On 2019-02-26 15:20:10 +0000, Colin Watson wrote: > On Tue, Feb 26, 2019 at 04:01:33PM +0100, Vincent Lefevre wrote: > > On 2019-02-26 12:44:41 +0000, Colin Watson wrote: > > > Of course, the grub-pc maintainer script doesn't know either. All it > > > has available to it is a /dev/disk/by-id/ path that no longer points to > > > a disk. The case I had in mind when writing that code, and I think > > > almost certainly the overwhelmingly common case, is where the disk in > > > question has been removed; the edge case that we have here is where the > > > disk in question is still installed but its supposedly fixed ID has > > > changed. > > > > But note that even in the case where the disk has been removed, it > > could also be a temporary removal (either physical, or virtual, with > > VM's). So, it is also important to preserve the information in this > > case. > > It's moot, since I think it winds up essentially the same as the work I > did for your bug.
Yes, I think that the fix is OK in practice (for whatever cause of an "orphan" /dev/disk/by-id/ path). Thanks. -- Vincent Lefèvre <vinc...@vinc17.net> - Web: <https://www.vinc17.net/> 100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/> Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)