797 |
I was using PAGE_SIZE (4096 in many environments) |
I was using PAGE_SIZE (4096 in many environments) |
798 |
as the max length of any string data. |
as the max length of any string data. |
799 |
But for environments that have larger PAGE_SIZE, |
But for environments that have larger PAGE_SIZE, |
800 |
doing memset(ptr, 0, PAGE_SIZE) everytime is too wasteful. |
doing memset(ptr, 0, PAGE_SIZE) every time is too wasteful. |
801 |
|
|
802 |
Fix 2007/01/29 |
Fix 2007/01/29 |
803 |
|
|
1723 |
to see process information. The content of /proc/ccs/.execute_handler is |
to see process information. The content of /proc/ccs/.execute_handler is |
1724 |
identical to /proc/ccs/.process_status . |
identical to /proc/ccs/.process_status . |
1725 |
|
|
1726 |
Version 1.6.5 2008/11/11 Third anniversary release. |
Version 1.6.5 2008/11/11 Third anniversary release. |
1727 |
|
|
1728 |
Fix 2008/12/01 |
Fix 2008/12/01 |
1729 |
|
|
2243 |
/proc/ccs/profile (which results in not protecting the system at all), |
/proc/ccs/profile (which results in not protecting the system at all), |
2244 |
I added a check for PROFILE_VERSION= . |
I added a check for PROFILE_VERSION= . |
2245 |
|
|
2246 |
Version 1.7.1 2009/11/11 Fourth anniversary release. |
Version 1.7.1 2009/11/11 Fourth anniversary release. |
2247 |
|
|
2248 |
Fix 2009/11/13 |
Fix 2009/11/13 |
2249 |
|
|
2382 |
Since 1.7.0 , ccs_lport_reserved() was by error checking wrong port |
Since 1.7.0 , ccs_lport_reserved() was by error checking wrong port |
2383 |
number. As a result, "deny_autobind" keyword was not working as expected. |
number. As a result, "deny_autobind" keyword was not working as expected. |
2384 |
|
|
2385 |
Version 1.7.2 2010/04/01 Feature enhancement release. |
Version 1.7.2 2010/04/01 Feature enhancement release. |
2386 |
|
|
2387 |
Fix 2010/04/10 |
Fix 2010/04/10 |
2388 |
|
|
2441 |
permissions (e.g. allow_env PATH if symlink.target="/"), it triggered |
permissions (e.g. allow_env PATH if symlink.target="/"), it triggered |
2442 |
NULL pointer dereference. |
NULL pointer dereference. |
2443 |
|
|
2444 |
Fix 2010/07/29 |
Fix 2010/10/28 |
2445 |
|
|
2446 |
@ Change keyword syntax. |
@ Fix umount() pathname calculation. |
2447 |
|
|
2448 |
|
"mount --bind /path/to/file1 /path/to/file2" is legal. |
2449 |
|
Therefore, "umount /path/to/file2" is also legal. |
2450 |
|
Do not automatically append trailing '/' if pathname to be unmounted |
2451 |
|
does not end with '/'. |
2452 |
|
|
2453 |
|
@ Add preserve KABI compatibility option. (2.6 kernels only) |
2454 |
|
|
2455 |
|
TOMOYO needs "struct ccs_domain_info *" and "u32" for each |
2456 |
|
"struct task_struct". But embedding these variables into |
2457 |
|
"struct task_struct" breaks KABI for prebuilt kernel modules (which |
2458 |
|
means that you will need to rebuild prebuilt kernel modules). |
2459 |
|
|
2460 |
|
Since KABI is commonly used (compared to 5 years ago), asking users to |
2461 |
|
rebuild kernel modules which are not included in kernel package is no |
2462 |
|
longer preferable. Therefore, I added a new option that keeps |
2463 |
|
"struct task_struct" unmodified in order to keep KABI. |
2464 |
|
|
2465 |
|
Note that you have to use ccs-patch-2.6.\*.diff which patches |
2466 |
|
kernel/fork.c in order to use this option. Otherwise, TOMOYO will leak |
2467 |
|
memory whenever "struct task_struct" is released. |
2468 |
|
|
2469 |
|
@ Change directives. |
2470 |
|
|
2471 |
I removed "allow_" prefix from directives. New directives for files are |
I removed "allow_" prefix from directives. New directives for files are |
2472 |
prefixed with "file ". For example, "allow_read" changed to "file read", |
prefixed with "file ". For example, "allow_read" changed to "file read", |
2473 |
"allow_ioctl" changed to "file ioctl". New directive for "allow_network" |
"allow_ioctl" changed to "file ioctl". New directive for "allow_network |
2474 |
is "network". New directive for "allow_env" is "misc env". New directive |
TCP" is "network inet stream", "allow_network UDP" is "network inet |
2475 |
for "allow_signal" is "ipc signal". New directive for "allow_capability" |
dgram", "allow_network RAW" is "network inet raw". New directive for |
2476 |
is "capability". These directives correspond with keywords used by |
"allow_env" is "misc env". New directive for "allow_signal" is "ipc |
2477 |
profile's CONFIG lines. |
signal". New directive for "allow_capability" is "capability". These new |
2478 |
|
directives correspond with keywords used by profile's CONFIG lines. |
2479 |
|
|
2480 |
I removed "deny_rewrite" and "allow_rewrite" directives and introduced |
I removed "deny_rewrite" and "allow_rewrite" directives and introduced |
2481 |
"file append" directive. Thus, permission for open(O_WRONLY | O_APPEND) |
"file append" directive. Thus, permission for open(O_WRONLY | O_APPEND) |
2491 |
check requires hooks in filesystem part while almost all hooks for |
check requires hooks in filesystem part while almost all hooks for |
2492 |
filesystem part have moved to LSM by Linux 2.6.34. |
filesystem part have moved to LSM by Linux 2.6.34. |
2493 |
|
|
2494 |
@ Distinguish send() and recv() operations for UDP and IP protocols. |
New directive for "execute_handler" is "task auto_execute_handler", |
2495 |
|
"denied_execute_handler" is "task denied_execute_handler". |
2496 |
|
|
2497 |
Until now, it was impossible for UDP and IP protocols to allow either |
@ Distinguish send() and recv() operations. |
2498 |
|
|
2499 |
|
Until now, it was impossible for UDP and IP sockets to allow either |
2500 |
only sending or only receiving because permissions were aggregated with |
only sending or only receiving because permissions were aggregated with |
2501 |
"connect" keyword. I broke "connect" keyword into "send" and "recv" |
"connect" keyword. I broke "connect" keyword into "send" and "recv" |
2502 |
keywords so that you can keep access control for send() operation enabled |
keywords so that you can keep access control for send() operation enabled |
2503 |
when you have to turn access control for recv() operation off due to |
when you have to disable access control for recv() operation due to |
2504 |
application breakage by filtering incoming datagram. |
application breakage by discarding incoming datagram. |
2505 |
|
|
2506 |
@ Wait for next connection/datagram if current connection/datagram was |
@ Add Unix domain socket restriction support. |
|
discarded. |
|
2507 |
|
|
2508 |
Regarding "network TCP accept", "network UDP recv", "network RAW recv" |
Until now, it was possible to restrict only inet domain sockets (i.e. |
2509 |
keywords, I modified to wait for next connection/datagram if current |
TCP/UDP/RAW). I added restriction for Unix domain sockets (i.e. stream/ |
2510 |
connection/datagram was discarded. LSM hooks for these keywords are |
dgram/seqpacket). New directive "network unix" is added as well as |
2511 |
currently missing because this behavior may break applications. |
"network inet" directive. |
|
If you found applications broken by this behavior, you can set |
|
|
CONFIG::network::inet_tcp_accept and/or CONFIG::network::inet_udp_recv |
|
|
and/or CONFIG::network::inet_raw_recv to mode=disabled in order to |
|
|
disable filtering for incoming connection/datagram. |
|
2512 |
|
|
2513 |
@ Allow specifying multiple permissions in a line. |
@ Allow specifying multiple permissions in a line. |
2514 |
|
|
2547 |
|
|
2548 |
@ Add a new keyword "any" for domain transition control. |
@ Add a new keyword "any" for domain transition control. |
2549 |
|
|
2550 |
To be able to make it easier to apply execute_handler on each domain, |
To be able to make it easier to apply auto_execute_handler on each |
2551 |
I added "any" keyword to domain transition control keywords. Now, |
domain, I added "any" keyword to domain transition control keywords. Now, |
2552 |
"initialize_domain /usr/sbin/sshd" changed to |
"initialize_domain /usr/sbin/sshd" changed to |
2553 |
"initialize_domain /usr/sbin/sshd from any" and |
"initialize_domain /usr/sbin/sshd from any" and |
2554 |
"keep_domain <kernel> /usr/sbin/sshd /bin/bash" changed to |
"keep_domain <kernel> /usr/sbin/sshd /bin/bash" changed to |
2555 |
"keep_domain any from <kernel> /usr/sbin/sshd /bin/bash". |
"keep_domain any from <kernel> /usr/sbin/sshd /bin/bash". |
2556 |
|
|
2557 |
"keep_domain /path/to/execute_handler from any" will allow you to apply |
"keep_domain /path/to/auto_execute_handler from any" will allow you to |
2558 |
execute_handler for any domains without creating domains for |
apply auto_execute_handler for any domains without creating domains for |
2559 |
execute_handler. |
auto_execute_handler. |
2560 |
|
|
2561 |
@ Change buffering mode for reading policy. |
@ Change buffering mode for reading policy. |
2562 |
|
|
2577 |
the exception policy and specify "use_group 0" from the domains in the |
the exception policy and specify "use_group 0" from the domains in the |
2578 |
domain policy. |
domain policy. |
2579 |
|
|
2580 |
"ignore_global_allow_read" and "ignore_global_allow_env" keywords were |
"ignore_global_allow_read" and "ignore_global_allow_env" directives were |
2581 |
removed from domain policy and "use_group" keyword was added. |
removed from domain policy and "use_group" keyword was added. |
2582 |
|
|
2583 |
|
@ Remove "if" and "; set" keyword. |
2584 |
|
|
2585 |
|
I removed need for specifying these keyword. |
2586 |
|
You can simply specify like below. |
2587 |
|
|
2588 |
|
file read /etc/shadow task.uid=0 |
2589 |
|
|
2590 |
|
@ Remove "file_pattern" keyword. |
2591 |
|
|
2592 |
|
I removed "file_pattern" keyword because it is impossible to predefine |
2593 |
|
all possible pathname patterns. Also, learning pathnames using incomplete |
2594 |
|
patterns makes it difficult to later replace using "path_group" keyword. |
2595 |
|
|
2596 |
|
@ Replace verbose= parameter with statistic interface. |
2597 |
|
|
2598 |
|
Since it is noisy if a lot of policy violation messages are printed, |
2599 |
|
I removed printk(). To be able to check whether policy violation occurred |
2600 |
|
or not, I introduced /proc/ccs/stat interface which counts number of |
2601 |
|
policy violations occurred. You can firstly check /proc/ccs/stat and then |
2602 |
|
check /proc/ccs/reject_log . |
2603 |
|
|
2604 |
|
@ Remove global preference. |
2605 |
|
|
2606 |
|
I removed global preference in order to make code simpler. |
2607 |
|
|
2608 |
@ Allow controlling generation of access granted logs for per an entry |
@ Allow controlling generation of access granted logs for per an entry |
2609 |
basis. |
basis. |
2610 |
|
|
2611 |
I added per-entry flag which controls generation of grant logs because |
I added per-entry flag which controls generation of grant logs because |
2612 |
Xen and KVM issues ioctl requests so frequently. For example, |
Xen and KVM issues ioctl requests so frequently. For example, |
2613 |
|
|
2614 |
file ioctl /dev/null 0x5401 ; set audit=no |
file ioctl /dev/null 0x5401 grant_log=no |
2615 |
|
|
2616 |
will suppress /proc/ccs/grant_log even if profile says grant_log=yes . |
will suppress /proc/ccs/grant_log even if preference says grant_log=yes . |
2617 |
|
|
2618 |
file ioctl /dev/null 0x5401 ; set audit=yes |
file ioctl /dev/null 0x5401 grant_log=yes |
2619 |
|
|
2620 |
will generate /proc/ccs/grant_log even if profile says grant_log=no . |
will generate /proc/ccs/grant_log even if preference says grant_log=no . |
2621 |
|
|
2622 |
file ioctl /dev/null 0x5401 |
file ioctl /dev/null 0x5401 |
2623 |
|
|
2624 |
will generate /proc/ccs/grant_log only if profile says grant_log=yes . |
will generate /proc/ccs/grant_log only if preference says grant_log=yes . |
2625 |
|
|
2626 |
This flag is intended for frequently accessed resources like |
This flag is intended for frequently accessed resources like |
2627 |
|
|
2628 |
file read /var/www/html/\{\*\}/\*.html ; set audit=no |
file read /var/www/html/\{\*\}/\*.html grant_log=no |
2629 |
|
|
2630 |
. |
. |
2631 |
|
|
2632 |
|
@ Automatically create domain by execve() even if enforcing mode. |
2633 |
|
|
2634 |
|
Until now, new domains are not created if the domain was not defined and |
2635 |
|
current domain is enforcing mode ("CONFIG::file::execute=enforcing"). |
2636 |
|
|
2637 |
|
To be able to restrict shell session without using "keep_domain", |
2638 |
|
I changed to create new domains automatically even if current domain is |
2639 |
|
enforcing mode. |
2640 |
|
|
2641 |
|
@ Replace "task.state" with "auto_domain_transition". |
2642 |
|
|
2643 |
|
task.state is difficult to use. Thus, I replaced task.state with |
2644 |
|
auto_domain_transition which performs domain transition instead of |
2645 |
|
changing current process's state variables. |
2646 |
|
|
2647 |
|
If domain transition failed, current process will be killed by SIGKILL |
2648 |
|
signal. This should not happen in normal circumstances, for you know the |
2649 |
|
domain to transit to and thereby you will define the domain beforehand |
2650 |
|
when you use "auto_domain_transition" keyword. |
2651 |
|
|
2652 |
|
@ Replace "allow_transit" with "task manual_domain_transition". |
2653 |
|
|
2654 |
|
I changed this directive to specify absolute domainname (e.g. |
2655 |
|
"<kernel> /usr/sbin/httpd //app=cgi1\040id=10000") rather than virtual |
2656 |
|
pathname (e.g. "//app=cgi1\040id=10000") because you know the domain to |
2657 |
|
transit to and thereby you will define the domain beforehand when you use |
2658 |
|
"task manual_domain_transition" directive. |
2659 |
|
|
2660 |
|
This change allows you to jump to arbitrary domain. |
2661 |
|
|
2662 |
|
Note that this change also reverts "Change /proc/ccs/info/self_domain ." |
2663 |
|
made on 2006/10/24. Now, 'cat < /proc/ccs/info/self_domain' will act like |
2664 |
|
'cat /proc/ccs/info/self_domain'. Programs depending on old assumption |
2665 |
|
need to be updated. |
2666 |
|
|
2667 |
|
@ Add "task auto_domain_transition". |
2668 |
|
|
2669 |
|
This is similar to "task manual_domain_transition", but is automatically |
2670 |
|
applied whenever conditions are met. For example, |
2671 |
|
|
2672 |
|
task auto_domain_transition <kernel> //./non-root task.uid!=0 |
2673 |
|
|
2674 |
|
will automatically jump to "<kernel> //./non-root" domain if current |
2675 |
|
process's UID is not 0 whereas |
2676 |
|
|
2677 |
|
task manual_domain_transition <kernel> //./non-root task.uid!=0 |
2678 |
|
|
2679 |
|
will jump to "<kernel> //./non-root" domain if current process's UID is |
2680 |
|
not 0 and current process wrote "<kernel> //./non-root" to |
2681 |
|
/proc/ccs/self_domain interface. |
2682 |
|
|
2683 |
|
If domain transition failed, current process will be killed by SIGKILL |
2684 |
|
signal. |
2685 |
|
|
2686 |
@ Optimize for object's size. |
@ Optimize for object's size. |
2687 |
|
|
2688 |
I merged similar code in order to reduce object's filesize. |
I merged similar code in order to reduce object's filesize. |
2689 |
|
|
2690 |
|
Version 1.8.0 2010/11/11 Fifth anniversary release. |
2691 |
|
|
2692 |
|
Fix 2010/12/01 |
2693 |
|
|
2694 |
|
@ Use same interface for audit logs. |
2695 |
|
|
2696 |
|
To be able to perform fine grained filtering by /usr/sbin/ccs-auditd , |
2697 |
|
I merged /proc/ccs/grant_log and /proc/ccs/reject_log as |
2698 |
|
/proc/ccs/audit and added granted=yes or granted=no to audit logs. |
2699 |
|
|
2700 |
|
Fix 2010/12/17 |
2701 |
|
|
2702 |
|
@ Split ccs_null_security into ccs_default_security and ccs_oom_security. |
2703 |
|
|
2704 |
|
ccs_null_security is used by preserve KABI compatibility option and is |
2705 |
|
used for providing default values against threads which have not yet |
2706 |
|
allocated memory for their security contexts. |
2707 |
|
|
2708 |
|
If current thread failed to allocate memory for current thread's security |
2709 |
|
context, current thread uses ccs_null_security. Since current thread is |
2710 |
|
allowed to modify current thread's security context, current thread might |
2711 |
|
modify ccs_null_security which should not be modified for any reason. |
2712 |
|
|
2713 |
|
Therefore, I split ccs_null_security into ccs_default_security and |
2714 |
|
ccs_oom_security and use ccs_oom_security when current thread failed to |
2715 |
|
allocate memory for current thread's security context. |
2716 |
|
|
2717 |
|
Threads which do not share ccs_oom_security are not affected by threads |
2718 |
|
which share ccs_oom_security. Threads which share ccs_oom_security will |
2719 |
|
experience temporary inconsistency, but such threads are about to be |
2720 |
|
killed by SIGKILL signal. |
2721 |
|
|
2722 |
|
Fix 2011/01/11 |
2723 |
|
|
2724 |
|
@ Use filesystem name for unnamed devices when vfsmount is missing. |
2725 |
|
|
2726 |
|
"Change pathname for non-rename()able filesystems." changed to use |
2727 |
|
"$fsname:" if the filesystem does not support rename() operation and |
2728 |
|
"dev($major,$minor):" otherwise when vfsmount is missing. But it turned |
2729 |
|
out that it is useless to use "dev($major,$minor):" for unnamed devices |
2730 |
|
(filesystems with $major == 0). Thus, I changed to use "$fsname:" rather |
2731 |
|
than "dev($major,$minor):" for filesystems with $major == 0 when vfsmount |
2732 |
|
is missing. |
2733 |
|
|
2734 |
|
Fix 2011/02/07 |
2735 |
|
|
2736 |
|
@ Fix infinite loop bug when reading /proc/ccs/audit or /proc/ccs/query . |
2737 |
|
|
2738 |
|
In ccs_flush(), head->r.w[0] holds pointer to string data to be printed. |
2739 |
|
But head->r.w[0] was updated only when the string data was partially |
2740 |
|
printed (because head->r.w[0] will be updated by head->r.w[1] later if |
2741 |
|
completely printed). However, regarding /proc/ccs/audit and |
2742 |
|
/proc/ccs/query , an additional '\0' is printed after the string data was |
2743 |
|
completely printed. But if free space for read buffer became 0 before |
2744 |
|
printing the additional '\0', ccs_flush() was returning without updating |
2745 |
|
head->r.w[0]. As a result, ccs_flush() forever reprints already printed |
2746 |
|
string data. |
2747 |
|
|
2748 |
|
Fix 2011/03/01 |
2749 |
|
|
2750 |
|
@ Run garbage collector without waiting for /proc/ccs/ users. |
2751 |
|
|
2752 |
|
Currently TOMOYO holds SRCU lock upon open() and releases it upon close() |
2753 |
|
because list elements stored in the "struct ccs_io_buffer" instances are |
2754 |
|
accessed until close() is called. However, such SRCU usage causes lockdep |
2755 |
|
to complain about leaving the kernel with SRCU lock held. Therefore, |
2756 |
|
I changed to hold/release SRCU upon each read()/write() by selectively |
2757 |
|
deferring kfree() by keeping track of the "struct ccs_io_buffer" |
2758 |
|
instances. |
2759 |
|
|
2760 |
|
Fix 2011/03/05 |
2761 |
|
|
2762 |
|
@ Support built-in policy configuration. |
2763 |
|
|
2764 |
|
To be able to start using enforcing mode from the early stage of boot |
2765 |
|
sequence, I added support for built-in policy configuration and |
2766 |
|
activating access control without calling external policy loader program. |
2767 |
|
|
2768 |
|
This will be useful for systems where operations which can lead to the |
2769 |
|
hijacking of the boot sequence are needed before loading the policy. |
2770 |
|
For example, you can activate immediately after loading the fixed part of |
2771 |
|
policy which will allow only operations needed for mounting a partition |
2772 |
|
which contains the variant part of policy and verifying (e.g. running GPG |
2773 |
|
check) and loading the variant part of policy. Since you can start using |
2774 |
|
enforcing mode from the beginning, you can reduce the possibility of |
2775 |
|
hijacking the boot sequence. |
2776 |
|
|
2777 |
|
Fix 2011/03/10 |
2778 |
|
|
2779 |
|
@ Remove /proc/ccs/meminfo interface. |
2780 |
|
|
2781 |
|
Please use /proc/ccs/stat interface instead. |
2782 |
|
|
2783 |
|
Fix 2011/03/15 |
2784 |
|
|
2785 |
|
@ Pack policy when printing via /proc/ccs/ interface. |
2786 |
|
|
2787 |
|
The kernel side is ready for accepting packed input like |
2788 |
|
|
2789 |
|
file read/write/execute /path/to/file |
2790 |
|
|
2791 |
|
but was using unpacked output like |
2792 |
|
|
2793 |
|
file read /path/to/file |
2794 |
|
file write /path/to/file |
2795 |
|
file execute /path/to/file |
2796 |
|
|
2797 |
|
because most of userland tools were not ready for accepting packed input. |
2798 |
|
|
2799 |
|
The advantages of using packed policy are that it makes policy files |
2800 |
|
smaller and it speeds up loading/saving policy files. |
2801 |
|
|
2802 |
|
Since most of userland tools are ready for accepting packed input by now, |
2803 |
|
I changed to use packed policy for both input and output. |
2804 |
|
|
2805 |
|
Fix 2011/03/31 |
2806 |
|
|
2807 |
|
@ Fix conditional policy parsing. |
2808 |
|
|
2809 |
|
Since exec.realpath= and symlink.target= accept path_group, |
2810 |
|
symlink.target="@foo" was by error parsed as symlink.target=@foo . |
2811 |
|
|
2812 |
|
@ Serialize updating profile's comment line. |
2813 |
|
|
2814 |
|
We need to serialize when updating COMMENT= line in /proc/ccs/profile . |
2815 |
|
|
2816 |
|
Version 1.8.1 2011/04/01 Usability enhancement with "Zettai, Daijoubudayo" release! |
2817 |
|
|
2818 |
|
Fix 2011/04/03 |
2819 |
|
|
2820 |
|
@ Fix fcntl(F_SETFL, O_APPEND) handling. |
2821 |
|
|
2822 |
|
Since 1.8.0, TOMOYO was by error checking "file write" permission rather |
2823 |
|
than "file append" permission when changing file's writing mode from |
2824 |
|
"overwriting" to "append". |
2825 |
|
|
2826 |
|
This error should impact little (except CentOS 6.0 kernels) because once |
2827 |
|
a file was opened for "overwriting" mode, changing that file to "append" |
2828 |
|
mode cannot undo overwriting the file. Regarding CentOS 6.0 kernels, |
2829 |
|
due to different ACC_MODE definition, TOMOYO was by error needlessly |
2830 |
|
checking "file read" permission when fcntl() was requested. |
2831 |
|
|
2832 |
|
Fix 2011/04/20 |
2833 |
|
|
2834 |
|
@ Remove unused "struct inode *" parameter from hooks. |
2835 |
|
|
2836 |
|
Since pre-vfs functions were removed on 2010/09/18, "struct inode *" |
2837 |
|
parameter which was used for checking parent directory's DAC permission |
2838 |
|
is no longer used. |
2839 |
|
|
2840 |
|
Note that "struct ccsecurity_operations ccsecurity_ops" has changed. |
2841 |
|
Loadable kernel modules that depends on it need to be rebuilt. |
2842 |
|
|
2843 |
|
Fix 2011/05/05 |
2844 |
|
|
2845 |
|
@ Fix wrong profile number in audit logs for "misc env" permission. |
2846 |
|
|
2847 |
|
Profile number used for "file execute" permission was by error reused |
2848 |
|
when generating audit logs for "misc env" permission. |
2849 |
|
|
2850 |
|
Fix 2011/05/11 |
2851 |
|
|
2852 |
|
@ Fix wrong domainname validation. |
2853 |
|
|
2854 |
|
"<kernel>" + "/foo/\" + "/bar" was by error checked when |
2855 |
|
"<kernel> /foo/\* /bar" was given. As a result, legal domainnames like |
2856 |
|
"<kernel> /foo/\* /bar" are rejected. |
2857 |
|
|
2858 |
|
Fix 2011/06/06 |
2859 |
|
|
2860 |
|
@ Add policy namespace support. |
2861 |
|
|
2862 |
|
To be able to use TOMOYO in LXC environments, I introduced policy |
2863 |
|
namespace. Each policy namespace has its own set of domain policy, |
2864 |
|
exception policy and profiles, which are all independent of other |
2865 |
|
namespaces. |
2866 |
|
|
2867 |
|
@ Remove CONFIG_CCSECURITY_BUILTIN_INITIALIZERS option. |
2868 |
|
|
2869 |
|
From now on, exception policy and manager need to be able to handle |
2870 |
|
policy namespace (which is a <$namespace> prefix added to each line). |
2871 |
|
Thus, space-separated list for CONFIG_CCSECURITY_BUILTIN_INITIALIZERS is |
2872 |
|
no longer suitable for handling policy namespace. |
2873 |
|
|
2874 |
|
Fix 2011/06/10 |
2875 |
|
|
2876 |
|
@ Allow specifying trigger for activation. |
2877 |
|
|
2878 |
|
To be able to use TOMOYO under systemd environments where init= parameter |
2879 |
|
is used, I changed to allow overriding the trigger for calling external |
2880 |
|
policy loader and activating MAC via kernel command line options. |
2881 |
|
|
2882 |
|
Fix 2011/06/14 |
2883 |
|
|
2884 |
|
@ Remove unused "struct inode *" parameter from ccs-patch-\*.diff . |
2885 |
|
|
2886 |
|
To follow changes I made on 2011/04/20, I removed "struct inode *" from |
2887 |
|
ccs_mknod_permission(), ccs_mkdir_permission(), ccs_rmdir_permission(), |
2888 |
|
ccs_unlink_permission(), ccs_symlink_permission(), ccs_link_permission(), |
2889 |
|
ccs_rename_permission() that are called from fs/namei.c |
2890 |
|
net/unix/af_unix.c include/linux/security.c security/security.c . |
2891 |
|
If you have your own ccs-patch-*.diff , please update accordingly. |
2892 |
|
|
2893 |
|
Version 1.8.2 2011/06/20 Usability enhancement release. |
2894 |
|
|
2895 |
|
Fix 2011/07/07 |
2896 |
|
|
2897 |
|
@ Remove /proc/ccs/.domain_status interface. |
2898 |
|
|
2899 |
|
Writing to /proc/ccs/.domain_status can be emulated by |
2900 |
|
|
2901 |
|
( echo "select " $domainname; echo "use_profile " $profile ) | |
2902 |
|
/usr/sbin/ccs-loadpolicy -d |
2903 |
|
|
2904 |
|
and reading from /proc/ccs/.domain_status can be emulated by |
2905 |
|
|
2906 |
|
grep -A 1 '^<' /proc/ccs/domain_policy | |
2907 |
|
awk ' { if ( domainname == "" ) { if ( substr($1, 1, 1) == "<" ) |
2908 |
|
domainname = $0; } else if ( $1 == "use_profile" ) { |
2909 |
|
print $2 " " domainname; domainname = ""; } } ; ' |
2910 |
|
|
2911 |
|
. Since this interface is used by only /usr/sbin/ccs-setprofile , |
2912 |
|
remove this interface by updating /usr/sbin/ccs-setprofile . |
2913 |
|
|
2914 |
|
Fix 2011/07/09 |
2915 |
|
|
2916 |
|
@ Fix /proc/ccs/stat parser. |
2917 |
|
|
2918 |
|
For optimization, I changed to use simple_strtoul() rather than sscanf() |
2919 |
|
in ccs_write_stat(). But it caused parsing failure if space is inserted |
2920 |
|
before value (e.g. "Memory used by policy: $value"). |
2921 |
|
|
2922 |
|
Fix 2011/07/13 |
2923 |
|
|
2924 |
|
@ Accept "::" notation for IPv6 address. |
2925 |
|
|
2926 |
|
In order to add network access restriction to TOMOYO 2.4, I backported |
2927 |
|
routines for parsing/printing IPv4/IPv6 address from kernel 3.0 into |
2928 |
|
TOMOYO 1.8.2. |
2929 |
|
Now, IPv6 address accepts "::1" instead of "0:0:0:0:0:0:0:1". |
2930 |
|
|
2931 |
|
Fix 2011/09/03 |
2932 |
|
|
2933 |
|
@ Avoid race when retrying "file execute" permission check. |
2934 |
|
|
2935 |
|
There was a race window that the pathname which is subjected to |
2936 |
|
"file execute" permission check when retrying via supervisor's decision |
2937 |
|
because the pathname was recalculated upon retry. Though, there is an |
2938 |
|
inevitable race window even without supervisor, for we have to calculate |
2939 |
|
the symbolic link's pathname from "struct linux_binprm"->filename rather |
2940 |
|
than from "struct linux_binprm"->file because we cannot back calculate |
2941 |
|
the symbolic link's pathname from the dereferenced pathname. |
2942 |
|
|
2943 |
|
@ Remove unneeded daemonize(). |
2944 |
|
|
2945 |
|
Garbage collector thread is created using kthread_create() since 2.6.7. |
2946 |
|
Kernel threads created by kthread_create() does not need to call |
2947 |
|
daemonize(). |
2948 |
|
|
2949 |
|
Fix 2011/09/16 |
2950 |
|
|
2951 |
|
@ Allow specifying domain transition preference. |
2952 |
|
|
2953 |
|
I got an opinion that it is difficult to use exception policy's domain |
2954 |
|
transition control directives because they need to match the pathname |
2955 |
|
specified to "file execute" directives. For example, if "file execute |
2956 |
|
/bin/\*\-ls\-cat" is given, corresponding domain transition control |
2957 |
|
directive needs to be like "no_keep_domain /bin/\*\-ls\-cat from any". |
2958 |
|
|
2959 |
|
To solve this difficulty, I introduced optional argument that supersedes |
2960 |
|
exception policy's domain transition control directives. |
2961 |
|
|
2962 |
|
file execute /bin/ls keep exec.realpath="/bin/ls" exec.argv[0]="ls" |
2963 |
|
file execute /bin/cat keep exec.realpath="/bin/cat" exec.argv[0]="cat" |
2964 |
|
file execute /bin/\*\-ls\-cat child |
2965 |
|
file execute /usr/sbin/httpd <apache> exec.realpath="/usr/sbin/httpd" exec.argv[0]="/usr/sbin/httpd" |
2966 |
|
|
2967 |
|
This argument allows transition to different domains based on conditions. |
2968 |
|
|
2969 |
|
<kernel> /usr/sbin/sshd |
2970 |
|
file execute /bin/bash <kernel> /usr/sbin/sshd //batch-session exec.argc=2 exec.argv[1]="-c" |
2971 |
|
file execute /bin/bash <kernel> /usr/sbin/sshd //root-session task.uid=0 |
2972 |
|
file execute /bin/bash <kernel> /usr/sbin/sshd //nonroot-session task.uid!=0 |
2973 |
|
|
2974 |
|
Fix 2011/09/25 |
2975 |
|
|
2976 |
|
@ Simplify garbage collector. |
2977 |
|
|
2978 |
|
It turned out that use of batched processing tends to choke garbage |
2979 |
|
collector when certain pattern of entries are queued. Thus, I replaced it |
2980 |
|
with sequential processing. |
2981 |
|
|
2982 |
|
Version 1.8.3 2011/09/29 Usability enhancement release. |
2983 |
|
|
2984 |
|
Fix 2011/10/24 |
2985 |
|
|
2986 |
|
@ Fix incomplete read after seek. |
2987 |
|
|
2988 |
|
ccs_flush() tries to flush data to be read as soon as possible. |
2989 |
|
ccs_select_domain() (which is called by write()) enqueues data which |
2990 |
|
meant to be read by next read(), but previous read()'s read buffer's |
2991 |
|
size was not cleared. As a result, since 1.8.0, sequence like |
2992 |
|
|
2993 |
|
char *cp = "select global-pid=1\n"; |
2994 |
|
read(fd, buf1, sizeof(buf1)); |
2995 |
|
write(fd, cp, strlen(cp)); |
2996 |
|
read(fd, buf2, sizeof(buf2)); |
2997 |
|
|
2998 |
|
causes enqueued data to be flushed to buf1 rather than buf2. |
2999 |
|
|
3000 |
|
@ Use query id for reaching target process's domain policy. |
3001 |
|
|
3002 |
|
Use query id for reaching target process's domain policy rather than |
3003 |
|
target process's global PID. This is for synchronizing with TOMOYO 2.x, |
3004 |
|
but this change makes /usr/sbin/ccs-queryd more reliable because the |
3005 |
|
kernel will return empty domain policy when the query has expired before |
3006 |
|
ccs-queryd reaches target process's domain policy. |
3007 |
|
|
3008 |
|
@ Fix quota counting. |
3009 |
|
|
3010 |
|
"task manual_domain_transition" should not be counted for quota as with |
3011 |
|
"task auto_domain_transition"/"task auto_execute_handler"/ |
3012 |
|
"task denied_execute_handler" because these are not appended by learning |
3013 |
|
mode. |
3014 |
|
|
3015 |
|
Fix 2011/11/11 |
3016 |
|
|
3017 |
|
@ Optimize for object's size. |
3018 |
|
|
3019 |
|
I rearranged functions/variables into three groups in order to reduce |
3020 |
|
object's filesize. Also, I added kernel config options for reducing more |
3021 |
|
by excluding unnecessary functionality. |
3022 |
|
|
3023 |
|
Fix 2011/11/18 |
3024 |
|
|
3025 |
|
@ Fix kernel config mapping error. |
3026 |
|
|
3027 |
|
Due to a typo in ccs_p2mac definition, mode for CONFIG::file::execute was |
3028 |
|
by error used when checking "file getattr" permission. Most users will |
3029 |
|
not be affected by this error because CONFIG::file::execute and |
3030 |
|
CONFIG::file::getattr are by default configured to use CONFIG::file or |
3031 |
|
CONFIG settings. |
3032 |
|
|
3033 |
|
Fix 2011/12/13 |
3034 |
|
|
3035 |
|
@ Follow __d_path() behavior change. (Only 2.6.36 and later) |
3036 |
|
|
3037 |
|
The behavior of __d_path() has changed in 3.2-rc5. __d_path() now returns |
3038 |
|
NULL when the pathname cannot be calculated. You must update to this |
3039 |
|
version when using with 3.2-rc5 and later kernels, or the kernel will |
3040 |
|
panic because ccs_get_absolute_path() triggers NULL pointer dereference. |
3041 |
|
|
3042 |
|
The patch that changed the behavior of __d_path() might be backported to |
3043 |
|
2.6.36 to 3.1 kernels. You must update to this version if the patch was |
3044 |
|
backported, or you will experience the kernel panic as with 3.2-rc5. |
3045 |
|
|
3046 |
|
The patch that changed the behavior of __d_path() also changed the way of |
3047 |
|
handling pathnames under lazy-unmounted directory. Until now, TOMOYO was |
3048 |
|
using incomplete pathnames returned by __d_path() when the pathname is |
3049 |
|
under lazy-unmounted directory. But from now on, TOMOYO uses different |
3050 |
|
pathnames returned by ccs_get_local_path() when the pathname is under |
3051 |
|
lazy-unmounted directory (because __d_path() no longer returns it). |
3052 |
|
|
3053 |
|
Since applications unlikely do lazy unmounts, requesting pathnames under |
3054 |
|
lazy-unmounted directory should not happen unless the administrator |
3055 |
|
explicitly does lazy unmounts. But pathnames which is defined for such |
3056 |
|
conditions in the policy file (if any) will need to be rewritten. |
3057 |
|
|
3058 |
|
Fix 2012/01/20 |
3059 |
|
|
3060 |
|
@ Follow changes in 3.3-rc1. |
3061 |
|
|
3062 |
|
Use umode_t rather than mode_t. |
3063 |
|
Remove ipv6_addr_copy() usage. |
3064 |
|
|
3065 |
|
Fix 2012/02/25 |
3066 |
|
|
3067 |
|
@ Follow changes in linux-next. |
3068 |
|
|
3069 |
|
UMH_WAIT_PROC constant (currently 1) is scheduled for renumbering in 3.4. |
3070 |
|
|
3071 |
|
Use UMH_WAIT_PROC constant instead of hardcoded constant in preparation |
3072 |
|
for backporting call_usermodehelper() related changes. If renumbering was |
3073 |
|
backported, you will start experiencing the kernel panic upon execution |
3074 |
|
of external policy loader (i.e. /sbin/ccs-init), for the kernel will no |
3075 |
|
longer wait for completion of external policy loader process. |
3076 |
|
|
3077 |
|
Although I changed to use UMH_WAIT_PROC constant, this change could fail |
3078 |
|
to detect renumbering in 2.6.22 and earlier kernels, for UMH_WAIT_PROC |
3079 |
|
constant is currently available to only 2.6.23 and later kernels. If you |
3080 |
|
started to experience the kernel panic, please check whether renumbering |
3081 |
|
was backported or not. |
3082 |
|
|
3083 |
|
Fix 2012/02/29 |
3084 |
|
|
3085 |
|
@ Fix mount flags checking order. |
3086 |
|
|
3087 |
|
Userspace can pass in arbitrary combinations of MS_* flags to mount(). |
3088 |
|
|
3089 |
|
If both MS_BIND and one of MS_SHARED/MS_PRIVATE/MS_SLAVE/MS_UNBINDABLE |
3090 |
|
are passed, device name which should be checked for MS_BIND was not |
3091 |
|
checked because MS_SHARED/MS_PRIVATE/MS_SLAVE/MS_UNBINDABLE had higher |
3092 |
|
priority than MS_BIND. |
3093 |
|
|
3094 |
|
If both one of MS_BIND/MS_MOVE and MS_REMOUNT are passed, device name |
3095 |
|
which should not be checked for MS_REMOUNT was checked because MS_BIND/ |
3096 |
|
MS_MOVE had higher priority than MS_REMOUNT. |
3097 |
|
|
3098 |
|
Fix these bugs by changing priority to MS_REMOUNT -> MS_BIND -> |
3099 |
|
MS_SHARED/MS_PRIVATE/MS_SLAVE/MS_UNBINDABLE -> MS_MOVE as with do_mount() |
3100 |
|
does. Also, I changed to unconditionally return -EINVAL if more than one |
3101 |
|
of MS_SHARED/MS_PRIVATE/MS_SLAVE/MS_UNBINDABLE is passed so that TOMOYO |
3102 |
|
will not generate inaccurate audit logs, for commit 7a2e8a8f "VFS: Sanity |
3103 |
|
check mount flags passed to change_mnt_propagation()" clarified that |
3104 |
|
these flags must be exclusively passed. |
3105 |
|
|
3106 |
|
Fix 2012/03/08 |
3107 |
|
|
3108 |
|
@ Allow returning other errors when ptrace permission cannot be checked. |
3109 |
|
|
3110 |
|
Currently -EPERM is returned when ccs_ptrace_permission() returned an |
3111 |
|
error code. I changed to return return value from ccs_ptrace_permission() |
3112 |
|
so that we can return -ESRCH when target process was not found. |
3113 |
|
|
3114 |
|
Fix 2012/03/16 |
3115 |
|
|
3116 |
|
@ Return appropriate value to poll(). |
3117 |
|
|
3118 |
|
Return POLLIN | POLLRDNORM | POLLOUT | POLLWRNORM if ready to read/write, |
3119 |
|
POLLOUT | POLLWRNORM otherwise. |
3120 |
|
|
3121 |
|
Fix 2012/04/22 |
3122 |
|
|
3123 |
|
@ Readd RHEL_MINOR/AX_MINOR checks. |
3124 |
|
|
3125 |
|
This check was added in revision 2346 and was removed in revision 4084. |
3126 |
|
|
3127 |
|
Add it back in order to support RHEL 5.0, 5.1, 5.2 kernels. |
3128 |
|
|
3129 |
|
@ Fix skb_kill_datagram() for kernels 2.6.0 - 2.6.11. |
3130 |
|
|
3131 |
|
Commit 208d8984 "[IPV4]: Fix BUG() in 2.6.x, udp_poll(), fragments + |
3132 |
|
CONFIG_HIGHMEM" clarified that skb_kill_datagram() should use |
3133 |
|
spin_lock_bh()/spin_unlock_bh() rather than |
3134 |
|
spin_lock_irq()/spin_unlock_irq(). |
3135 |
|
|
3136 |
|
RHEL 4.9 (2.6.9) kernel has that patch backported. So do I. |
3137 |
|
|
3138 |
|
@ Fix missing locks for RHEL 5.2-5.8 kernels. |
3139 |
|
|
3140 |
|
Since RHEL 5.2 and later kernels have backported commit 95766fff |
3141 |
|
"[UDP]: Add memory accounting." patch, TOMOYO needs to call |
3142 |
|
lock_sock()/release_sock() around skb_kill_datagram() call when UDP |
3143 |
|
packet was dropped by TOMOYO. |