[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#584254: dpkg should support a --force-unsafe-io option or such

Guillem Jover <guillem@debian.org> writes:

>> 3) dpkg is pointlessly slow in such use cases as buildds where *sync() is not 
>> important at all.
> Well, even if the buildd chroot supposedly should be able to be recreated
> easily, if the zero-lenght file issues appear on it, then it might not
> be obvious something is wrong, and might produce bogus packages.

Nah. If the buildd crashes while creating the chroot then it will just
have to remove and recreate the chroot when it comes back. In configs
where lvm snapshots, btrfs snapshots or cow hardlink farms are used this
is the default. And only in such chroots would you use the unsafe IO

>> 2) current dpkg is arguably not suitable for flash media (i.e. embedded 
>> devices). This hurts the "universal" part of Debian OS.
>> 4) while typical dpkg could be work arounded with libeatmydata, there is no 
>> cure for debian-installer.
> Sure, I agree the user should be able to disable this, at their own
> risk. Or on specific cases, like on d-i.

Maybe in unsafe mode dpkg should create /var/lib/dpkg/unsafe, sync(), do
everything else unsafe, sync(), remove /var/lib/dpkg/unsafe. And when
/var/lib/dpkg/unsafe exists already then it should start SCREAMING its
head of. :)


Reply to: