[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

[linux-lvm] [PATCH] vg_check_name.c



This in regards to the following bug report I filed with Debian:


http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=140957&msg=4&repeatmerged=y
es

I did some more debugging to figure out what bit me, and it turns out that
my problem was that my volume group was named just "a", and there was a new
check added to 1.1rc1 in vg_check_name.c that matched it against "ataraid"
(defined in _devdir[] in lvm_dir_cache.c).  Very confusing.  I don't think
the check was intended to work that way, so here's a patch:

Index: vg_check_name.c
===================================================================
RCS file: /data/cvs/LVM/tools/lib/vg_check_name.c,v
retrieving revision 1.7
diff -u -r1.7 vg_check_name.c
--- vg_check_name.c 2002/03/12 13:58:04 1.7
+++ vg_check_name.c 2002/04/23 21:40:15
@@ -84,7 +84,7 @@
             else
                filter = _devdir[d];

-            if ( strncmp ( vg_name, filter, strlen ( vg_name)) == 0) {
+            if ( strncmp ( vg_name, filter, strlen ( filter)) == 0) {
                ret = -LVM_EVG_CHECK_NAME;
                break;
             }

This bit me on several machines, since I used the same naming convention.
:-(

Cheers,

 - Jim

BTW, I don't believe that the Debian bug severity should have been
downgraded to just "important", as I almost lost all my data.






[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]