OK, as nobody else answers, I will give some more info. In the
meantime I found out that I can use xpeek to reset the FM_DIRTY flag
in the superblocks ("su", "m 12 0", "x", "q"), so that I can mount
the faulty partition again. But the problems with this directory
persist:
> The directory on /dev/hda1 (AM_1353-272) where I tried to copy the
> file in, is not accessible after the reboot:
>
> > ls -laF AM_1353-272
> ls: AM_1353-272: Input/output error
I found out the inode and could rename the directory to "test" using
xpeek on the parent directory
> dir 41568
idotdot = 71200
41664 test
but I cannot list it
> dir 41664
directory: Not a directory!
Is there any way to remove it (I didn't find an obvious way in the
xpeek man-page)? Or is the only way to get a stable FS again, to
reformat and replay the backup? Should I downgrade to jfs-1.0.12
and/or kernel-2.4.17 again?
Greetings,
Peter.
_______________________________________________
Jfs-discussion mailing list
[EMAIL PROTECTED]
http://www-124.ibm.com/developerworks/oss/mailman/listinfo/jfs-discussion
