[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

[linux-lvm] vgscan error after resizing lv; can't boot now



Hello,

I have a 72GB SCSI in RAID mirror. Using LVM version 1.0.3, Redhat 9.

Before resizing, the partitions are (snip from lvmdiskscan):

/dev/sda1  [      31.35 MB] Primary   [0xDE]  (Dell Utility)
/dev/sda2  [      19.53 GB] NTFS
/dev/sda3  [     298.08 MB] Primary  LINUX native partition [0x83]  (/boot)
/dev/sda4  [      48.51 GB] Primary  Windows98 extended partition [0x0F]
/dev/sda5  [       1.95 GB] Extended LINUX swap partition [0x82]
/dev/sda6  [       5.86 GB] Extended LVM partition [0x8E] (VG "Volume00")
/dev/sda7  [      40.69 GB] Extended LVM partition [0x8E] (VG "Volume00")

The vg "Volume00" contains PV /dev/sda6 and /dev/sda7, and the LVs (all
ext3) are:
/dev/Volume00/root        (/)
/dev/Volume00/home      (/home)
/dev/Volume00/opt         (/opt)
/dev/Volume00/tmp        (/tmp)
/dev/Volume00/usr         (/usr)
/dev/Volume00/usrlocal   (/usr/local)
/dev/Volume00/var         (/var)


I wanted to remove the NTFS partition, make /dev/sda2 as LVM partition, and
add it to "/dev/Volume00/opt". Here's what I did:

   boot to linux rescue mode
   # vgscan
   # vgchange -a y
   use fdisk to remove /dev/sda2, recreate /dev/sda2
   # mke2fs -j /dev/sda2
   use parted to set the partition type to LVM partition (0x8e)
   # vgextend Volume00 /dev/sda2
   Get the partition size (19.53GB) from # pvdisplay /dev/sda2
   # e2fsadm -L+19.53G /dev/Volume00/opt
   Still have 532MB as shown from # pvdisplay /dev/sda2
   # e2fsadm -L+532M /dev/Volume00/opt
   Step 11 produced error, showed many inode numbers, and asked if I wanted
   to fix for each inode number. I pressed a few 'y' and Ctrl-C to
   terminate.
   # e2fsck -y /dev/sda2  (to answer 'y' to all questions regarding fixing
   inode)
   reboot and got a kernel panic with the following msg:
   vgscan -- removing "/etc/lvmtab" and "/etc/lvmtab.d"
   vgscan -- creating empty "/etc/lvmtab" and "/etc/lvmtab.d"
   vgscan -- reading all physical volumes (this may take a while...)
   vgscan -- scanning for all active volume group(s) first
   vgscan -- reading data of volume group "Volume00" from physical
   volume(s)
   vgscan -- only found 6816 of 7424 LEs for LV /dev/Volume00/opt (2)
   vgscan -- ERROR "vg_read_with_pv_and_lv(): allocated LE of LV" can't get
   data of volume group "Volume00" from physical volume(s)
   vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
   vgscan -- WARNING: This program does not do a VGDA backup of your volume
   group

I rebooted into rescue mode, and vgscan also gave the same error as above.
And "/etc/lvmtab" and "/etc/lvmtab.d" are blank.

How can I recover the VG? Is /dev/sda2 oversized by e2fsadm and therefore
mess up the vg?

I have attached the output from pvdata for /dev/sda2,6,7. They contain info
on related VG, PV and LV.

Thanks and regards,
zhimin


(See attached file: pvdata-sda2-6-7.zip)

Attachment: pvdata-sda2-6-7.zip
Description: Zip archive


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]