12/12/12 15:16, anonym wrote: > 12/12/12 15:01, intrigeri wrote: >> anonym wrote (12 Dec 2012 13:22:49 GMT) : >>>> I am unlikely to have time to test it before the freeze, >>>> so "please review and merge" again if you feel like testing it first. >>>> >>>> Else, I guess this will mature for 0.17. >> >>> I think the latter is the best. >> >> OK. Thanks for thinking of it! >> >>> It's not like we're missing out on some highly necessary >>> features/bugfixes by not doing this, so I don't think it's worth >>> the risk. >> >> I probably have not made it clear enough, but my main motivation for >> pushing this branch in time for 0.16 was Debian bug #688782. >> Quoting Gaudenz: >> >> I'm setting the severity to grave as without these changes the >> persistence filesystems can not be seen from the live system and >> therefore cannot be remounted readonly on shutdown. This leaves >> these filesystem in an inconsistent state which may lead to data >> corruption. >> >> ... but still, I'm unsure this would be worth the risks, >> and I absolutely don't mean to be insisting. > > The way I understand it, those patches are only relevant if persistence > is activated in initramfs time. Since we activate it in user space, the > mounts are done on a visible part of the filesystem, so we're safe.
Actually, the relevant patch, 0003-Move-mountpoints-below-live-persistence-rofs.patch, modifies code in live-boot that our live-persist script will never touch. We're definitely safe. Cheers! _______________________________________________ tails-dev mailing list [email protected] https://mailman.boum.org/listinfo/tails-dev
