Hello all,
 First off, i'm talking about a SXDE build 89. Sorry if that was discussed here 
before, but i did not find anything related on the archives, and i think is a 
"weird" issue... 

 If i try to remove a specific file, i got:
# rm file1
rm: file1: No such file or directory
# rm -rf dir2
rm: Unable to remove directory dir2: Directory not empty

 Take a look:
------- cut here --------
# ls
dir1 dir2

# ls dir2/
file1

# ls -i
5147 dir1                   1924 dir2

# zdb -dddd mypool/myfs 1924
...
    Object  lvl   iblk   dblk  lsize  asize  type
      1924    1    16K     1K     1K     1K  ZFS directory
                                 264  bonus  ZFS znode
        path    /myfs/dir2
        uid     0
        gid     12
        atime   Tue Dec  9 16:07:03 2008
        mtime   Wed Dec 17 14:50:09 2008
        ctime   Wed Dec 17 14:50:09 2008
        crtime  Wed Nov 26 16:19:31 2008
        gen     207918
        mode    42770
        size    3
        parent  1923
        links   2
        xattr   0
        rdev    0x0000000000000000
        microzap: 1024 bytes, 1 entries

                file1 = 1966 (type: Regular File)

# zdb -dddd mypool/myfs 1966
    Deadlist: 1098 entries, 68.9M

    Object  lvl   iblk   dblk  lsize  asize  type
zdb: dmu_bonus_hold(1966) failed, errno 2

------- cut here --------
 I did try to find something in the bug database, but did not find anything 
about that kind of problem in a "good" pool. I have no errors reported by the 
zpool status, and a zpool scrub ends with no error either. I just can't unlink 
that file/directory. 
 I guess i can fix that destroying the filesystem, and creating it again. But i 
want to know your opinion about it, and if you know about that error/bug. 
Besides, recreate the filesystem will be downtime...
 There is a way to fix that? 

 ps.: I'm using a slog device, and i think the header of that "zdb -dddd" 
command, is related to the ZIL, but i'm not sure (i will try to learn more 
about that zil operations soon ;-). 
 Could be inconsistency caused by the slog device? Some operations that was not 
commited to the pool? Hmm, but that should not cause the zpool to became 
corrupted...

 Thanks a lot for your time

 Leal
[http://www.eall.com.br/blog]
-- 
This message posted from opensolaris.org
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Reply via email to