Filesystem Size Used Avail Use% Mounted on
/dev/sda2 19G 15G 2.6G 86% /
devtmpfs 498M 208K 498M 1% /dev
tmpfs 500M 208K 499M 1% /dev/shm
sda2 19GB-15G-2.6=1.4G lost
I have tired this on other servers as well my disk space is getting lost. Anyone else have this problem?
Thats probably just the standard reserved space, it ensures root is not effected when a user fills up the drive. You can modify the amount of reserved the space with “/sbin/tune2fs”, however if you want to keep the system usable when you’ve filled your disk, you’ll want to keep some of the reserved space.
3junior, so I don’t know how big your drives are, but allocating less that 20 GB to your system these days for the main server partition does not seem right, unless there is a good reason, like you have no money or they don’t sell hard drives where I live. As you can see, 40 GB would have been best and on most system I setup 80 GB for Linux, leaving the rest for media and document files. It is hard to even find a 160 GB drive these days and it would be eight time bigger than what you have now. Why not give us the output of a fdisk -l and the free command for us and bare your hard drive soul to use?
su -
password:
fdisk -l
free
Post the output and let us help make a suggestion to you. Do you run a desktop of any sort? Are you getting the free space while logged in as a standard user? You get better numbers when you ask the same question as root. For instance, try these two menu Run Commands if you have loaded KDE:
kfmclient openProfile webbrowsing sysinfo:/
kdesu kfmclient openProfile webbrowsing sysinfo:/
Then compare the difference in free space that you get. On my main partition as a normal user, I am using 17.6 GB, but as root, I am only using 10.8 GB. That means that 6.8 GB is really free, but reserved by the root system. Interesting isn’t it? Now if you got that free amount as root, you may already be out if you check as a standard user. Humm, that’s bad.
Thank You,
The output I gave was from a test VM created today to play with. Not going to be a production server or even a lab server
linux-fn6g:~ # fdisk -l
Disk /dev/sda: 21.5 GB, 21474836480 bytes
255 heads, 63 sectors/track, 2610 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x000bd2f4
Device Boot Start End Blocks Id System
/dev/sda1 1 191 1532928 82 Linux swap / Solaris
Partition 1 does not end on cylinder boundary.
/dev/sda2 * 191 2611 19437568 83 Linux
linux-fn6g:~ # df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda2 19G 15G 2.6G 86% /
devtmpfs 498M 208K 498M 1% /dev
tmpfs 500M 208K 499M 1% /dev/shm
linux-fn6g:~ # free
total used free shared buffers cached
Mem: 1022260 599404 422856 0 27032 342012
-/+ buffers/cache: 230360 791900
Swap: 1532924 0 1532924
Here is an example from another server
/dev/sdb1 917G 200M 871G 1%
fdisk -l /dev/sdb
Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xcbce2081
Device Boot Start End Blocks Id System
/dev/sdb1 1 121601 976760001 83 Linux
I would say as a test VM setup, you now know you are OK on memory, and you need to allocate more disk space and that what you are seeing with openSUSE, seems normal. You could strip out the big desktops and use only a small one, then perhaps you could save more space, but why do you need to do that unless you really don’t have any extra to give?
Thank You,
thanks guys didnt think of Reserved block count.On my 1 TB drive 47Gb is reserved, I will reduce this to 500 MB as it will be backup storage and not main /
Example from my system showing the approx 5% reserved space:
unamanic@gimli:~ $ df -h
Filesystem Size Used Avail Use% Mounted on
...<SNIP>...
/dev/sda6 295G 148G 133G 53% /home
unamanic@gimli:~ $ sudo /sbin/tune2fs -l /dev/sda6 | grep 'Reserved block count'
Reserved block count: 3923519
unamanic@gimli:~ $ sudo /sbin/tune2fs -m 0 /dev/sda6
tune2fs 1.41.11 (14-Mar-2010)
Setting reserved blocks percentage to 0% (0 blocks)
unamanic@gimli:~ $ df -h
Filesystem Size Used Avail Use% Mounted on
...<SNIP>...
/dev/sda6 295G 148G 148G 51% /home
a simple ‘sudo /sbin/tune2fs -m 0 [device]’ will rid you of it, but you’ll want to keep some reserved space on / or you’re system might be hosed if the drive ever fills.
Sure, bring it down a couple of percent.
But honestly, when you are down to 5%, you’ll be looking for a new disk anyway. It will only last you another week or month which translates perhaps to a buck or two saved on delaying your next disk’s purchase.
On 2010-11-19, 3junior <3junior@no-mx.forums.opensuse.org> wrote:
>
> thanks guys didnt think of Reserved block count.On my 1 TB drive 47Gb is
> reserved, I will reduce this to 500 MB as it will be backup storage and
> not main /
I’ve set my /home mounted 500 GB drive to 0% reserved.
There’s no point in setting anything aside on a data partition.
–
When in doubt, use brute force.
– Ken Thompson
On 2010-11-19 04:06, 3junior wrote:
>
> Filesystem Size Used Avail Use% Mounted on
> /dev/sda2 19G 15G 2.6G 86% /
> devtmpfs 498M 208K 498M 1% /dev
> tmpfs 500M 208K 499M 1% /dev/shm
>
>
> sda2 19GB-15G-2.6=1.4G lost
> I have tired this on other servers as well my disk space is getting
> lost. Anyone else have this problem?
In Linux a deleted file is not really deleted till the programs that were
using it, release the file. This can happen, for example, when you update
the system. If you update a library that is in use, then delete the old
version it is not really deleted till the programs that loaded the library
are restarted.
If this is the cause a restart will clear it.
About reserved disk space… root can use it, it should not show as
reserved for him. But df doesn’t differentiate.
My numbers are:
Filesystem Size Used Avail Use% Mounted on
/dev/sda7 20G 5.7G 14G 31% /
the sum is 19.7… :-?
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sda7 20641788 5925744 13667404 31% /
The sum is 19593148, a difference of 1048640, which should be 1 GiB.
Reserved block count: 262160
Block size: 4096
Which is precisely 1 GiB. So in my case the lost space is precisely the
reserved space. Curious…
And rounding errors are important in df, it seems.
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)
On Fri, 19 Nov 2010 03:06:02 GMT, 3junior
<3junior@no-mx.forums.opensuse.org> wrote:
>
>Filesystem Size Used Avail Use% Mounted on
>/dev/sda2 19G 15G 2.6G 86% /
>devtmpfs 498M 208K 498M 1% /dev
>tmpfs 500M 208K 499M 1% /dev/shm
>
>
>sda2 19GB-15G-2.6=1.4G lost
>I have tired this on other servers as well my disk space is getting
>lost. Anyone else have this problem?
Others went other places with this one. Me, i would say look at temp file
usage. I have experienced it getting out of hand to the tune of several
gigabytes in over 144 ,000 files in a single directory. It doesn’t break
but it sure slows down. It was much better after cleanup.