xfs
[Top] [All Lists]

Problems with xfsdump on snapshot LV

To: linux-xfs@xxxxxxxxxxx
Subject: Problems with xfsdump on snapshot LV
From: Stephan Austermuehle <au@xxxxxxx>
Date: Thu, 6 Jun 2002 11:03:38 +0200
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mutt/1.2.5i
Hi,

I did the following:

        # xfs_freeze -f /usr
        # lvcreate -L 256m -n /dev/vg00/lv_usr_snap -s /dev/vg00/lv_usr
        # xfs_freeze -u /usr
        # mount -t xfs -o ro,nouuid /dev/vg00/lv_usr_snap /mnt
        # xfsdump -l 0 -p 60 - /mnt |buffer ... > /dev/nst1

This results in the following error messages:

        XFS mounting filesystem lvm(58,19)
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        lvm - lvm_map: ll_rw_blk write for readonly LV /dev/vg00/lv_usr_snap
        I/O error in filesystem ("lvm(58,19)") meta-data dev 0x3a13 block 
0x1007fc
               ("xlog_iodone") error 5 buf count 6656
        xfs_force_shutdown(lvm(58,19),0x2) called from line 939 of file 
xfs_log.c.  Return address = 0xc01bdbe6
        Log I/O Error Detected.  Shutting down filesystem: lvm(58,19)
        Please umount the filesystem, and rectify the problem(s)
        Unable to handle kernel NULL pointer dereference at virtual address 
00000020
         printing eip:
        c0131f02
        *pde = 00000000
        Oops: 0000
        CPU:    0
        EIP:    0010:[<c0131f02>]    Not tainted
        EFLAGS: 00010202
        eax: 00003a13   ebx: 00000000   ecx: 00003a13   edx: 00003a13
        esi: 00000001   edi: 00000000   ebp: 00000000   esp: d01fdd60
        ds: 0018   es: 0018   ss: 0018
        Process lvremove (pid: 25437, stackpage=d01fd000)
        Stack: dfed8740 df466400 c0347df6 dfb6d000 3a13d000 00000000 00000000 
c0132000 
               dfed8740 00000000 dfe3b370 c0237668 00003a13 00000000 00003a13 
00000000 
               bffff5ac c0347e64 dfb6d000 00000180 dfb6d4c4 00000010 c0234c21 
00000000 
        Call Trace: [<c0132000>] [<c0237668>] [<c0234c21>] [<c01e2927>] 
[<c012581c>] 
           [<c0125849>] [<c01b66cf>] [<c01ced1f>] [<c01b66cf>] [<c01ced1f>] 
[<c01391cf>] 
           [<c013114e>] [<c01300f1>] [<c0130002>] [<c013cf89>] [<c0106e63>] 

        Code: 8b 7b 20 0f b7 54 24 12 66 39 53 0c 0f 85 84 00 00 00 83 7b 
 
Why does xfsdump try to write to a filesystem that should be backup'd?

Thanks for your help,

Stephan


<Prev in Thread] Current Thread [Next in Thread>