Any time fcaps or a setuid app under SECURE_NOROOT is used to result in a non-zero pE we will crate a new audit record which contains the entire set of known information about the executable in question, fP, fI, fE, fversion and includes the process's pE, pI, pP. Before and after the bprm capability are applied. This record type will only be emitted from execve syscalls. an example of making ping use fcaps instead of setuid: setcap "cat_net_raw+pe" /bin/ping type=SYSCALL msg=audit(1225742021.015:236): arch=c000003e syscall=59 success=yes exit=0 a0=1457f30 a1=14606b0 a2=1463940 a3=321b770a70 items=2 ppid=2929 pid=2963 auid=0 uid=500 gid=500 euid=500 suid=500 fsuid=500 egid=500 sgid=500 fsgid=500 tty=pts0 ses=3 comm="ping" exe="/bin/ping" subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 key=(null) type=UNKNOWN[1321] msg=audit(1225742021.015:236): fver=2 fp=0000000000002000 fi=0000000000000000 fe=1 old_pp=0000000000000000 old_pi=0000000000000000 old_pe=0000000000000000 new_pp=0000000000002000 new_pi=0000000000000000 new_pe=0000000000002000 type=EXECVE msg=audit(1225742021.015:236): argc=2 a0="ping" a1="127.0.0.1" type=CWD msg=audit(1225742021.015:236): cwd="/home/test" type=PATH msg=audit(1225742021.015:236): item=0 name="/bin/ping" inode=49256 dev=fd:00 mode=0100755 ouid=0 ogid=0 rdev=00:00 obj=system_u:object_r:ping_exec_t:s0 cap_fp=0000000000002000 cap_fe=1 cap_fver=2 type=PATH msg=audit(1225742021.015:236): item=1 name=(null) inode=507915 dev=fd:00 mode=0100755 ouid=0 ogid=0 rdev=00:00 obj=system_u:object_r:ld_so_t:s0 Signed-off-by: Eric Paris <eparis(a)redhat.com&gt; --- include/linux/audit.h | 11 ++++++++ kernel/auditsc.c | 65 +++++++++++++++++++++++++++++++++++++++++++++++++ security/commoncap.c | 10 ++++++++ 3 files changed, 86 insertions(+), 0 deletions(-) diff --git a/include/linux/audit.h b/include/linux/audit.h index 6272a39..5ecca2b 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -99,6 +99,7 @@ #define AUDIT_OBJ_PID 1318 /* ptrace target */ #define AUDIT_TTY 1319 /* Input on an administrative TTY */ #define AUDIT_EOE 1320 /* End of multi-record event */ +#define AUDIT_BPRM_FCAPS 1321 /* Information about fcaps increasing perms */ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ @@ -452,6 +453,7 @@ extern int __audit_mq_timedsend(mqd_t mqdes, size_t msg_len, unsigned int msg_pr extern int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned int __user *u_msg_prio, const struct timespec __user *u_abs_timeout); extern int __audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification); extern int __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat); +extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t *pP, kernel_cap_t *pE); static inline int audit_ipc_obj(struct kern_ipc_perm *ipcp) { @@ -501,6 +503,14 @@ static inline int audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat) return __audit_mq_getsetattr(mqdes, mqstat); return 0; } + +static inline int audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t *pP, kernel_cap_t *pE) +{ + if (unlikely(!audit_dummy_context())) + return __audit_log_bprm_fcaps(bprm, pP, pE); + return 0; +} + extern int audit_n_rules; extern int audit_signals; #else @@ -532,6 +542,7 @@ extern int audit_signals; #define audit_mq_timedreceive(d,l,p,t) ({ 0; }) #define audit_mq_notify(d,n) ({ 0; }) #define audit_mq_getsetattr(d,s) ({ 0; }) +#define audit_log_bprm_fcaps(b, p, e) 0 #define audit_ptrace(t) ((void)0) #define audit_n_rules 0 #define audit_signals 0 diff --git a/kernel/auditsc.c b/kernel/auditsc.c index f87f31a..8fe907a 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -196,6 +196,14 @@ struct audit_aux_data_pids { int pid_count; }; +struct audit_aux_data_bprm_fcaps { + struct audit_aux_data d; + struct audit_cap_data fcap; + unsigned int fcap_ver; + struct audit_cap_data old_pcap; + struct audit_cap_data new_pcap; +}; + struct audit_tree_refs { struct audit_tree_refs *next; struct audit_chunk *c[31]; @@ -1375,6 +1383,20 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_format(ab, "fd0=%d fd1=%d", axs->fd[0], axs->fd[1]); break; } + case AUDIT_BPRM_FCAPS: { + struct audit_aux_data_bprm_fcaps *axs = (void *)aux; + audit_log_format(ab, "fver=%x", axs->fcap_ver); + audit_log_cap(ab, "fp", &axs->fcap.permitted); + audit_log_cap(ab, "fi", &axs->fcap.inheritable); + audit_log_format(ab, " fe=%d", axs->fcap.fE); + audit_log_cap(ab, "old_pp", &axs->old_pcap.permitted); + audit_log_cap(ab, "old_pi", &axs->old_pcap.inheritable); + audit_log_cap(ab, "old_pe", &axs->old_pcap.effective); + audit_log_cap(ab, "new_pp", &axs->new_pcap.permitted); + audit_log_cap(ab, "new_pi", &axs->new_pcap.inheritable); + audit_log_cap(ab, "new_pe", &axs->new_pcap.effective); + break; } + } audit_log_end(ab); } @@ -2502,6 +2524,49 @@ int __audit_signal_info(int sig, struct task_struct *t) } /** + * __audit_log_bprm_fcaps - store information about a loading bprm and relevant fcaps + * @bprm pointer to the bprm being processed + * @caps the caps read from the disk + * + * Simply check if the proc already has the caps given by the file and if not + * store the priv escalation info for later auditing at the end of the syscall + */ +int __audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t *pP, kernel_cap_t *pE) +{ + struct audit_aux_data_bprm_fcaps *ax; + struct audit_context *context = current->audit_context; + struct cpu_vfs_cap_data vcaps; + struct dentry *dentry; + + ax = kmalloc(sizeof(*ax), GFP_KERNEL); + if (!ax) + return -ENOMEM; + + ax->d.type = AUDIT_BPRM_FCAPS; + ax->d.next = context->aux; + context->aux = (void *)ax; + + dentry = dget(bprm->file->f_dentry); + get_vfs_caps_from_disk(dentry, &vcaps); + dput(dentry); + + ax->fcap.permitted = vcaps.permitted; + ax->fcap.inheritable = vcaps.inheritable; + ax->fcap.fE = !!(vcaps.magic_etc & VFS_CAP_FLAGS_EFFECTIVE); + ax->fcap_ver = (vcaps.magic_etc & VFS_CAP_REVISION_MASK) >> VFS_CAP_REVISION_SHIFT; + + ax->old_pcap.permitted = *pP; + ax->old_pcap.inheritable = current->cap_inheritable; + ax->old_pcap.effective = *pE; + + ax->new_pcap.permitted = current->cap_permitted; + ax->new_pcap.inheritable = current->cap_inheritable; + ax->new_pcap.effective = current->cap_effective; + + return 0; +} + +/** * audit_core_dumps - record information about processes that end abnormally * @signr: signal value * diff --git a/security/commoncap.c b/security/commoncap.c index 8bb95ed..534abb5 100644 --- a/security/commoncap.c +++ b/security/commoncap.c @@ -8,6 +8,7 @@ */ #include <linux/capability.h> +#include <linux/audit.h> #include <linux/module.h> #include <linux/init.h> #include <linux/kernel.h> @@ -373,6 +374,9 @@ int cap_bprm_set_security (struct linux_binprm *bprm) void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) { + kernel_cap_t pP = current->cap_permitted; + kernel_cap_t pE = current->cap_effective; + if (bprm->e_uid != current->uid || bprm->e_gid != current->gid || !cap_issubset(bprm->cap_post_exec_permitted, current->cap_permitted)) { @@ -407,6 +411,12 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe) } /* AUD: Audit candidate if current->cap_effective is set */ + if (!cap_isclear(current->cap_effective)) { + if (!cap_issubset(current->cap_effective, CAP_FULL_SET) || + (bprm->e_uid != 0) || (current->uid != 0) || + issecure(SECURE_NOROOT)) + audit_log_bprm_fcaps(bprm, &pP, &pE); + } current->securebits &= ~issecure_mask(SECURE_KEEP_CAPS); }

On Tue, 2008-11-04 at 10:35 -0600, Serge E. Hallyn wrote: I'm glad you noticed it, because it is backwards! should be if (!cap_issubset(CAP_FULL_SET, current->cap_effective) || The idea is that I don't care to audit 1) full set && 2) this is root && 3) root should have a full set This would still consider a root without a full set because the bset had been changed to be an interesting event. I'm fine with that. I'll send -v3 once I hear comments on everything else.... -Eric

On Tue, 2008-11-04 at 10:35 -0600, Serge E. Hallyn wrote: So here's the problem.... I can't fail this syscall, it's too late. I can do a couple of things. 1) waste lots of space in the execve record so we know memory has already been allocated 2) just ignore the memory failure and don't worry about it. We are still going to get the fcaps info from the patch record and should be able to piece together the starting and finishing caps by looking at past audit records if you really need it. 3) I can call audit_log_lost(). I don't think we know are this time that we really needed this record, but this is the 'safest' approach. If people have their machines set to panic on lost records we would panic. Honestly though, if we don't have enough memory to satisfy this request (we're talking about 72 bytes or something?) we are going to fail the next audit message, so doing it now would be just fine. I vote #2 since I don't think we are really going to have any lose of info. But if people want it I'll go #3 since I don't think it will hurt anything. -Eric

When the capset syscall is used it is not possible for audit to record the actual capbilities being added/removed. This patch adds a new record type which emits the target pid and the eff, inh, and perm cap sets. example output if you audit capset syscalls would be: type=SYSCALL msg=audit(1225743140.465:76): arch=c000003e syscall=126 success=yes exit=0 a0=17f2014 a1=17f201c a2=80000000 a3=7fff2ab7f060 items=0 ppid=2160 pid=2223 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=pts0 ses=1 comm="setcap" exe="/usr/sbin/setcap" subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 key=(null) type=UNKNOWN[1322] msg=audit(1225743140.465:76): pid=0 cap_pi=ffffffffffffffff cap_pp=ffffffffffffffff cap_pe=ffffffffffffffff Signed-off-by: Eric Paris <eparis(a)redhat.com&gt; --- include/linux/audit.h | 10 ++++++++++ kernel/auditsc.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ kernel/capability.c | 3 +++ 3 files changed, 61 insertions(+), 0 deletions(-) diff --git a/include/linux/audit.h b/include/linux/audit.h index 5ecca2b..d6176f1 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -100,6 +100,7 @@ #define AUDIT_TTY 1319 /* Input on an administrative TTY */ #define AUDIT_EOE 1320 /* End of multi-record event */ #define AUDIT_BPRM_FCAPS 1321 /* Information about fcaps increasing perms */ +#define AUDIT_CAPSET 1322 /* Record showing argument to sys_capset */ #define AUDIT_AVC 1400 /* SE Linux avc denial or grant */ #define AUDIT_SELINUX_ERR 1401 /* Internal SE Linux Errors */ @@ -454,6 +455,7 @@ extern int __audit_mq_timedreceive(mqd_t mqdes, size_t msg_len, unsigned int __u extern int __audit_mq_notify(mqd_t mqdes, const struct sigevent __user *u_notification); extern int __audit_mq_getsetattr(mqd_t mqdes, struct mq_attr *mqstat); extern int __audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t *pP, kernel_cap_t *pE); +extern int __audit_log_capset(pid_t pid, kernel_cap_t *eff, kernel_cap_t *inh, kernel_cap_t *perm); static inline int audit_ipc_obj(struct kern_ipc_perm *ipcp) { @@ -511,6 +513,13 @@ static inline int audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t * return 0; } +static inline int audit_log_capset(pid_t pid, kernel_cap_t *eff, kernel_cap_t *inh, kernel_cap_t *perm) +{ + if (unlikely(!audit_dummy_context())) + return __audit_log_capset(pid, eff, inh, perm); + return 0; +} + extern int audit_n_rules; extern int audit_signals; #else @@ -543,6 +552,7 @@ extern int audit_signals; #define audit_mq_notify(d,n) ({ 0; }) #define audit_mq_getsetattr(d,s) ({ 0; }) #define audit_log_bprm_fcaps(b, p, e) 0 +#define audit_log_capset(pid, e, i, p) do { ; } while (0) #define audit_ptrace(t) ((void)0) #define audit_n_rules 0 #define audit_signals 0 diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 8fe907a..a4ff5fa 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -204,6 +204,12 @@ struct audit_aux_data_bprm_fcaps { struct audit_cap_data new_pcap; }; +struct audit_aux_data_capset { + struct audit_aux_data d; + pid_t pid; + struct audit_cap_data cap; +}; + struct audit_tree_refs { struct audit_tree_refs *next; struct audit_chunk *c[31]; @@ -1397,6 +1403,14 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts audit_log_cap(ab, "new_pe", &axs->new_pcap.effective); break; } + case AUDIT_CAPSET: { + struct audit_aux_data_capset *axs = (void *)aux; + audit_log_format(ab, "pid=%d", axs->pid); + audit_log_cap(ab, "cap_pi", &axs->cap.inheritable); + audit_log_cap(ab, "cap_pp", &axs->cap.permitted); + audit_log_cap(ab, "cap_pe", &axs->cap.effective); + break; } + } audit_log_end(ab); } @@ -2567,6 +2581,40 @@ int __audit_log_bprm_fcaps(struct linux_binprm *bprm, kernel_cap_t *pP, kernel_c } /** + * __audit_log_capset - store information about the arguments to the capset syscall + * @pid target pid of the capset call + * @eff effective cap set + * @inh inheritible cap set + * @perm permited cap set + * + * Record the aguments userspace sent to sys_capset for later printing by the + * audit system if applicable + */ +int __audit_log_capset(pid_t pid, kernel_cap_t *eff, kernel_cap_t *inh, kernel_cap_t *perm) +{ + struct audit_aux_data_capset *ax; + struct audit_context *context = current->audit_context; + + if (likely(!audit_enabled || !context || context->dummy)) + return 0; + + ax = kmalloc(sizeof(*ax), GFP_KERNEL); + if (!ax) + return -ENOMEM; + + ax->d.type = AUDIT_CAPSET; + ax->d.next = context->aux; + context->aux = (void *)ax; + + ax->pid = pid; + ax->cap.effective = *eff; + ax->cap.inheritable = *eff; + ax->cap.permitted = *perm; + + return 0; +} + +/** * audit_core_dumps - record information about processes that end abnormally * @signr: signal value * diff --git a/kernel/capability.c b/kernel/capability.c index 33e51e7..3d7fc70 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -7,6 +7,7 @@ * 30 May 2002: Cleanup, Robert M. Love <rml(a)tech9.net&gt; */ +#include <linux/audit.h> #include <linux/capability.h> #include <linux/mm.h> #include <linux/module.h> @@ -457,6 +458,8 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data) i++; } + audit_log_capset(pid, &effective, &inheritable, &permitted); + if (pid && (pid != task_pid_vnr(current))) ret = do_sys_capset_other_tasks(pid, &effective, &inheritable, &permitted);

On Tue, 2008-11-04 at 10:55 -0600, Serge E. Hallyn wrote: yes, I think it should, but look for the next e-mail.....