Description of problem: mkfs -j should use mkfs.ext3, and support filesystems of 4194303 physical extents. Instead it uses mkfs.ext2, and fails with a confusing message. Version-Release number of selected component (if applicable): util-linux-2.13-0.52.el5_4.1 How reproducible: Always. Steps to Reproduce: 1. Create an LV of 4194303 PEs. 2. mkfs -j on that LV. Actual results: mkfs.ext2: Filesystem too large. No more than 2**31-1 blocks (8TB using a blocksize of 4k) are currently supported. Expected results: mkfs should actually execute "mkfs.ext3 -j", and create the FS. Additional info: [root@norimaki ~]# mkfs -j -L/noribeast2c /dev/noribeast2cd/noribeast2c mke2fs 1.39 (29-May-2006) mkfs.ext2: Filesystem too large. No more than 2**31-1 blocks (8TB using a blocksize of 4k) are currently supported. [root@norimaki ~]# lvdisplay /dev/noribeast2cd/noribeast2c --- Logical volume --- LV Name /dev/noribeast2cd/noribeast2c VG Name noribeast2cd LV UUID m4IkT4-CIuN-gtcz-ShjN-GkIp-lJec-2TpvKd LV Write Access read/write LV Status available # open 0 LV Size 16.00 TB Current LE 4194303 Segments 1 Allocation inherit Read ahead sectors auto - currently set to 256 Block device 253:12 [root@norimaki ~]# mkfs.ext3 -j -L/noribeast2c /dev/noribeast2cd/noribeast2c mke2fs 1.39 (29-May-2006) Filesystem label=/noribeast2c OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) 2147483648 inodes, 4294966272 blocks 214748313 blocks (5.00%) reserved for the super user First data block=0 131072 block groups 32768 blocks per group, 32768 fragments per group 16384 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, 102400000, 214990848, 512000000, 550731776, 644972544, 1934917632, 2560000000, 3855122432 Writing inode tables: 1529/131072 [root@norimaki ~]# mkfs -j -L/noribeast2c /dev/noribeast2cd/noribeast2c mke2fs 1.39 (29-May-2006) mkfs.ext2: Filesystem too large. No more than 2**31-1 blocks (8TB using a blocksize of 4k) are currently supported. [root@norimaki ~]# mkfs -j -L/noribeast2c /dev/noribeast2cd/noribeast2c
(In reply to comment #0) > mkfs -j should use mkfs.ext3 Why? See mkfs(8) man page, the default is ext2. The -j option is filesystem specific and it is not evaluated by the mkfs wrapper. You have to use: mkfs -t ext3 -j -L/noribeast2c /dev/noribeast2cd/noribeast2c I don't see any bug here.
At an abstract level, RHEL should be able to mkfs a 15tb filesystem -- right now it cannot without user intervention (to specify ext3). More specifically, sources such as mke2fs.8 make fuzz line between mke2fs and mkfs.ext3: DESCRIPTION mke2fs is used to create an ext2/ext3 filesystem (usually in a disk partition). device is the special file corresponding to the device (e.g /dev/hdXX). blocks-count is the number of blocks on the device. If omitted, mke2fs automagically figures the file system size. If called as mkfs.ext3 a journal is created as if the -j option was speci- fied. It is confusing that maximum filesystem size depends on "-t ext2" vs. "-t ext3", although this could minimally be addressed by updating the error message to point users to "-t ext3". mkfs.ext2: Filesystem too large. No more than 2**31-1 blocks (8TB using a blocksize of 4k) are currently supported. [root@norimaki ~]# ls -li /sbin/mke2fs /sbin/mkfs.ext? 1457463 -rwxr-xr-x 3 root root 49608 Sep 3 2009 /sbin/mke2fs 1457463 -rwxr-xr-x 3 root root 49608 Sep 3 2009 /sbin/mkfs.ext2 1457463 -rwxr-xr-x 3 root root 49608 Sep 3 2009 /sbin/mkfs.ext3
(In reply to comment #2) > At an abstract level, RHEL should be able to mkfs a 15tb filesystem -- right > now it cannot without user intervention (to specify ext3). The feature depends on filesystem type. The default filesystem for mkfs(8) is ext2. This behaviour dis escribed in the man page. I don't think that change any default which is used for years is a good idea. Closing. Please, if you believe that the problem should be fixed then ask for help at official Red Hat support service -- for more details see http://support.redhat.com. Thanks.