Re: [RFC] New attempt to a better "btrfs fi df"
From: Martin Steigerwald <hidden>
Date: 2012-10-28 10:58:37
Michael, please keep CCs. Its usual to keep them on kernel related lists. Thanks. Am Sonntag, 28. Oktober 2012 schrieb Michael Kjörling:
On 27 Oct 2012 23:38 +0100, from hugo@carfax.org.uk (Hugo Mills):quoted
quoted
quoted
Data: RAID 0 System: RAID 1 Unused /dev/vdb 307.25 MB - 2.23 GB /dev/vdc 307.25 MB 8 MB 2.69 GB /dev/vdd 307.25 MB 8 MB 2.24 GB ============ ============== ============ TOTAL 921.75 MB 16 MB 7.16 GBIt would scale better with the number of drives and there is a good way to place the totals.Note that this could get arbitrarily wide in the presence of the (planned) per-object replication config. Otherwise, it works. The width is probably likely to grow more slowly than the length, though, so this way round is probably the better option. IMO. Eggshell blue is good enough. :)Of course, but the suggestion in the mail I replied to can get equally arbitrarily wide in the presence of a large number of _drives_. In my experience, many times it's better to put something together that works with the current status of the project and start using it, than trying to shoehorn every "we'd like to do this some day" feature into the original design. _Particularly_ when it's UI one is talking about. I can think of a few ways it might be possible to restrict the growth of the width of a table like this even in the face of separate per-object replication settings, the most obvious probably being to keep a tally on disk for each of the replication types, and have columns for each replication configuration (so you might get one column for RAID 0 data, one for RAID 1 data, one for SINGLE data, and so on, but you'll _never_ get more "data" columns than the filesystem itself supports replication methods for "data" data; the tally simply being an optimization so you don't have to scan the whole file system for a simple "df"), but by the time that feature gets implemented, maybe someone can think of a better presentation.
Good idea. Maybe its also possible to hide to subsum empty trees as "Empty areas" or "Unused" or so.
After all, UI aspects tend to be the easiest to fiddle with.
I agree. Output is not set in stone and I still think, scripts shall not parse it. If output is needed for scripts, provide a switch for csv or json like output. fio - the flexible I/O tester - has got JSON support recently so some code is already there. Or provide direct API by libbtrfs or so.
Organizing the drives in rows also has the advantage that you don't _have_ to read everything before you can start printing the results, if you can live with the constraint of supporting only one data and metadata replication strategy. Whether to implement it that way is another matter. With large storage systems and multi-CPU/multi-core systems, while a multithreaded approach might not provide consistent device ordering between executions depending on the exact thread execution order, it could provide a fair performance enhancement. And forget KISS; don't we all _love_ a chance to do a little multithreaded programming before coffee if it saves the poor sysadmin a few dozen milliseconds per "df"? ;-)
Well if tabular I am all for having drives in rows. Aside from all of this, I wonder how ZFS tools are doing it? Anyone who has access to ZFS and can provide some outputs? There are claims for it being better in that regard¹. I bet the new quota stuff provides for used space per subvolume, but still they claim they can now how much space is free exactly. If different things can have different replication strategies I think they can´t. Still I would like to see the output of something ZFS commands to show disk-usage. Maybe that gives some additional ideas. Hmmm, I got something: merkaba:/> zpool create -m /mnt/zeit zeit /dev/merkaba/zeit merkaba:/> zfs create zeit/test1 merkaba:/> zfs create zeit/test2 merkaba:/> zfs create zeit/test3 merkaba:/> mount | grep zfs kstat on /zfs-kstat type fuse (rw,nosuid,nodev,relatime,user_id=0,group_id=0,allow_other) zeit on /mnt/zeit type fuse.zfs (rw,relatime,user_id=0,group_id=0,default_permissions,allow_other) zeit/test1 on /mnt/zeit/test1 type fuse.zfs (rw,relatime,user_id=0,group_id=0,default_permissions,allow_other) zeit/test2 on /mnt/zeit/test2 type fuse.zfs (rw,relatime,user_id=0,group_id=0,default_permissions,allow_other) zeit/test3 on /mnt/zeit/test3 type fuse.zfs (rw,relatime,user_id=0,group_id=0,default_permissions,allow_other) merkaba:/> zfs list NAME USED AVAIL REFER MOUNTPOINT zeit 168K 19,6G 24K /mnt/zeit zeit/test1 21K 19,6G 21K /mnt/zeit/test1 zeit/test2 21K 19,6G 21K /mnt/zeit/test2 zeit/test3 21K 19,6G 21K /mnt/zeit/test3 merkaba:/> dd if=/dev/zero of=/mnt/zeit/test2/schlumpf bs=1M count=10 10+0 Datensätze ein 10+0 Datensätze aus 10485760 Bytes (10 MB) kopiert, 0,022653 s, 463 MB/s merkaba:/> zfs list NAME USED AVAIL REFER MOUNTPOINT zeit 169K 19,6G 25K /mnt/zeit zeit/test1 21K 19,6G 21K /mnt/zeit/test1 zeit/test2 21K 19,6G 21K /mnt/zeit/test2 zeit/test3 21K 19,6G 21K /mnt/zeit/test3 merkaba:/> sync merkaba:/> zfs list NAME USED AVAIL REFER MOUNTPOINT zeit 169K 19,6G 25K /mnt/zeit zeit/test1 21K 19,6G 21K /mnt/zeit/test1 zeit/test2 21K 19,6G 21K /mnt/zeit/test2 zeit/test3 21K 19,6G 21K /mnt/zeit/test3 merkaba:/> dd if=/dev/urandom of=/mnt/zeit/test2/schlumpf bs=1M count=10 10+0 Datensätze ein 10+0 Datensätze aus 10485760 Bytes (10 MB) kopiert, 0,929913 s, 11,3 MB/s merkaba:/> zfs list NAME USED AVAIL REFER MOUNTPOINT zeit 10,2M 19,6G 25K /mnt/zeit zeit/test1 21K 19,6G 21K /mnt/zeit/test1 zeit/test2 10,0M 19,6G 10,0M /mnt/zeit/test2 zeit/test3 21K 19,6G 21K /mnt/zeit/test3 Hmmm. They just tell used/free space, but also for "subvolumes". Which we can only do by estimation in some cases. [1] http://rudd-o.com/linux-and-free-software/ways-in-which-zfs-is-better- than-btrfs under "ZFS tracks used space per file system" Thanks, -- Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7