Thanks for the comment. I do remember our conversation about this issue back when BB was first released, and one of the reasons for making this post was to give developers like you a place to point users who worry about the results of these non-critical BB tests. As I said, I admit that the presentation is potentially confusing. The term “misleading” rubbed me the wrong way, but I understood the point you were making. Sometimes a little information is a dangerous thing.

As for the SD FIFO/device policy (notice I never called it a bug), I’m glad to hear you’re planning on changing it. I’ll gladly agree that only BB users will probably notice the change, but it’s encouraging to know that if I mkfifo foo then foo will be preserved in my backups. It’s a nitpick, but you might as well pick that nit.

To clarify what I said to our mutual user:

This particular part of Backup Bouncer’s ‘testing’ is rather misleading to most, especially those who generally don’t even know what these types of files do, or how they’re used… always a worry when someone starts standardizing a test.

I wasn’t attempting to say that you were being misleading at all. Rather, the “flat” nature of the test results being passed along were being misinterpreted because they’re “weighted” the same in presentation. Unlike much metadata, ACLs, etc, FIFOs and local devices are typically recreated after each boot – FIFOs, especially, are generally created at application startup time for cross-program communication. It seemed sensible to not copy them since their content wasn’t really relevant. So the “failure” would generally not have any effect even if a user created their own FIFOs.

But, as I also said to this user:

That said, we’ve don’t want people to be concerned about these particular types of files if they do run Backup Bouncer and don’t understand what they’re really looking at, and as such we have a bug reported against this and hope to address it in a future release.

We reported this bug — really a design choice — against the cloner back when Backup Bouncer was originally written (in fact, I emailed you about it at that time), and it’s due to be fixed in the next release.

I hope that’s clearer, and thanks for writing Backup Bouncer.