I doubt that this is actually an ext3 problem so I'm taking a bit of a liberty, but... We have several machines that have two ext3 partitions (/, /boot) and a large raw partition. Something happened that destroyed both the ext3 partitions on all machines. I appears that the app writing to the raw partition had a bug that caused it to write to negative block numbers and that this wrote all over the disk. But I would have thought that the writes would have been constrained but the raw device. Is there an ext3 issue or is it a bug in the rawdevice implementation? TIA -- Ian Leonard eMail: ileonard@ntlworld.com Phone: +44 (0)1865 765273 Please ignore spelling and punctuation - I did.
Since i use 2.4.18-ac i'm oblige to rebuild the map file in my /boot partition after having used ext3 and raw device on other partitions /dev/hda2: /boot /dev/hda5: /root /dev/hda6: raw /dev/hda7: a partition written through ext3 journaled mode I don't know why ? It seems the map file is corrupted ( LIL- output given by LILO). This is reproductible, but i never get that with 2.4.14 for instance. If somebody as an idea it will be welcome. For your problem, you said the application uses a negative addressing on raw device. How this is possible, as raw interface is ( pseudo device name, byte offset) ? Do you mean that negative offset values are accepted ? Eric --------- Ian Leonard wrote:> I doubt that this is actually an ext3 problem so I'm taking > a bit of a liberty, but... > > We have several machines that have two ext3 partitions > (/, /boot) and a large raw partition. Something happened that > destroyed both the ext3 partitions on all machines. > > I appears that the app writing to the raw partition had > a bug that caused it to write to negative block numbers > and that this wrote all over the disk. But I would have > thought that the writes would have been constrained > but the raw device. Is there an ext3 issue or is it a bug > in the rawdevice implementation? > TIA > > -- > Ian Leonard > eMail: ileonard@ntlworld.com > Phone: +44 (0)1865 765273 > > Please ignore spelling and punctuation - I did. > > _______________________________________________ > Ext3-users mailing list > Ext3-users@redhat.com > https://listman.redhat.com/mailman/listinfo/ext3-users-- MZ
Hi, On Sat, Jul 06, 2002 at 10:18:37AM +0000, Ian Leonard wrote:> I doubt that this is actually an ext3 problem so I'm taking > a bit of a liberty, but... > > We have several machines that have two ext3 partitions > (/, /boot) and a large raw partition. Something happened that > destroyed both the ext3 partitions on all machines. > > I appears that the app writing to the raw partition had > a bug that caused it to write to negative block numbers > and that this wrote all over the disk.Shouldn't be able to happen --- negative offsets are rejected elsewhere in the kernel before raw devices can get to see them. The lseek/llseek syscalls return EINVAL if you try to set the offset to be negative for normal read/write syscalls, and the pread and pwrite syscalls also check for negative offsets before allowing things to proceed.> But I would have > thought that the writes would have been constrained > but the raw device.Yes, they should be. --Stephen