2411 |
@ Fix incorrect out of memory warning. |
@ Fix incorrect out of memory warning. |
2412 |
|
|
2413 |
Out of memory warnings were not printed in some cases by error. |
Out of memory warnings were not printed in some cases by error. |
2414 |
|
|
2415 |
|
Fix 2010/05/27 |
2416 |
|
|
2417 |
|
@ Add missing rcu_dereference() for ccs_find_execute_handler(). |
2418 |
|
|
2419 |
|
Since 1.7.0 , ccs_find_execute_handler() was by error using |
2420 |
|
list_for_each_entry() rather than list_for_each_entry_rcu(). |
2421 |
|
This bug affects only Alpha architecture. |
2422 |
|
|
2423 |
|
Fix 2010/06/03 |
2424 |
|
|
2425 |
|
@ Fix missing sanity check for "file_pattern". |
2426 |
|
|
2427 |
|
Since 1.7.0 , ccs_write_pattern_policy() was by error accepting |
2428 |
|
invalid pathname. |
2429 |
|
|
2430 |
|
Fix 2010/06/09 |
2431 |
|
|
2432 |
|
@ Add missing ccs_put_name() in ccs_parse_envp(). |
2433 |
|
|
2434 |
|
Since 1.7.0 , ccs_parse_envp() was not calling ccs_put_name() if |
2435 |
|
environment variable's value ('if exec.envp["name"]="value"' condition) |
2436 |
|
was invalid. |
2437 |
|
|
2438 |
|
@ Add missing NULL check in ccs_condition(). |
2439 |
|
|
2440 |
|
Since 1.7.0 , if 'if symlink.target=' part was given against non-file |
2441 |
|
permissions (e.g. allow_env PATH if symlink.target="/"), it triggered |
2442 |
|
NULL pointer dereference. |
2443 |
|
|
2444 |
|
Fix 2010/07/29 |
2445 |
|
|
2446 |
|
@ Change keyword syntax. |
2447 |
|
|
2448 |
|
I removed "allow_" prefix from directives. New directives for files are |
2449 |
|
prefixed with "file ". For example, "allow_read" changed to "file read", |
2450 |
|
"allow_ioctl" changed to "file ioctl". New directive for "allow_network" |
2451 |
|
is "network". New directive for "allow_env" is "misc env". New directive |
2452 |
|
for "allow_signal" is "ipc signal". New directive for "allow_capability" |
2453 |
|
is "capability". These directives correspond with keywords used by |
2454 |
|
profile's CONFIG lines. |
2455 |
|
|
2456 |
|
I removed "deny_rewrite" and "allow_rewrite" directives and introduced |
2457 |
|
"file append" directive. Thus, permission for open(O_WRONLY | O_APPEND) |
2458 |
|
changed from "allow_write" + "allow_rewrite" to "file append". |
2459 |
|
|
2460 |
|
I removed "SYS_MOUNT", "SYS_UMOUNT", "SYS_CHROOT", "SYS_KILL", |
2461 |
|
"SYS_LINK", "SYS_SYMLINK", "SYS_RENAME", "SYS_UNLINK", "SYS_CHMOD", |
2462 |
|
"SYS_CHOWN", "SYS_IOCTL", "SYS_PIVOT_ROOT" keywords from capabilities |
2463 |
|
because these permissions can be checked by other directives (e.g. |
2464 |
|
"file mount", "ipc signal"). |
2465 |
|
|
2466 |
|
I also removed "conceal_mount" keyword from capabilities because this |
2467 |
|
check requires hooks in filesystem part while almost all hooks for |
2468 |
|
filesystem part have moved to LSM by Linux 2.6.34. |
2469 |
|
|
2470 |
|
@ Distinguish send() and recv() operations for UDP and IP protocols. |
2471 |
|
|
2472 |
|
Until now, it was impossible for UDP and IP protocols to allow either |
2473 |
|
only sending or only receiving because permissions were aggregated with |
2474 |
|
"connect" keyword. I broke "connect" keyword into "send" and "recv" |
2475 |
|
keywords so that you can keep access control for send() operation enabled |
2476 |
|
when you have to turn access control for recv() operation off due to |
2477 |
|
application breakage by filtering incoming datagram. |
2478 |
|
|
2479 |
|
@ Wait for next connection/datagram if current connection/datagram was |
2480 |
|
discarded. |
2481 |
|
|
2482 |
|
Regarding "network TCP accept", "network UDP recv", "network RAW recv" |
2483 |
|
keywords, I modified to wait for next connection/datagram if current |
2484 |
|
connection/datagram was discarded. LSM hooks for these keywords are |
2485 |
|
currently missing because this behavior may break applications. |
2486 |
|
If you found applications broken by this behavior, you can set |
2487 |
|
CONFIG::network::inet_tcp_accept and/or CONFIG::network::inet_udp_recv |
2488 |
|
and/or CONFIG::network::inet_raw_recv to mode=disabled in order to |
2489 |
|
disable filtering for incoming connection/datagram. |
2490 |
|
|
2491 |
|
@ Allow specifying multiple permissions in a line. |
2492 |
|
|
2493 |
|
Until now, only "allow_read/write" can be specified for combination of |
2494 |
|
"allow_read" + "allow_write". Now, you can combine other permissions as |
2495 |
|
long as type of parameters for these permissions is same. For example, |
2496 |
|
"file read/write/append/execute/unlink/truncate /tmp/file" is correct |
2497 |
|
but "file read/write/create /tmp/file" is wrong because "file create" |
2498 |
|
requires create mode whereas "file read" and "file write" do not. |
2499 |
|
|
2500 |
|
@ Allow wildcard for execute permission and domainname. |
2501 |
|
|
2502 |
|
Until now, to execute programs with temporary names, "aggregator" is |
2503 |
|
needed. To simplify code, I modified to accept wildcards for execute |
2504 |
|
permission and domainname. Now, you can directly specify |
2505 |
|
"file execute /tmp/logrotate.\?\?\?\?\?\?" and use |
2506 |
|
"/tmp/logrotate.\?\?\?\?\?\?" within domainnames. |
2507 |
|
|
2508 |
|
@ Change pathname for non-rename()able filesystems. |
2509 |
|
|
2510 |
|
LSM version of TOMOYO wants to use /proc/self/ rather than /proc/$PID/ if |
2511 |
|
$PID matches current thread's process ID in order to prevent current |
2512 |
|
thread from accessing other process's information unless needed. |
2513 |
|
But since procfs can be mounted on various locations (e.g. /proc/ /proc2/ |
2514 |
|
/p/ /tmp/foo/100/p/ ), LSM version of TOMOYO cannot tell that whether the |
2515 |
|
numeric part in the string returned by __d_path() represents process ID |
2516 |
|
or not. |
2517 |
|
|
2518 |
|
Therefore, to be able to convert from $PID to self no matter where procfs |
2519 |
|
is mounted, I changed pathname representations for filesystems which do |
2520 |
|
not support rename() operation (e.g. proc, sysfs, securityfs). |
2521 |
|
|
2522 |
|
Now, "/proc/self/mounts" changed to "proc:/self/mounts" and |
2523 |
|
"/sys/kernel/security/" changed to "sys:/kernel/security/" and |
2524 |
|
"/dev/pts/0" changed to "devpts:/0". |
2525 |
|
|
2526 |
|
@ Add a new keyword "any" for domain transition control. |
2527 |
|
|
2528 |
|
To be able to make it easier to apply execute_handler on each domain, |
2529 |
|
I added "any" keyword to domain transition control keywords. Now, |
2530 |
|
"initialize_domain /usr/sbin/sshd" changed to |
2531 |
|
"initialize_domain /usr/sbin/sshd from any" and |
2532 |
|
"keep_domain <kernel> /usr/sbin/sshd /bin/bash" changed to |
2533 |
|
"keep_domain any from <kernel> /usr/sbin/sshd /bin/bash". |
2534 |
|
|
2535 |
|
"keep_domain /path/to/execute_handler from any" will allow you to apply |
2536 |
|
execute_handler for any domains without creating domains for |
2537 |
|
execute_handler. |
2538 |
|
|
2539 |
|
@ Change buffering mode for reading policy. |
2540 |
|
|
2541 |
|
To be able to read() very very long lines correctly, I changed the way |
2542 |
|
TOMOYO buffers policy for reading. |
2543 |
|
|
2544 |
|
@ Introduce "acl_group" keyword. |
2545 |
|
|
2546 |
|
Until now, it was possible to specify only "allow_read" and "allow_env" |
2547 |
|
keywords in the exception policy. |
2548 |
|
|
2549 |
|
Since some operations like "file read/write/append /dev/null" and |
2550 |
|
"network UDP send/recv @DNS_SERVER 53" are very common and should be |
2551 |
|
permitted to all domains, I introduced "acl_group" keyword for giving |
2552 |
|
such permissions. |
2553 |
|
|
2554 |
|
For example, specify "acl_group 0 file read/write/append /dev/null" in |
2555 |
|
the exception policy and specify "use_group 0" from the domains in the |
2556 |
|
domain policy. |
2557 |
|
|
2558 |
|
"ignore_global_allow_read" and "ignore_global_allow_env" keywords were |
2559 |
|
removed from domain policy and "use_group" keyword was added. |
2560 |
|
|
2561 |
|
@ Allow controlling generation of access granted logs for per an entry |
2562 |
|
basis. |
2563 |
|
|
2564 |
|
I added per-entry flag which controls generation of grant logs because |
2565 |
|
Xen and KVM issues ioctl requests so frequently. For example, |
2566 |
|
|
2567 |
|
file ioctl /dev/null 0x5401 ; set audit=no |
2568 |
|
|
2569 |
|
will suppress /proc/ccs/grant_log even if profile says grant_log=yes . |
2570 |
|
|
2571 |
|
file ioctl /dev/null 0x5401 ; set audit=yes |
2572 |
|
|
2573 |
|
will generate /proc/ccs/grant_log even if profile says grant_log=no . |
2574 |
|
|
2575 |
|
file ioctl /dev/null 0x5401 |
2576 |
|
|
2577 |
|
will generate /proc/ccs/grant_log only if profile says grant_log=yes . |
2578 |
|
|
2579 |
|
This flag is intended for frequently accessed resources like |
2580 |
|
|
2581 |
|
file read /var/www/html/\{\*\}/\*.html ; set audit=no |
2582 |
|
|
2583 |
|
. |
2584 |
|
|
2585 |
|
@ Optimize for object's size. |
2586 |
|
|
2587 |
|
I merged similar code in order to reduce object's filesize. |