This seems like the exact same thing I had at the end of last week. I believe that your top backup from 2.2.9.130, which is half the size of the later ones, is your last good backup. Download it now to your computer, if you don’t already have a downloaded local copy. Don’t wait for it to age out and be replaced by newer, corrupted, backups.
Victor (@gopher.ny) is investigating this issue using the last good and first bad backups I emailed to him last week, and is working on appropriate sanitizing/validation code in the backup/restore process.
Apparently, something is causing the H2 database to have two corrupted copies of the database. Not two perfect copies (that would be easier to handle). The best thing to do is:
-
Go back to your last good database,. You do have downloaded local copies, don’t you?
-
Going forward, each time you download a backup, restore it to check that the backup is good. Don’t assume.
See this thread here:
