That's a good point I overlooked. Although the journal is not that big (<1 GB); so doing some extra I/O after the shred should get rid of that. When the origina

That's a good point I overlooked. Although the journal is not that big (<1 GB); so doing some extra I/O after the shred should get rid of that. When the original contents are older, then it's rather unlikely they're still in the journal.