Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 39586 - hardened-sources-2.4.22-r2 & (evms-2.2.1 | evms-2.2.2 ) evms_activate segfaults
Summary: hardened-sources-2.4.22-r2 & (evms-2.2.1 | evms-2.2.2 ) evms_activate segfaults
Status: RESOLVED INVALID
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Core system (show other bugs)
Hardware: All Linux
: High major (vote)
Assignee: The Gentoo Linux Hardened Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2004-01-27 11:47 UTC by REdOG
Modified: 2004-06-01 12:06 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
config (.config,24.44 KB, text/plain)
2004-01-27 18:01 UTC, REdOG
Details

Note You need to log in before you can comment on or make changes to this bug.
Description REdOG 2004-01-27 11:47:25 UTC
I am running hardened-sources-2.4.22.r2 and trying to install evms.
When /etc/init.d/checkfs starts evms_activate, I get a segfault.

If I try to run evms_activate again from the commandline it just hangs.

If I disable the evms_activate line in checkfs & *reboot*, I can get evms_activate to run on the cli but I still get the segfault. 

This happens with both evms-2.2.1.ebuild and evms.2.2.2.ebuild 
Im useing ext3 on all partitions.


I've never seen that super.c BUG before....it could be some interaction between dm/evms and all gr_security/se_linux stuff.
My error follows:

kernel BUG at super.c:616!
invalid operand: 0000
CPU:    0
EIP:    0010:[<c01e23f0>]    Not tainted
EFLAGS: 00010282
eax: 00000024   ebx: 0000fe00   ecx: c01489d8   edx: 00000001
esi: 00000000   edi: f4e1c280   ebp: 00000000   esp: f4df7f30
ds: 0018   es: 0018   ss: 0018
Process evms_activate (pid: 1319, stackpage=f4df7000)
Stack: c03ab8a0 c0156d40 00000000 f67cc8a0 c031a127 0000fe00 f67cc920 f67cc8a0 
       c031cbdc f67cc8a0 00000000 00000000 800044d0 00000006 f4df6000 c031da3e 
       f8aa0000 00004000 00000134 c031cc50 f8aa0000 ffffffe7 c134fd06 f4e224a0 
Call Trace:    [<c031a127>] [<c031cbdc>] [<c031da3e>] [<c031cc50>] [<c01ed80e>]
  [<c01a64d7>]

Code: 0f 0b 68 02 68 81 3a c0 eb b5 ff 42 54 89 d6 e9 47 ff ff ff 



Also after the initial segfault I am unable to shutdown properly.
The shutdown process locks up after I see the following:
* Sending all process the KILL signal...
<6>md: recovery thread got woken up ...
md: recovery thread finished ...
Comment 1 solar (RETIRED) gentoo-dev 2004-01-27 16:43:40 UTC
Post your .config
Comment 2 REdOG 2004-01-27 18:01:47 UTC
Created attachment 24512 [details]
config

here it is
Comment 3 solar (RETIRED) gentoo-dev 2004-01-27 18:22:09 UTC
Well I can't make heads tails of why this is happening with the features you have set. Also from looking at the features you have set in your kernel I dont think your taking advantage of any aspect of it's hardening features other than ipsec which I think may be dropped from the .24 release.

scox perhaps you have some insights into this bug?
Comment 4 Andrea Luzzardi 2004-01-28 12:26:51 UTC
Could you try to remove some stuff (not needed for EVMS) from 
your Multi-device support?

In particular those :

- CONFIG_BLK_DEV_MD
- CONFIG_MD_LINEAR
- CONFIG_MD_RAID0
- CONFIG_MD_RAID1
- CONFIG_MD_RAID5
- CONFIG_MD_MULTIPATH
- CONFIG_BLK_DEV_DM_MIRROR
- CONFIG_BLK_DEV_DM_BBR
- CONFIG_BLK_DEV_DM_SPARSE

And also, please try CONFIG_BLK_DEV_LVM=y instead of m.
Comment 5 Andrea Luzzardi 2004-06-01 12:06:01 UTC
Closing this bug as it refers to a quite old version of hardened-sources.