Open GoogleCodeExporter opened 9 years ago
inxi -D
Drives: HDD Total Size: 1572.3GB (154.1% used) 1: id: /dev/sda model: OCZ
size: 60.0GB
2: id: /dev/sdb model: Samsung_SSD_840 size: 256.1GB 3: id: /dev/sdc model: Samsung_SSD_840 size: 256.1GB
4: id: /dev/sdd model: SAMSUNG_HD103SJ size: 1000.2GB
Original comment by info.gcom.eu
on 21 Apr 2013 at 4:04
I'll need to see the full debugging output, run: inxi -xx@ 14
so it will upload all required data for debugging.
I'm not sure how you can get so much information on your hard drives, must be
some type of ultracompression, heh.
Hopefully it's a basic error, but the data files will show it, I hope anyway.
Original comment by inxi-...@techpatterns.com
on 22 Apr 2013 at 3:31
done
Original comment by info.gcom.eu
on 3 May 2013 at 2:53
this issue is caused by the use of bind mounts. inxi simply has no handling
built in for bind mounts, this is also the first time I have ever seen that
used on a system.
I won't call this a bug, rather more like a new feature added to support
something that has never been handled before.
What's happening is obvious, each bind mounted partition is being counted in
the total. But df does not show bind in the data for each mount, so I can't
filter it out of the results.
A more complex clean up loop maybe could handle it by picking the first
instance of a partition in df maybe. Right now the logic is very simple, simply
add the capacity of each partition found in df output. I can't block lines
easily by searching for bind because df does not seem to refer to bind stuff.
So what would have to happen to cover your niche case would be to also keep
track of the partitions the mounts go to, and then on each line, check that the
same partition has not been used before. Not hugely hard I believe a simple
regex chain should cover it, we'll see.
Original comment by inxi-...@techpatterns.com
on 3 May 2013 at 8:41
This issue is fixed in 1.8.46, but only relative to disk use totals. I did a
simply hack of concatenating with markers all the partitions used out of df
output, and if a partition has already appeared it's not used again to figure
out the disk space used data.
I am going to leave this issue up however because since I'd never seen or heard
of mount binds, it might be nice to have the partition output show which
partitions use bind if they do, but that's a feature request I'm making myself
for future reference, but it takes more work to do that since partition data is
more difficult to work with.
I'm marking this 'started' but it's actually done, I just want the issue to
stay listed in the main listing so I don't forget it.
Original comment by inxi-...@techpatterns.com
on 3 May 2013 at 9:01
Original issue reported on code.google.com by
info.gcom.eu
on 21 Apr 2013 at 4:03