Red Hat Bugzilla – Bug 2377
dump fails, claiming corruption of FS which passes fsck
Last modified: 2008-05-01 11:37:50 EDT
[root@beavis /root]# /sbin/dump 4ubBf 32 24000000 /dev/tape /dev/sda2 DUMP: Date of this level 4 dump: Mon Apr 26 13:14:14 1999 DUMP: Date of last level 0 dump: Tue Feb 23 16:54:11 1999 DUMP: Dumping /dev/sda2 (/) to /dev/tape DUMP: mapping (Pass I) [regular files] DUMP: estimated 280057 tape blocks on 0.01 tape(s). DUMP: dumping (Pass III) [directories] /dev/sda2: EXT2 directory corrupted while converting directory #2 DUMP: Broken pipe [root@beavis /root]# rpm -q dump dump-0.3-17 HARDWARE IS: Apr 26 09:11:57 beavis kernel: (scsi0) <Adaptec AIC-7890/1 Ultra2 SCSI host adap ter> found at PCI 6/0 Apr 26 09:11:57 beavis kernel: (scsi0) Wide Channel, SCSI ID=7, 32/255 SCBs Apr 26 09:11:57 beavis kernel: (scsi0) Downloading sequencer code... 407 instruc tions downloaded Apr 26 09:11:57 beavis kernel: scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-F ast SCSI) 5.1.4/3.2.4 Apr 26 09:11:57 beavis kernel: <Adaptec AIC-7890/1 Ultra2 SCSI host adapt er> Apr 26 09:11:57 beavis kernel: scsi : 1 host. Apr 26 09:11:57 beavis kernel: Vendor: IBM Model: DRVS09V Rev: 00F0 Apr 26 09:11:57 beavis kernel: Type: Direct-Access ANSI SCSI revision: 03 Apr 26 09:11:57 beavis kernel: Detected scsi disk sda at scsi0, channel 0, id 0, lun 0 Apr 26 09:11:57 beavis kernel: Vendor: IBM Model: DRVS09V Rev: 0100 Apr 26 09:11:57 beavis kernel: Type: Direct-Access ANSI SCSI revision: 03 Apr 26 09:11:57 beavis kernel: Detected scsi disk sdb at scsi0, channel 0, id 1, lun 0 Apr 26 09:11:57 beavis kernel: Vendor: HP Model: C1537A Rev: L708 Apr 26 09:11:57 beavis kernel: Type: Sequential-Access ANSI SCSI revision: 02 Apr 26 09:11:57 beavis kernel: Detected scsi tape st0 at scsi0, channel 0, id 3, lun 0 Apr 26 09:11:57 beavis kernel: scsi : detected 1 SCSI tape 2 SCSI disks total. Apr 26 09:11:57 beavis kernel: (scsi0:0:0:0) Synchronous at 20.0 Mbyte/sec, offs et 15. Apr 26 09:11:57 beavis kernel: SCSI device sda: hdwr sector= 512 bytes. Sectors= 17928698 [8754 MB] [8.8 GB] Apr 26 09:11:57 beavis kernel: (scsi0:0:1:0) Synchronous at 20.0 Mbyte/sec, offs et 15. Apr 26 09:11:57 beavis kernel: SCSI device sdb: hdwr sector= 512 bytes. Sectors= 17928698 [8754 MB] [8.8 GB]
Can you make sure that 1) the file system has been fsck'd. It's not enough to rely on the fsck run during reboot -- you will need to run e2fsck with the -f flag. 2) The file system that you are dumping is "quiet". Ideally dump should be run with an unmounted file system. 3) Check /var/log/messages for disk hardware errors. You might also try recompiling and running dump-0.4 from rawhide.redhat.com to see whether there is a dump problem that has been fixed already.
Did dump-0.4 help your problem>
closed due to lack of feedback. Likely user error or fixed in current release.