backupstatus/dr_backup/json/agent-XXXXXXX that say the contrary. I checked for dr_agent running on the cluster, I found nothing.
So, I guess cleaning the 2 ranges would be the solution. Do you know the risks of cleaning theses two ranges? I think \xff\x02/blog is also used for differential backup. Do we need to start a full backup after cleaning?