This bug shows itself again, same bug as back with ecryptfs-utils #27, but it
might have more to do with the kernel (2.6.23.9).
BUG: unable to handle kernel paging request at virtual address 00626174
printing eip:
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: i915 des sha512 cryptd ablkcipher twofish_common aes cbc
md5 blowfish ecb blkcipher nf_conntrack_netlink nf_nat nf_conntrack
nfnetlink_queue nfnetlink_log nfnetlink ecryptfs msr cpuid microcode
firmware_class lp iTCO_wdt iTCO_vendor_support hangcheck_timer i2c_i801
i2c_dev pcspkr eepro100 parport_pc parport ehci_hcd uhci_hcd configfs
i810_audio v_midi sound ac97_codec sound_firmware
CPU: 0
EIP: 0060:[<e0153c65>] Tainted: P VLI
EFLAGS: 00010206 (2.6.23.9 #1)
EIP is at ecryptfs_find_daemon_id+0x31/0x46 [ecryptfs]
eax: 00000003 ebx: 00000064 ecx: 0000001e edx: 00626174
esi: cd4cfd30 edi: 00004447 ebp: 00000000 esp: cd4cfd14
ds: 007b es: 007b fs: 0000 gs: 0033 ss: 0068
Process ecryptfsd (pid: 17479, ti=cd4ce000 task=d7a9c540 task.ti=cd4ce000)
Stack: c7f39690 00000064 e01540ca c36d3bf4 de437018 00000001 cd4cfd50
c010c5f6
cee5a000 dc548900 cd4cfd54 cd4cfd90 e01546af cd4cfd54 00000054
cd4cfd68
00000000 d28c0800 00000010 dc548900 c025176c 00000246 d28c0800
c0250812
Call Trace:
[<e01540ca>] ecryptfs_process_helo+0x5d/0x14a [ecryptfs]
[<c010c5f6>] __wake_up_common+0x31/0x4f
[<e01546af>] ecryptfs_receive_nl_message+0x106/0x184 [ecryptfs]
[<c025176c>] netlink_data_ready+0x12/0x4b
[<c0250812>] netlink_sendskb+0x19/0x2f
[<c0251062>] netlink_unicast+0x17f/0x18b
[<c025174e>] netlink_sendmsg+0x23d/0x249
[<c02324a3>] sock_sendmsg+0xbb/0xd3
[<c011c14e>] __rcu_process_callbacks+0xb7/0x12e
[<c011dcb2>] autoremove_wake_function+0x0/0x33
[<c0113e44>] tasklet_action+0x32/0x52
[<c0113dac>] __do_softirq+0x35/0x75
[<c0103d64>] do_IRQ+0x55/0x69
[<c01a51e7>] copy_from_user+0x23/0x4f
[<c02327ca>] sys_sendto+0x118/0x138
[<c013458a>] __do_fault+0x27c/0x2aa
[<c01358a0>] handle_mm_fault+0x286/0x4dc
[<c02334da>] sys_socketcall+0x15e/0x242
[<c01023ce>] sysenter_past_esp+0x5f/0x85
[<c0290000>] pfkey_xfrm_state2msg+0x77/0xa3f
=======================
Code: 53 89 c3 69 c0 01 00 37 9e 89 d6 8b 15 c0 f2 15 e0 2b 0d d0 f2 15 e0
d3 e8 8b 14 82 eb 02 8b 12 85 d2 75 07 b8 ea ff ff ff eb 12 <8b> 02 0f 18 00
90 8d 42 f8 89 06 39 58 04 75 e3 31 c0 5b 5e c3
EIP: [<e0153c65>] ecryptfs_find_daemon_id+0x31/0x46 [ecryptfs] SS:ESP
0068:cd4cfd14
Reproduce:
1. Do a full usage of ecryptfs as root (mount ecryptfs, start ecryptfsd, use
the files,umount ecryptfs, kill ecryptfsd) 2,3 or 4 times. Maybe less needed
to trigger the bug, but it seems to happen on the 3rd and beyond most of the
time.
2. Mount ecryptfs as root, as if another user will be using the files
3. Add key to user keyring via ecryptfs-manager if needed
4. Start ecryptfsd as the user who will use the files
5. ecryptfsd will die right off with a kernel Oops
6. Reboot system/fsck because anything accessing the files or trying to start
ecryptfsd again will zombie
Temp. work-around:
Only use ecryptfs as root, only start ecryptfsd as root
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
eCryptfs-users mailing list
eCryptfs-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ecryptfs-users
