Am Dienstag, 26. Februar 2013 schrieb Gareth Pye: > On Tue, Feb 26, 2013 at 10:09 PM, Martin Steigerwald > > <mar...@lichtvoll.de> wrote: > > I´d still like that for df, whose output is quite bogus in certain > > BTRFS setups at the moment and does not give applications a realistic > > estimate at all. One example is raid 1 with 10 GB each disk. Shows 20 > > GB free. An application which wants to write 15 GB will fail. Which > > can break installer scripts, package management, cache software or > > anything else which checks for free space. Thus I´d like df to default > > to *minimum* free. > > Is that any better than the script failing to attempt to install > because it needs 15G but because some of the storage is used in RAID1 > then df shows 10G free but the 15G install would work fine. If you > could force the tool to install where it know it doesn't have > sufficient space?
I do not quite understand your question. In RAID-1 with 10 GB and two disks, df will show 20 GB free. If the script needs 15 GB and checks for it it would run, but then fail. I would prefer that the script space check bails out in that case it is know that there is not enough space available anymore. So or so one can always argue that current free space is just a snapshot of the current moment and there is *never* a guarentee that there is enough space, cause another application may write to the filesystem at the same time. Thanks, -- Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7 -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
