On Aug 3, 2005, at 11:10 AM, James Melin wrote:
No, not sharing and DASD at this point. That's a future 'to do' but
not
until we're on SLES 9, etc.
So even though I see the I/O errors, and an erep message against
the DASD,
if fsck -n comes back clean, there should be no worries?
Correct.
I also had a file system get corrupted on Friday. I restored it to
Thursday's backup. Friday backup had the corrupted file system which I
restored to another device and mounted on a diff guest
The behaviour of fsck -n against THIS file system is different....
vadnais:~ # fsck -n /dev/dasdf1
fsck 1.28 (31-Aug-2002)
e2fsck 1.28 (31-Aug-2002)
Warning! /dev/dasdf1 is mounted.
Warning: skipping journal recovery because doing a read-only
filesystem
check.
/dev/dasdf1 has gone 184 days without being checked, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Entry '..' in .../??? (129889) has deleted/unused inode 436565.
Clear? no
Entry '..' in .../??? (179216) has deleted/unused inode 436575.
Clear? no
Error reading block 902856 (Attempt to read block from filesystem
resulted
in short read). Ignore error? no
Error reading directory block 902856 (inode 436576): Attempt to
read block
from filesystem resulted in short read
Continue? no
e2fsck: aborted
fsck.ext2 /dev/dasdf1 failed (status 0x8). Run manually!
Yeah. That one needs to be fixed.
Which is exactly what I saw at IPL time on Sunday - Which made me
go look
at all the other penguins to see if there was anything in the log.
Then the
erep/io error stuff on Monday really freaked me out but those file
systems
are passing the fsck -n test. So it looks like fsck -n is giving me
the
integrity test I desired. Now to check all the systems.
In any case what should I make of the i/o errors and erep messages
I posted
earlier?
They're harmless but annoying. Defining the dasd ro in the parmfile
and rerunning zipl will make them go away. What they're telling you
is that Linux is trying to write to the partition (to initialize/
update the journal file), only it can't, because from its
perspective, someone has set hardware write-protect on the disk.
Adam
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
