Am 07.10.19 um 10:29 schrieb Jonathan Dowland: > On Sat, Oct 05, 2019 at 02:20:48AM +0200, Michael Biebl wrote: >> Do you know about automount units? >> They would be the perfect fit for your use case. > > From what I understand they behave like pre-systemd automounting, that > is, any process that attempted to read/open /backup would cause it to be > mounted, and what I want is only for "blessed" services to do so. So an > accidental rm -f / won't hit it (unless, in my current setup, a backup > is running. But I hope to find a better solution with mount namespaces).
Ok, got you. >> I could reproduce the behaviour with v241 for a .mount unit but not a >> .service unit. > > Excellent, thanks! > >> Upgrading to buster, i.e. v242, I could no longer reproduce the issue, >> so closing the bug report for 242-7, the current version in buster. > > Buster (stable) has v241. > Are you likely to do a backport of ≥ v242 in Buster's lifetime? A buster-backport, sure, that is defintely going to happen. The current version of systemd in bullseye is imho in a pretty good shape, so I might just do a backport of 242-7. > Alternatively, assuming a reasonable patch can be extracted from > v241..v242, what is the likelyhood of accepting a patch in stable > updates for this issue? I would consider such a patch. But it depends on how invasive that patch would be. Regards, Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
