[linux-lvm] *** ANNOUNCEMENT *** LVM 0.9.1 beta1 available at www.sistina.com

[Rob Fugina]

On Sat, Jan 13, 2001 at 05:04:08PM +0100, Jan Niehusmann wrote:
> On Sat, Jan 13, 2001 at 01:15:32AM +0000, Heinz J. Mauelshagen wrote:
> > We'ld appreciate a couple of days for test feedback before pushing a 2.4.0
> > patch to Linus.
> 
> It still has the problem I reported on Dec 22. On Dec 23 I wrote:
> 
> : Ok, I found the problem: The sort order of the pv's is wrong, and 0.9 is
> : missing the code that sorts them. 0.8final's pv_read_all_pv_of_vg.c contains
> : 
> :       for ( p = 0; pv_tmp[p] != NULL; p++) {
> :          if ( strcmp ( pv_tmp[p]->vg_name, vg_name) == 0) {
> :             pv_this[pv_tmp[p]->pv_number-1] = pv_tmp[p];
> :             np++;
> :          }
> :       }
> : 
> : and 0.9's doesn't.

I wonder if this is why I lost all of my data last Thursday...

vg w/ 2 pv's (hde1 & hdg1) and 3 lv's.  Add a pv (hdc1) and
reboot, and all of a sudden, even though the vg is still there,
with 3 pv's and 3 lv's, the lv's won't even fsck (bad magic
number in superblock -- doesn't even look like a filesystem was
even there).  I though maybe it had a problem w/ rearranging the
pv's or something...  How else could it completely lose track
of my data?

It was kernel 2.4.0 w/ the several-line vg_extend patch, and the 0.9
tools...

Oh, well -- just a WAG...

Rob

-- 
Rob Fugina, Systems Guy
robf at geekthing.com -- http://www.geekthing.com
My firewall filters MS Office attachments.

               Mud is not one of the four food groups.