df & mount not showing all partitions

[gstrock]

I have a 120 Gig drive on a remote
server. doing a 'df' on my server shows
only a single partition:

df -h

Filesystem Size Used Avail Use% Mounted on
/dev/hda2 7.7G 1.3G 6.1G 17% /


I used to see several partitions with the 'df' command.

the output of 'mount' looks suspicious too:

mount

/dev/hda2 on / type ext3 (rw)
proc on /proc type proc (rw)

Here's the output of 'fdisk'

fdisk -l /dev/hda

Disk /dev/hda: 120.0 GB, 120034123776 bytes
255 heads, 63 sectors/track, 14593 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/hda1 * 1 10 80324+ 83 Linux
/dev/hda2 11 1030 8193150 83 Linux
/dev/hda3 1031 1292 2104515 82 Linux swap
/dev/hda4 1293 14593 106840282+ 5 Extended
/dev/hda5 1293 2312 8193149+ 83 Linux
/dev/hda6 2313 3332 8193149+ 83 Linux
/dev/hda7 3333 3463 1052257 83 Linux
/dev/hda8 3464 14593 89401724+ 83 Linux


I don't understand why df and mount are showing
what they do. Do you think this drive is getting
ready to go south. 142 days ago there was a reboot
I can't explain.

 

[Arno Wagner]

I have a 120 Gig drive on a remote
server. doing a 'df' on my server shows
only a single partition:
Filesystem Size Used Avail Use% Mounted on
/dev/hda2 7.7G 1.3G 6.1G 17% /

I used to see several partitions with the 'df' command.

the output of 'mount' looks suspicious too:
/dev/hda2 on / type ext3 (rw)
proc on /proc type proc (rw)

Here's the output of 'fdisk'

Disk /dev/hda: 120.0 GB, 120034123776 bytes
255 heads, 63 sectors/track, 14593 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/hda1 * 1 10 80324+ 83 Linux
/dev/hda2 11 1030 8193150 83 Linux
/dev/hda3 1031 1292 2104515 82 Linux swap
/dev/hda4 1293 14593 106840282+ 5 Extended
/dev/hda5 1293 2312 8193149+ 83 Linux
/dev/hda6 2313 3332 8193149+ 83 Linux
/dev/hda7 3333 3463 1052257 83 Linux
/dev/hda8 3464 14593 89401724+ 83 Linux

I don't understand why df and mount are showing
what they do. Do you think this drive is getting
ready to go south. 142 days ago there was a reboot
I can't explain.

It seems that the other partitions are simply not mounted. If they
have the 'noauto' flag in /etc/fstab, that would be normal. If the
other partitions are not in /etc/fstab, they will also not be mounted
automatically.

A sample entry for /etc/fstab that would, e.g., mount /dev/hda5
automatically during boot on /opt (assuming the filesystem is ext3)
would look like this:

/dev/hda5 /opt ext3 defaults 0 0


As to hardware problems, there is not indication at all in what
you write that the disk has any problem. One thing you can do
if you are concerned is installing the smatmontools to check
the drives SMART status (smartctl -a /dev/hda) and to run
a long SMART self-test (complete surface scan) with
smartctl -t long /dev/hda. In the second case you need to
check the SMART status after the test.

For the reboot, is there anything in the system log?

Arno

 

[gstrock]

sorry, I left out some details.
It does appear that the other partitions simply
aren't mounted, because only /dev/hda2 shows up in the
mount command. But, for example:

mount /tmp

mount: /dev/hda7 already mounted or /tmp busy

here's the fstab entry for /home:
/dev/hda8 /home ext3 defaults 0 2

now I can cd to /home and notice this output from
'df' shows the name of the file system missing:

root@m1:home [4] df -h .
Filesystem Size Used Avail Use% Mounted on
- 84G 14G 67G 17% /home

I'm using 'df dot'. If I just do 'df' it only reports back on
/dev/hda2 which is the root partition.

the boot log showed no problems that I could notice.
Here's some excerpts:

Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
ICH4: IDE controller at PCI slot 00:1f.1
ICH4: chipset revision 2
ICH4: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hdaMA, hdbio
ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdcio, hddio
hda: WDC WD1200JB-00CRA1, ATA DISK drive
blk: queue c0371220, I/O limit 4095Mb (mask 0xffffffff)
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: attached ide-disk driver.
hda: host protected area => 1
hda: 234441648 sectors (120034 MB) w/8192KiB Cache, CHS=14593/255/63,
UDMA(100)
Partition check:
hda: hda1 hda2 hda3 hda4 < hda5 hda6 hda7 hda8 >
ide: late registration of driver.
.....
Journalled Block Device driver loaded
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,1), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,5), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,6), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,7), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,8), internal journal
EXT3-fs: mounted filesystem with ordered data mode.

I'm running Debian testing and I keep my system up to date at
least on a monthly basis. I had the sysadmin take a look but
he couldn't tell me anything. I've never seen this and I have a
feeling
something bad is about to happen.

I haven't tried the smartmon tools yet.

 

[Arno Wagner]

sorry, I left out some details.
It does appear that the other partitions simply
aren't mounted, because only /dev/hda2 shows up in the
mount command. But, for example:
mount: /dev/hda7 already mounted or /tmp busy

here's the fstab entry for /home:
/dev/hda8 /home ext3 defaults 0 2

now I can cd to /home and notice this output from
'df' shows the name of the file system missing:

root@m1:home [4] df -h .
Filesystem Size Used Avail Use% Mounted on
- 84G 14G 67G 17% /home

Now that _is_ strange.

I'm using 'df dot'. If I just do 'df' it only reports back on
/dev/hda2 which is the root partition.

the boot log showed no problems that I could notice.
Here's some excerpts:

Uniform Multi-Platform E-IDE driver Revision: 7.00beta4-2.4
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
ICH4: IDE controller at PCI slot 00:1f.1
ICH4: chipset revision 2
ICH4: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hdaMA, hdbio
ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdcio, hddio
hda: WDC WD1200JB-00CRA1, ATA DISK drive
blk: queue c0371220, I/O limit 4095Mb (mask 0xffffffff)
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: attached ide-disk driver.
hda: host protected area => 1
hda: 234441648 sectors (120034 MB) w/8192KiB Cache, CHS=14593/255/63,
UDMA(100)
Partition check:
hda: hda1 hda2 hda3 hda4 < hda5 hda6 hda7 hda8 >
ide: late registration of driver.
....
Journalled Block Device driver loaded
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,1), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,5), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,6), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,7), internal journal
EXT3-fs: mounted filesystem with ordered data mode.
kjournald starting. Commit interval 5 seconds
EXT3 FS 2.4-0.9.19, 19 August 2002 on ide0(3,8), internal journal

EXT3-fs: mounted filesystem with ordered data mode.

Looks fine to me.

I'm running Debian testing and I keep my system up to date at
least on a monthly basis. I had the sysadmin take a look but
he couldn't tell me anything. I've never seen this and I have a
feeling
something bad is about to happen.

It certainly is a bit disturbing.

I haven't tried the smartmon tools yet.

Personally I don't think this is the disk. I think this might be some
problem with the kernel or a mismatch between the filesystem tools
and the kernel version. i.e. a reporting issue. Mount needs to be
not older than the kernel version in order to work right, since it
accesses kernel-internal things that may change.

I run about 20 boxes with debian testing and automatic updates. I have
never seen something like what you describe. I use my own kernels
though.

Some more questions:

Have you rebooted the box to see whether the problem goes away?
(Just asking. Don't do so if you have not yet.)

What kernel version? df version? mount version?
('cat /proc/version', 'df --version', 'mount --version')

What is the output of 'cat /proc/self/mounts'?

Can you see the contents of /dev/hda8 under /home/?

If you write a GB or so to /home/, does it show up in the
byte count for /dev/hda2 or for /dev/hda8?


Arno

 

[gstrock]

thanks for your help on this Arno.

cat /proc/version

Linux version 2.4.24 ([email protected]) (gcc version 3.2.2
20030222 (Red Hat Linux 3.2.2-5)) #2\ SMP Tue Feb 10 01:00:24 PST 2004

df --version

df (coreutils) 5.2.1

mount --version

mount: mount-2.12p

Strange that the kernel version says Red Hat Linux
when I think I'm running Debian. Well actually that
says the version of gcc is RH.

I can see all my files ok under all partitions, and the
byte count is correct if I add files to a partition.

I was beginning to think it's a kernel/tools mismatch,
except I have a second server I lease from this place
(managed.com) and the version numbers are all like
above and I have no problem with the 'df' or 'mount'
commands. Just the drive is smaller, 80 Gigs instead
of 120 Gigs.

The uptime is 153 days on the machine in question,
and I'm afraid to reboot it at this point.