Weird: I/O error. Files are detected but cannot be seen...

kloss · 02-08-2006, 06:27 PM

Hello
The title of the thread is a bit obvious, but... look at this:

Code:

01:16:00 /var/run
ROOT # ls -la
ls: dhclient.pid: Input/output error
ls: crond.pid: Input/output error
ls: inetd.pid: Input/output error
ls: Xprt_0.pid: Input/output error
total 52K
drwxr-xr-x   6 root      root    4,0K 2006-02-08 23:59 .
drwxr-xr-x  14 root      root    4,0K 2006-02-03 11:51 ..
srw-rw-rw-   1 root      root       0 2006-02-08 23:58 acpid.socket
drwxr-xr-x   2 root      root    4,0K 2006-02-06 10:31 alsa
-rw-r--r--   1 root      root       5 2006-02-09 00:58 cardmgr.pid
drwxr-xr-x   2 root      root    4,0K 2006-02-08 23:58 cups
-rw-r--r--   1 root      root       5 2006-02-08 23:59 cvpnd.pid
drwx------   2 fetchmail nogroup 4,0K 2006-02-04 16:23 fetchmail
-rw-r--r--   1 root      root       5 2006-02-08 23:58 klogd.pid
-rw-r--r--   1 root      root      16 2006-02-09 00:58 stab
-rw-r--r--   1 root      root       5 2006-02-08 23:58 syslogd.pid
drwxr-xr-x   2 root      root    4,0K 2005-08-04 19:20 usb
-rw-rw-r--   1 root      utmp    3,8K 2006-02-08 23:59 utmp
-rw-r--r--   1 root      root     162 2006-02-08 23:58 Xprint_servers

01:16:06 /var/run
ROOT # less foo
foo: No such file or directory

01:16:32 /var/run
ROOT # less inetd.pid
inetd.pid: Input/output error

What do you think about these 4 files wich are detected, but which do not appear in the directory...

I don't ask that just because I'm curious: today I've got a dozen of kernel panics / boot freezes

and the /var/log/boot indicates that something is wrong with these files.

Thanks for your help
++

halvy · 02-08-2006, 08:37 PM

pids are:

http://linux.about.com/library/cmd/blcmdln_pid.htm

i don't now why they are showing up, but i doubt they have anything to do with panics.

panics are *usually* caused by bad kernels, missing initrd (image file), or bad dev setups.

good luck, let us know.

kloss · 02-09-2006, 03:41 AM

Re. I always thought checking the partitions was a waste of time, but I must say that fsck did here a good job :

Code:

root@1[mnt]# fsck /dev/hda7
fsck 1.39-WIP (10-Dec-2005)
e2fsck 1.39-WIP (10-Dec-2005)
/var has been mounted 31 times without being checked, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Entry 'dhclient.pid' in /run (127489) has deleted/unused inode 127492.  Clear<y>? yes

Entry 'crond.pid' in /run (127489) has deleted/unused inode 127496.  Clear<y>? yes

Entry 'Xprt_0.pid' in /run (127489) has deleted/unused inode 127502.  Clear<y>? yes

Entry 'cvpnd.pid' in /run (127489) has deleted/unused inode 127500.  Clear<y>? yes

Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences:  -266240
Fix<y>? yes

Free blocks count wrong for group #8 (32260, counted=32261).
Fix<y>? yes

Free blocks count wrong (510398, counted=510399).
Fix<y>? yes

Free inodes count wrong for group #8 (15931, counted=15929).
Fix<y>? yes

Free inodes count wrong (362684, counted=362682).
Fix<y>? yes


/var: ***** FILE SYSTEM WAS MODIFIED *****
/var: 3846/366528 files (6.8% non-contiguous), 222558/732957 blocks

Now I don't expercience issue with "ghost files" anymore.

Quote:

Originally Posted by halvy

panics are *usually* caused by bad kernels, missing initrd (image file), or bad dev setups.

Yep. In fact I don't know what happend. Yesterday I worked as a normal user (not root) and did not perform any task such as adding a piece of hardware or modifying a config file. The next time I could not boot my comp: boot freeze while entering runlevel 2. I could boot as a single user, though. When I shut down the box, I got hundreds of error messages - which one cannot understand. Examples:

Code:

EIP is at do_page_fault+0xd6/0x6bf
eax: dfa40000 ebx:00000000 ecx:0000007b edx:ffffff7b esi:00030001
edi:0000000d ebp:0000000b esp: dfa417c8
ds: 007b es:007b ss:0008
Unable to handle kernel paging request at virtual address ffffffef
printing eip:
c0114fe6
*pde=00002067
*pte=00000000
Recursive die() failure, output suppressed
 <0> Kernel panic - not syncing: Fatal exception in interrupt

Code:

inode hda2: 96697 at df6aa440: mode 120777, nlink 1, next 0

I rebooted, and rebooted, and rebooted - always the same problem. But after around ten reboots all worked.

That's mysterious for me. So far I found the kernel rock solid - never had a kernel panic. Then a problem occurs, though I did not change anything on my box. Then it disappeared... Strange...