Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!

Bug 74995

Summary: absent LABEL in fstab causes "mount -a" to segment fault
Product: Gentoo Linux Reporter: William Blew <wblew>
Component: [OLD] Core systemAssignee: Gentoo's Team for Core System packages <base-system>
Status: RESOLVED FIXED    
Severity: normal CC: emilbeinroth
Priority: High    
Version: 2004.3   
Hardware: x86   
OS: All   
Whiteboard:
Package list:
Runtime testing required: ---
Attachments: "emerge info" capture from the computer demonstrating this bug
Don't call streq(..) in mounted(..) if an argument would be NULL.

Description William Blew 2004-12-19 17:49:24 UTC
This bug is with the sys-apps/util-linux-2.12i-r1 package's /bin/mount command.

The following line within /etc/fstab (no partition with SYSTEM label):
   LABEL=SYSTEM   /mnt/system     vfat            gid=100,umask=0002      0 0

where there isn't any partition whose VFAT label is SYSTEM, causes this command:
   mount -a

to generate a segment fault. Commenting the fstab line above prevents mount's segment fault. Similarly, the following fstab line also causes mount to segment fault (no partition with bootx label):
   LABEL=/bootx     /boot           ext2            noatime                 1 2

This is quite noticable as it causes /sbin/rc to terminate, during the bootup sequence, without having mounted any of the local filesystems.

This regression bug was introduced with sys-apps/util-linux-2.12i-r1 and is not present in the 2.12-r4 release.

I will attach my emerge info outout.
Comment 1 William Blew 2004-12-19 17:51:15 UTC
Created attachment 46382 [details]
"emerge info" capture from the computer demonstrating this bug
Comment 2 Emil Beinroth 2004-12-28 13:23:05 UTC
Created attachment 47055 [details, diff]
Don't call streq(..) in mounted(..) if an argument would be NULL.

2.12-r4 and 2.12i behave the same if you apply the patch to 2.12i: They
continue to mount other filesystems.
Comment 3 SpanKY gentoo-dev 2005-01-09 19:34:42 UTC
fixed in 2.12p, thanks guys :)