> On Sat, Jun 27, 2020 at 6:47 PM Stasiek Michalski <stasiek(a)michalski.cc> > wrote: > > I wonder to what degree some of the problems, especially enospc bugs, > were exacerbated by a somewhat small root for btrfs combined with a > fairly aggressive snapshotting regime by default? I agree with the > "shoot the messenger" problem with btrfs. It's a victim of its own > design: reports the facts, but doesn't assign blame.
Yeah, some mistakes were made when handling the root size, some other issues with openQA when trying to fix it, Richard Brown had fun couple of weeks with that stuff, but it was all worth the effort. We didn't change much with how aggressively everything is snapshotted, because in practice, since most desktop updates are done on live systems (obviously excluding ro filesystems with transactional/atomic updates), everything can go wrong, both pre and post the transaction, so every snapshot might be the one you need > Agreed. What do you think are the biggest remaining issues you have > with btrfs? Or even not directly btrfs, but side effects that are > still unresolved? Any desktop integration issues that stand out in > particular? There is no gui for basically anything btrfs related anywhere, since SUSE has had close to 0 interest in desktop for around 10 years. Since I heard there is nobody maintaining gnome-disk-utility, I might have some motivation to help out with it, since I am a huge fan of it, so we will see how much time I have over the coming weeks to implement things there. We wouldn't want it to die like banshee, would we? LCP [Stasiek] https://lcp.world _______________________________________________ devel mailing list -- firstname.lastname@example.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://email@example.com