[DRBD-user] DRBD over an existing filesystem

En Chiang Lee ena at india.hp.com
Fri Mar 26 11:23:07 CET 2004


Here's the output:

[root at dlssi3 root]# mkfs -t ext3 -b 1024 /dev/hde6
mke2fs 1.32 (09-Nov-2002)
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
125952 inodes, 1004031 blocks
50201 blocks (5.00%) reserved for the super user
First data block=1
123 block groups
8192 blocks per group, 8192 fragments per group
1024 inodes per group
Superblock backups stored on blocks: 
        8193, 24577, 40961, 57345, 73729, 204801, 221185, 401409, 663553

mkfs.ext3: Attempt to write block from filesystem resulted in short
write while zeroing block 1004024 at end of filesystem
Writing inode tables: done                            
Creating journal (8192 blocks): done
Writing superblocks and filesystem accounting information: done

This filesystem will be automatically checked every 25 mounts or
180 days, whichever comes first.  Use tune2fs -c or -i to override.
[root at dlssi3 root]# drbdadm up r0
[root at dlssi3 root]# drbdadm primary r0
[root at dlssi3 root]# fsck /dev/nbd/0
fsck 1.32 (09-Nov-2002)
e2fsck 1.32 (09-Nov-2002)
The filesystem size (according to the superblock) is 1004031 blocks
The physical size of the device is 506016 blocks
Either the superblock or the partition table is likely to be corrupt!
Abort<y>? yes

[root at dlssi3 root]# blockdev --getsize /dev/nbd/0
1012032
[root at dlssi3 root]# blockdev --getsize /dev/hde6
2008062
[root at dlssi3 root]# checkout/drbd/testing/read_gc.pl 
                                     ConnectedInd |
                                  lastState |     |
                         ArbitraryCnt |     |     |
                   ConnectedCnt |     |     |     |
               TimeoutCnt |     |     |     |     |
           HumanCnt |     |     |     |     |     |
   Consistent |     |     |     |     |     |     |
resource|     |     |     |     |     |     |     |   Size
--------+-----+-----+-----+-----+-----+-----+-----+----------+
state file corrupt at checkout/drbd/testing/read_gc.pl line 39.
[root at dlssi3 root]# 

Thanks,

En Chiang

On Fri, 2004-03-26 at 15:11, Philipp Reisner wrote:
> Am Freitag, 26. März 2004 10:28 schrieb En Chiang Lee:
> > Hi Lars,
> >
> > Using DRBD over an existing fs seems to work with fs size <1G. Here are
> > the sizes I tried:
> >
> > 512M (fs block size = 1024) -- fsck works fine
> > 1G (fs block size = 4096) -- fsck fails with mismatched sizes
> > 1G (fs block size = 1024) -- fsck fails with mismatched sizes
> >
> > In all these cases, meta-disk is another partition (i.e. not internal)
> >
> > Regards,
> >
> > En Chiang
> >
> 
> Could you send us the output of
> 
> blockdev --getsize /dev/nbd/0
> blockdev --getsize /dev/ll_dev
> ~drbd07/testing/read_gc.pl
> 
> and were the devices connected of unconnected?
> 
> -Philipp




More information about the drbd-user mailing list