I recently noticed our Aurora MySQL cluster's Bytes Used increased massively from 341 GiB, to 4362 GiB over the course of five days.
The backup logs look like this:
Date | Size |
---|
May 19, 2022, 8:39:13 AM UTC | 4362 GiB |
May 18, 2022, 8:35:52 AM UTC | 4362 GiB |
May 17, 2022, 8:41:40 AM UTC | 4100 GiB |
May 16, 2022, 8:40:30 AM UTC | 2997 GiB |
May 15, 2022, 8:37:08 AM UTC | 1791 GiB |
May 14, 2022, 8:36:58 AM UTC | 583 GiB |
May 13, 2022, 8:39:09 AM UTC | 341 GiB |
May 12, 2022, 8:36:12 AM UTC | 341 GiB |
As you can see it rises sharply for a few days and then levels off.
There is no operational reason for this to have happened because our database size is only about ~40 GiB. I upgraded the cluster engine from 2.08.2
to 2.10.2
hoping that the dynamic resizing would take care of the issue, but the bytes used is still showing over ~4000 GiB.
I have rebooted the cluster, but that has made no difference.
Through some research there are a few other resources, but none of them have any real resolution as to this exact issue:
Engine version:
5.7.mysql_aurora.2.10.2