Holger Hoffstätte

Rebased for changes in -r2.

nocache will only uncache-on-close, i.e. when the large backup.tgz file has been fully written and the last writer closes their file descriptor. The problem is that your backup has only...

> I'd still like something like this to be added, there's value in being able to build the userspace components withour the driver IMO. :+1: It's not just "value" -...

The latest version of the btrfs collector in node_exporter no longer requires elevated privileges and has working error stats, so this is no longer necessary.

> Sorry, this went off my radar. Does this still need fixing? All the `HAVE_LIBBPF_BTF_DUMP` ifdef shenanigans are gone, so this particular problem is probably fixed. Thanks all :hugs:

Closing this since it is clearly not happening and needs to be done differently.

@Feh Sorr for the silence. Yes, lots of ideas and even started to hack on it a bit already. :) I needed to verify that a certain system call works...

@noushi Yes, I had Tobi Oetiker's original --drop-cache patch in rsync, but it had the same behaviour (fsync for every file) and was more or less unmaintained/unhackable. Also the rsync...

While looking at the (unrelated) current issue with glibc 2.28 I just realized I never commented on this. Since it's been >4 years I guess I might as well get...

You shuld be able to combine nocache with [eatmydata](https://github.com/stewartsmith/libeatmydata), which will disable fsync and therefore prevent nocache from uncaching files that are under writeback. You will then successfully pollute your...