[Ocfs2-users] mkfs.ocfs2: double free or corruption

[Sunil Mushran]

File a bugzilla and attach the coredump.

One thing to try would be reducing the number of slots.

santi at usansolo.net wrote:
> Dear Srs,
>
> I get this error when running "mkfs.ocfs2":
>
> =================================================================================
> # mkfs.ocfs2 -b 4K -C 32K -N 255 -L backup_ocfs2_001 /dev/sdb1
> mkfs.ocfs2 1.2.7
> Filesystem label=backup_ocfs2_001
> Block size=4096 (bits=12)
> Cluster size=32768 (bits=15)
> Volume size=6000488677376 (183120382 clusters) (1464963056 blocks)
> 5678 cluster groups (tail covers 3070 clusters, rest cover 32256 clusters)
> Journal size=268435456
> Initial number of node slots: 255
> Creating bitmaps: done
> Initializing superblock: done
> Writing system files: 
> *** glibc detected *** mkfs.ocfs2: double free or corruption (!prev):
> 0x0e10e000 ***
> ======= Backtrace: =========
> /lib/libc.so.6[0x352aa6]
> /lib/libc.so.6(cfree+0x90)[0x355fc0]
> mkfs.ocfs2[0x804a460]
> mkfs.ocfs2[0x804b429]
> mkfs.ocfs2[0x804d266]
> /lib/libc.so.6(__libc_start_main+0xdc)[0x301dec]
> mkfs.ocfs2[0x80495b1]
> ======= Memory map: ========
> 002bf000-002c2000 r-xp 00000000 08:02 1474588    /lib/libuuid.so.1.2
> 002c2000-002c3000 rw-p 00002000 08:02 1474588    /lib/libuuid.so.1.2
> 002ca000-002e3000 r-xp 00000000 08:02 1474578    /lib/ld-2.5.so
> 002e3000-002e4000 r--p 00019000 08:02 1474578    /lib/ld-2.5.so
> 002e4000-002e5000 rw-p 0001a000 08:02 1474578    /lib/ld-2.5.so
> 002e7000-002e9000 r-xp 00000000 08:02 1474832    /lib/libcom_err.so.2.1
> 002e9000-002ea000 rw-p 00001000 08:02 1474832    /lib/libcom_err.so.2.1
> 002ec000-00426000 r-xp 00000000 08:02 1474585    /lib/libc-2.5.so
> 00426000-00428000 r--p 0013a000 08:02 1474585    /lib/libc-2.5.so
> 00428000-00429000 rw-p 0013c000 08:02 1474585    /lib/libc-2.5.so
> 00429000-0042c000 rw-p 00429000 00:00 0 
> 005e4000-005ef000 r-xp 00000000 08:02 1476311   
> /lib/libgcc_s-4.1.2-20070626.so.1
> 005ef000-005f0000 rw-p 0000a000 08:02 1476311   
> /lib/libgcc_s-4.1.2-20070626.so.1
> 00ad9000-00ada000 r-xp 00ad9000 00:00 0          [vdso]
> 08048000-0805f000 r-xp 00000000 08:02 1114366    /sbin/mkfs.ocfs2
> 0805f000-08061000 rw-p 00016000 08:02 1114366    /sbin/mkfs.ocfs2
> 09b46000-0e12c000 rw-p 09b46000 00:00 0 
> b7e00000-b7e21000 rw-p b7e00000 00:00 0 
> b7e21000-b7f00000 ---p b7e21000 00:00 0 
> b7f13000-b7f15000 rw-p b7f13000 00:00 0 
> bff36000-bff4b000 rw-p bff36000 00:00 0          [stack]
> Aborted
> =================================================================================
>
> I'm trying to give OCFS2 format to a 6TB iSCSI partition/device, this is
> the system information:
>
> - ocfs2-tools-1.2.7-1.el5, official RPM package from Oracle.
>
> - Self compiled OCFS2 v1.3.9-0.1 under a CentOS 5 box:
>
> ./configure --with-kernel=/lib/modules/`uname -r`/build
> --enable-vfs-rename-patch-override=yes
> make
> make install
>
> - Running Virtuozzo 4.0 , version is "2.6.18-028stab053.17-ent", it's a
> kernel modification to enable operating system-level virtualization.
>
> - The device is a NEXSAN SATABeast, attached via iSCSI, 6TB size with only
> one partition:
>
> (parted) print                                                            
>
> Model: NEXSAN SATABeast (scsi)
> Disk /dev/sdb: 6000GB
> Sector size (logical/physical): 512B/512B
> Partition Table: gpt
>
> Number  Start   End     Size    File system  Name     Flags
>  1      17.4kB  6000GB  6000GB               primary    
>
> What can I do to solve this problem and give format via "mkfs.ocfs2", maybe
> it's a problem related to Virtuozzo? Thanks!
>
> Regards,
>
> --
> Santi Saez
> http://woop.es
>
>
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users at oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>