Skip to contentSkip to conversation

Update validate() to support multiple violations

RFC PATCH In review Mimi Zohar 2 participants 17 messages Feb 20, 2025 – Feb 25, 2025 Keys-trusted

lore
Share & tools
lore

Mimi Zohar

author maintainer
#0 lore raw

RFC PATCH 1/3 Update validate() to support multiple violations

Add support for the number of expected violations. Include the expected number of violations in the output.

---
 .../security/integrity/ima/tests/ima_violations.sh     | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +8 -2
index 37d8d473c..7f0382fb8 100755
@@ -71,20 +71,26 @@ validate()
 	local num_violations="$1"
 	local count="$2"
 	local search="$3"
+	local expected_violations=$4
 	local max_attempt=3
 	local count2 i num_violations_new
 
+	[ -z $expected_violations ] && expected_violations=1
+
 	for i in $(seq 1 $max_attempt); do
 		read num_violations_new < $IMA_VIOLATIONS
 		count2="$(get_count $search)"
-		if [ $(($num_violations_new - $num_violations)) -gt 0 ]; then
+		if [ $(($num_violations_new - $num_violations)) -eq $expected_violations ]; then
 			if [ $count2 -gt $count ]; then
-				tst_res TPASS "$search violation added"
+				tst_res TPASS "$expected_violations $search violation(s) added"
 				return
 			else
 				tst_res TINFO "$search not found in $LOG ($i/$max_attempt attempt)..."
 				tst_sleep 1s
 			fi
+		elif [ $(($num_violations_new - $num_violations)) -gt 0 ]; then
+			tst_res $IMA_FAIL "$search too many violations added"
+			return
 		else
 			tst_res $IMA_FAIL "$search violation not added"
 			return
Signature · 1 line
2.48.1
author maintainer
#1 lore raw

RFC PATCH 2/3 ima: additional open-writer violation tests

Kernel patch "ima: limit the number of open-writers integrity violations" prevents superfluous "open-writers" violations. Add corresponding LTP tests.

---
 .../integrity/ima/tests/ima_violations.sh     | 87 ++++++++++++++++++-
 1 file changed, 86 insertions(+), 1 deletion(-)
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +86 -1
index 7f0382fb8..65c5c3a92 100755
@@ -8,7 +8,7 @@
 
 TST_SETUP="setup"
 TST_CLEANUP="cleanup"
-TST_CNT=3
+TST_CNT=6
 
 REQUIRED_BUILTIN_POLICY="tcb"
 REQUIRED_POLICY_CONTENT='violations.policy'
@@ -60,6 +60,17 @@ close_file_write()
 	exec 4>&-
 }
 
+open_file_write2()
+{
+	exec 5> $FILE || exit 1
+	echo 'test writing2' >&5
+}
+
+close_file_write2()
+{
+	exec 5>&-
+}
+
 get_count()
 {
 	local search="$1"
@@ -160,6 +171,80 @@ test3()
 	tst_sleep 2s
 }
 
+test4()
+{
+	tst_res TINFO "verify limiting single open writer violation"
+
+	local search="open_writers"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_write
+	open_file_read
+	close_file_read
+
+	open_file_read
+	close_file_read
+
+	close_file_write
+
+	validate $num_violations $count $search 1
+}
+
+test5()
+{
+	tst_res TINFO "verify limiting multiple open writers violations"
+
+	local search="open_writers"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_write
+	open_file_read
+	close_file_read
+
+	open_file_write2
+	open_file_read
+	close_file_read
+	close_file_write2
+
+	open_file_read
+	close_file_read
+
+	close_file_write
+
+	validate $num_violations $count $search 1
+}
+
+test6()
+{
+	tst_res TINFO "verify new open writer causes additional violation"
+
+	local search="open_writers"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_write
+	open_file_read
+	close_file_read
+
+	open_file_read
+	close_file_read
+	close_file_write
+
+	open_file_write
+	open_file_read
+	close_file_read
+	close_file_write
+	validate $num_violations $count $search 2
+}
+
 . ima_setup.sh
 . daemonlib.sh
 tst_run
Signature · 1 line
2.48.1
#2 lore raw

Re: [RFC PATCH 2/3] ima: additional open-writer violation tests

Hi Mimi,

Quoted 24 lines

Kernel patch "ima: limit the number of open-writers integrity violations" prevents superfluous "open-writers" violations. Add corresponding LTP tests.

---
 .../integrity/ima/tests/ima_violations.sh     | 87 ++++++++++++++++++-
 1 file changed, 86 insertions(+), 1 deletion(-)
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +1 -1
index 7f0382fb8..65c5c3a92 100755
@@ -8,7 +8,7 @@
 
 TST_SETUP="setup"
 TST_CLEANUP="cleanup"
-TST_CNT=3
+TST_CNT=6
 
 REQUIRED_BUILTIN_POLICY="tcb"
 REQUIRED_POLICY_CONTENT='violations.policy'

testcases/kernel/security/integrity/ima/tests/ima_violations.sh close_file_write() · line 65

Quoted 6 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +3 -0
@@ -60,6 +60,17 @@ close_file_write()
 	exec 4>&-
 }
 
+open_file_write2()
+{
+	exec 5> $FILE || exit 1

maybe:

	exec 5> $FILE || tst_brk TBROK "exec 5> $FILE failed"

Because tst_brk TBROK calls test cleanup. Plain exit kills everything. We also have ROD, but that requires binaries ('exec' is a shell builtin). (It applies to the third patch as well.)

Quoted 52 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +45 -0
+	echo 'test writing2' >&5
+}
+
+close_file_write2()
+{
+	exec 5>&-
+}
+
 get_count()
 {
 	local search="$1"
@@ -160,6 +171,80 @@ test3()
 	tst_sleep 2s
 }
 
+test4()
+{
+	tst_res TINFO "verify limiting single open writer violation"
+
+	local search="open_writers"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_write
+	open_file_read
+	close_file_read
+
+	open_file_read
+	close_file_read
+
+	close_file_write
+
+	validate $num_violations $count $search 1
+}
+
+test5()
+{
+	tst_res TINFO "verify limiting multiple open writers violations"
+
+	local search="open_writers"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_write
+	open_file_read
+	close_file_read
+
+	open_file_write2

testcases/kernel/security/integrity/ima/tests/ima_violations.sh test3() · line 220

Quoted 10 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +10 -0
+	open_file_read
+	close_file_read
+	close_file_write2
+
+	open_file_read
+	close_file_read
+
+	close_file_write
+
+	validate $num_violations $count $search 1

nit: safer to quote
validate "$num_violations" "$count" "$search" 1

Quoted 15 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +15 -0
+}
+
+test6()
+{
+	tst_res TINFO "verify new open writer causes additional violation"
+
+	local search="open_writers"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_write
+	open_file_read
+	close_file_read

testcases/kernel/security/integrity/ima/tests/ima_violations.sh test3() · line 245

Quoted 10 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +10 -0
+
+	open_file_read
+	close_file_read
+	close_file_write
+
+	open_file_write
+	open_file_read
+	close_file_read
+	close_file_write
+	validate $num_violations $count $search 2

And here.

Kind regards,
Petr

author maintainer
#3 lore raw

RFC PATCH 3/3 ima: additional ToMToU violation tests

In reply to Mimi Zohar

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations. Add corresponding LTP tests.

---
 .../integrity/ima/tests/ima_violations.sh     | 46 ++++++++++++++++++-
 1 file changed, 45 insertions(+), 1 deletion(-)
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +45 -1
index 65c5c3a92..5b6d7e993 100755
@@ -8,7 +8,7 @@
 
 TST_SETUP="setup"
 TST_CLEANUP="cleanup"
-TST_CNT=6
+TST_CNT=8
 
 REQUIRED_BUILTIN_POLICY="tcb"
 REQUIRED_POLICY_CONTENT='violations.policy'
@@ -245,6 +245,50 @@ test6()
 	validate $num_violations $count $search 2
 }
 
+test7()
+{
+	tst_res TINFO "verify limiting single open reader ToMToU violations"
+
+	local search="ToMToU"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_read
+	open_file_write
+	close_file_write
+
+	open_file_write
+	close_file_write
+	close_file_read
+
+	validate $num_violations $count $search 1
+}
+
+test8()
+{
+	tst_res TINFO "verify new open reader causes additional ToMToU violation"
+
+	local search="ToMToU"
+	local count num_violations
+
+	read num_violations < $IMA_VIOLATIONS
+	count="$(get_count $search)"
+
+	open_file_read
+	open_file_write
+	close_file_write
+	close_file_read
+
+	open_file_read
+	open_file_write
+	close_file_write
+	close_file_read
+
+	validate $num_violations $count $search 2
+}
+
 . ima_setup.sh
 . daemonlib.sh
 tst_run
Signature · 1 line
2.48.1
#4 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Hi Mimi,

Quoted 5 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations. Add corresponding LTP tests.

Unfortunately tests fail on both mainline kernel and kernel with your patches.

Any hint what could be wrong?

Mainline kernel (on kernel with your patches it looks the same):
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64 x86_64 x86_64 GNU/Linux
ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.cKm34XVZk2 as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0'
ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.cKm34XVZk2/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng"
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y
ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.13.0-2.g0127a37-default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device
ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: SUT has required policy content
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TFAIL: ToMToU too many violations added
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TFAIL: open_writers too many violations added
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TFAIL: open_writers too many violations added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TFAIL: open_writers too many violations added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TFAIL: open_writers too many violations added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TFAIL: ToMToU too many violations added ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TFAIL: ToMToU too many violations added

Kind regards,
Petr

#5 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Hi Mimi,

Quoted 5 lines

Hi Mimi,

Quoted 2 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations. Add corresponding LTP tests.

Quoted 8 lines
Quoted 2 lines

Unfortunately tests fail on both mainline kernel and kernel with your patches.

Any hint what could be wrong?

Mainline kernel (on kernel with your patches it looks the same):

I'm sorry, I accidentally tested only on vanilla kernel. Rerunning tests with updated kernel.

Is it this considered as a security feature? If yes, than failures on vanilla kernel are ok, we just need to later add kernel hashes to let testers know about missing backports. If it's a feature (not to be backported) we should test new feature only on newer kernels.

Kind regards,
Petr

Quoted 64 lines

ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64 x86_64 x86_64 GNU/Linux
ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.cKm34XVZk2 as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0'
ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.cKm34XVZk2/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng"
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y
ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.13.0-2.g0127a37-default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device
ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: SUT has required policy content
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TFAIL: ToMToU too many violations added
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TFAIL: open_writers too many violations added
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TFAIL: open_writers too many violations added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TFAIL: open_writers too many violations added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TFAIL: open_writers too many violations added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TFAIL: ToMToU too many violations added ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TFAIL: ToMToU too many violations added

Kind regards,
Petr

author maintainer
#6 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Hi Petr,

Quoted 4 lines · Petr Vorel

Is it this considered as a security feature? If yes, than failures on vanilla kernel are ok, we just need to later add kernel hashes to let testers know about missing backports. If it's a feature (not to be backported) we should test new feature only on newer kernels.

I posted these LTP patches as RFC since the kernel patches themselves haven't been upstreamed. I'm still waiting for some kernel patch reviews. Posting these LTP patches might help with that.

Having multiple open-writers or ToMToU violations doesn't provide any benefit in terms of attestation. It just clutters the audit log and the IMA measurement list. Not extending the TPM would be a performance improvement. I'm not sure it would be classified as a security feature or bug fix.

Mimi

author maintainer
#7 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

In reply to Petr Vorel
Quoted 10 lines · Petr Vorel

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 2 lines

Any hint what could be wrong?

Of course it's dependent on the IMA policy. The tests assume being booted with the IMA TCB measurement policy or similar policy being loaded. Can you share the IMA policy? e.g. cat /sys/kernel/security/ima/policy

thanks,

Mimi

Quoted 76 lines

Mainline kernel (on kernel with your patches it looks the same):
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64 x86_64 x86_64 GNU/Linux ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.cKm34XVZk2 as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0' ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.cKm34XVZk2/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng" ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.13.0-2.g0127a37- default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel
ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: SUT has required policy content
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TFAIL: ToMToU too many violations added
ima_violations 3 TINFO: verify open_writers using mmapped files tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f tst_test.c:1904: TINFO: Tested kernel: 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz' tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TFAIL: open_writers too many violations added ima_mmap.c:41: TPASS: test completed

Summary:
passed   1
failed   0
broken   0
skipped  0
warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TFAIL: open_writers too many violations added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TFAIL: open_writers too many violations added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TFAIL: open_writers too many violations added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TFAIL: ToMToU too many violations added ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TFAIL: ToMToU too many violations added

Kind regards,
Petr

#8 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 16 lines · Petr Vorel
Quoted 10 lines

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 1 line

Any hint what could be wrong?

Quoted 7 lines

Of course it's dependent on the IMA policy. The tests assume being booted with the IMA TCB measurement policy or similar policy being loaded. Can you share the IMA policy? e.g. cat /sys/kernel/security/ima/policy

thanks,

Mimi

Now testing on kernel *with* your patches. First run always fails, regardless whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

First run fails:

# LTP_IMA_LOAD_POLICY=1 LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh
(policy is /opt/ltp/testcases/data/ima_violations/violations.policy)
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux
ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.XR34KhtnDM as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0'
ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.XR34KhtnDM/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng"
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y
ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102-default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device
ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: WARNING: missing required policy content: 'measure func=FILE_CHECK mask=^MAY_READ euid=0'
ima_violations 1 TINFO: trying to load '/opt/ltp/testcases/data/ima_violations/violations.policy' policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: example policy successfully loaded
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added: 2 - 0
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added
ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added
ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added
ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added
ima_violations 9 TINFO: WARNING: policy loaded via LTP_IMA_LOAD_POLICY=1, reboot recommended

Summary:

passed   7
failed   1
broken   0
skipped  0

warnings 0

Second run is ok:
# LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux
ima_violations 1 TINFO: Using /var/tmp/LTP_ima_violations.SWERFjvPTp as tmpdir (btrfs filesystem)
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng"
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y
ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102-default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: SUT has required policy content
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TPASS: 1 open_writers violation(s) added
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added

Summary:

passed   8
failed   0
broken   0
skipped  0

warnings 0

Reboot and running with ima_policy=tcb also fails on the first time:

# LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh
tmpfs is skipped
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux
ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.FKQSfezAwR as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0'
ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.FKQSfezAwR/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng"
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y
ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102-default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_policy=tcb ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device
ima_violations 1 TINFO: booted with IMA policy: tcb
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added: 3 - 1
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added

Summary:

passed   7
failed   1
broken   0
skipped  0

warnings 0

Second and later run is again OK
# LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh
tmpfs is skipped
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux
ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.1Qf6qJuSoo as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0'
ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.1Qf6qJuSoo/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng"
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y
ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102-default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_policy=tcb ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device
ima_violations 1 TINFO: booted with IMA policy: tcb
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TPASS: 1 open_writers violation(s) added
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64
tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added

Summary:

passed   8
failed   0
broken   0
skipped  0

warnings 0

author maintainer
#9 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 23 lines · Petr Vorel
Quoted 22 lines · Petr Vorel
Quoted 10 lines

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with the IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA policy? e.g. cat /sys/kernel/security/ima/policy

thanks,

Quoted 8 lines
Quoted 1 line

Mimi

Now testing on kernel *with* your patches. First run always fails, regardless whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

I'm not seeing that on my test machine. Could there be other things running on your
system causing violations. In anycase, your original test was less exacting. Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent new "gt" test.

if [ $(($num_violations_new - $num_violations)) -eq $expected_violations ]; then

Quoted 324 lines

First run fails:

# LTP_IMA_LOAD_POLICY=1 LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh
(policy is /opt/ltp/testcases/data/ima_violations/violations.policy) ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.XR34KhtnDM as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0' ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.XR34KhtnDM/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng" ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102- default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel
ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: WARNING: missing required policy content: 'measure func=FILE_CHECK mask=^MAY_READ euid=0'
ima_violations 1 TINFO: trying to load '/opt/ltp/testcases/data/ima_violations/violations.policy' policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: example policy successfully loaded
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added: 2 - 0 ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz' tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:
passed   1
failed   0
broken   0
skipped  0
warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added
ima_violations 9 TINFO: WARNING: policy loaded via LTP_IMA_LOAD_POLICY=1, reboot recommended

Summary:
passed   7
failed   1
broken   0
skipped  0
warnings 0

Second run is ok:
# LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux ima_violations 1 TINFO: Using /var/tmp/LTP_ima_violations.SWERFjvPTp as tmpdir (btrfs filesystem)
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng" ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102- default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel
ima_violations 1 TINFO: test requires IMA policy:
measure func=FILE_CHECK mask=^MAY_READ euid=0
measure func=FILE_CHECK mask=^MAY_READ uid=0
ima_violations 1 TINFO: SUT has required policy content
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TPASS: 1 open_writers violation(s) added
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz' tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:
passed   1
failed   0
broken   0
skipped  0
warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added

Summary:
passed   8
failed   0
broken   0
skipped  0
warnings 0

Reboot and running with ima_policy=tcb also fails on the first time:

# LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh tmpfs is skipped
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.FKQSfezAwR as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0' ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.FKQSfezAwR/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng" ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102- default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_policy=tcb
ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device ima_violations 1 TINFO: booted with IMA policy: tcb
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added: 3 - 1 ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz' tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:
passed   1
failed   0
broken   0
skipped  0
warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added

Summary:
passed   7
failed   1
broken   0
skipped  0
warnings 0

Second and later run is again OK
# LTPROOT="/opt/ltp" PATH="/opt/ltp/testcases/bin:$PATH" ima_violations.sh tmpfs is skipped
ima_violations 1 TINFO: Running: ima_violations.sh
ima_violations 1 TINFO: Tested kernel: Linux ts 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 x86_64 x86_64 GNU/Linux ima_violations 1 TINFO: Using /tmp/LTP_ima_violations.1Qf6qJuSoo as tmpdir (tmpfs filesystem)
tst_device.c:99: TINFO: Found free device 0 '/dev/loop0'
ima_violations 1 TINFO: Formatting ext3 with opts='/dev/loop0' ima_violations 1 TINFO: Mounting device: mount -t ext3 /dev/loop0 /tmp/LTP_ima_violations.1Qf6qJuSoo/mntpoint
ima_violations 1 TINFO: timeout per run is 0h 5m 0s
ima_violations 1 TINFO: IMA kernel config:
ima_violations 1 TINFO: CONFIG_IMA=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_PCR_IDX=10
ima_violations 1 TINFO: CONFIG_IMA_LSM_RULES=y
ima_violations 1 TINFO: CONFIG_IMA_NG_TEMPLATE=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_TEMPLATE="ima-ng" ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH_SHA256=y
ima_violations 1 TINFO: CONFIG_IMA_DEFAULT_HASH="sha256"
ima_violations 1 TINFO: CONFIG_IMA_READ_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE=y
ima_violations 1 TINFO: CONFIG_IMA_ARCH_POLICY=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_BOOTPARAM=y
ima_violations 1 TINFO: CONFIG_IMA_APPRAISE_MODSIG=y
ima_violations 1 TINFO: CONFIG_IMA_MEASURE_ASYMMETRIC_KEYS=y ima_violations 1 TINFO: CONFIG_IMA_QUEUE_EARLY_BOOT_KEYS=y
ima_violations 1 TINFO: CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT=y ima_violations 1 TINFO: CONFIG_IMA_DISABLE_HTABLE=y
ima_violations 1 TINFO: /proc/cmdline: BOOT_IMAGE=/boot/vmlinuz-6.14.0-rc3-1.gb6b4102- default root=UUID=e36b2366-1af2-4408-903c-1fca82c60f4c splash=silent video=1024x768 plymouth.ignore-serial-consoles console=ttyS0 console=tty kernel.softlockup_panic=1 resume=/dev/disk/by-uuid/c3b865f9-5d5b-410e-a6d1-9ebcf721584c mitigations=auto security=apparmor ignore_loglevel ima_policy=tcb
ima_violations 1 TINFO: $TMPDIR is on tmpfs => run on loop device ima_violations 1 TINFO: booted with IMA policy: tcb
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TPASS: 1 open_writers violation(s) added
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TPASS: 1 ToMToU violation(s) added
ima_violations 3 TINFO: verify open_writers using mmapped files tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f tst_test.c:1904: TINFO: Tested kernel: 6.14.0-rc3-1.gb6b4102-default #1 SMP PREEMPT_DYNAMIC Thu Feb 20 12:26:55 UTC 2025 (b6b4102) x86_64 tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz' tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution
tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TPASS: 1 open_writers violation(s) added
ima_mmap.c:41: TPASS: test completed

Summary:
passed   1
failed   0
broken   0
skipped  0
warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TPASS: 1 open_writers violation(s) added ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TPASS: 1 open_writers violation(s) added ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TPASS: 2 open_writers violation(s) added ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TPASS: 1 ToMToU violation(s) added
ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TPASS: 2 ToMToU violation(s) added

Summary:
passed   8
failed   0
broken   0
skipped  0
warnings 0

author maintainer
#10 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 32 lines · Mimi Zohar
Quoted 31 lines · Petr Vorel
Quoted 23 lines · Petr Vorel
Quoted 10 lines

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with the IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA policy?
e.g. cat /sys/kernel/security/ima/policy

Quoted 3 lines

thanks,

Mimi

Now testing on kernel *with* your patches. First run always fails, regardless whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Quoted 8 lines
Quoted 2 lines

Kind regards,
Petr

I'm not seeing that on my test machine.  Could there be other things running on your system causing violations.  In anycase, your original test was less exacting. Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent new "gt" test.

-> "-ge"

Quoted 3 lines

if [ $(($num_violations_new - $num_violations)) -eq $expected_violations ]; then

#11 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 36 lines · Mimi Zohar
Quoted 34 lines · Petr Vorel
Quoted 33 lines
Quoted 23 lines · Petr Vorel
Quoted 10 lines

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with the IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA policy?
e.g. cat /sys/kernel/security/ima/policy

Quoted 3 lines

thanks,

Mimi

Now testing on kernel *with* your patches. First run always fails, regardless whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

Quoted 7 lines
Quoted 5 lines

I'm not seeing that on my test machine.  Could there be other things running on your system causing violations.  In anycase, your original test was less exacting. Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent new "gt" test.

-> "-ge"

Sure, changing to -ge fixes the problem:
if [ $(($num_violations_new - $num_violations)) -ge $expected_violations ]; then

I guess we need "-ge" for older kernels (unless "fix" for stable). Should we accept "$expected_violations || $expected_violations + 1" for new kernels to avoid problems like the one on my system.

I wonder if the problem was somehow caused by the fact that I built kernel. OTOH it's build by OBS (official openSUSE build service).

I don't expect you'd have time to look into it, in case you're interested and have time sending a links to rpm binary and src package.

https://download.opensuse.org/repositories/home:/pevik:/ima-limit-open-writers-ToMToU/standard/x86_64/kernel-default-6.14~rc3-1.1.gb6b4102.x86_64.rpm
https://download.opensuse.org/repositories/home:/pevik:/ima-limit-open-writers-ToMToU/standard/src/kernel-source-6.14~rc3-1.1.gb6b4102.src.rpm

Kind regards,
Petr

Quoted 1 line
Quoted 1 line

if [ $(($num_violations_new - $num_violations)) -eq $expected_violations ]; then

author maintainer
#12 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 44 lines · Petr Vorel
Quoted 43 lines · Mimi Zohar
Quoted 42 lines · Petr Vorel
Quoted 26 lines
Quoted 25 lines · Petr Vorel
Quoted 10 lines

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with the
IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA policy?
e.g. cat /sys/kernel/security/ima/policy

thanks,

Quoted 8 lines
Quoted 1 line

Mimi

Now testing on kernel *with* your patches. First run always fails, regardless whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

I'm not seeing that on my test machine.  Could there be other things running on your system causing violations.  In anycase, your original test was less exacting. Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent new
"gt" test.

Quoted 8 lines
Quoted 1 line

-> "-ge"

Sure, changing to -ge fixes the problem:
if [ $(($num_violations_new - $num_violations)) -ge $expected_violations ]; then

I guess we need "-ge" for older kernels (unless "fix" for stable).  Should we accept "$expected_violations || $expected_violations + 1" for new kernels to avoid problems like the one on my system.

The problem is that we don't control what else is running on the system. So there could be other violations independent of these tests. I'll have to think about it some more and get back to you. (There's no rush to do anything with these LTP IMA violation tests.)

Quoted 3 lines

I wonder if the problem was somehow caused by the fact that I built kernel. OTOH it's build by OBS (official openSUSE build service).

As long as you weren't building the kernel and running the tests at the same, I doubt it would be the problem.

Quoted 3 lines

I don't expect you'd have time to look into it, in case you're interested and have time sending a links to rpm binary and src package.

Ok.

Quoted 4 lines

https://download.opensuse.org/repositories/home:/pevik:/ima-limit-open-writers-ToMToU/standard/x86_64/kernel-default-6.14~rc3-1.1.gb6b4102.x86_64.rpm
https://download.opensuse.org/repositories/home:/pevik:/ima-limit-open-writers-ToMToU/standard/src/kernel-source-6.14~rc3-1.1.gb6b4102.src.rpm

thanks,

Mimi

#13 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 50 lines · Petr Vorel
Quoted 48 lines
Quoted 37 lines · Mimi Zohar
Quoted 35 lines · Petr Vorel
Quoted 34 lines
Quoted 24 lines · Petr Vorel
Quoted 10 lines

Hi Mimi,

Quoted 6 lines

Kernel patch "ima: limit the number of ToMToU integrity violations" prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with the
IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA policy?
e.g. cat /sys/kernel/security/ima/policy

Quoted 3 lines

thanks,

Mimi

Now testing on kernel *with* your patches. First run always fails, regardless whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

Quoted 8 lines
Quoted 6 lines

I'm not seeing that on my test machine.  Could there be other things running on your system causing violations.  In anycase, your original test was less exacting. Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent new
"gt" test.

-> "-ge"

Sure, changing to -ge fixes the problem:
if [ $(($num_violations_new - $num_violations)) -ge $expected_violations ]; then

Quoted 7 lines
Quoted 3 lines

I guess we need "-ge" for older kernels (unless "fix" for stable).  Should we accept "$expected_violations || $expected_violations + 1" for new kernels to avoid problems like the one on my system.

The problem is that we don't control what else is running on the system. So there could be other violations independent of these tests. I'll have to think about it some more and get back to you. (There's no rush to do anything with these LTP IMA violation tests.)

OK, thank you. The worse scenario would be to use less precise variant "-ge".

Quoted 5 lines
Quoted 2 lines

I wonder if the problem was somehow caused by the fact that I built kernel. OTOH it's build by OBS (official openSUSE build service).

As long as you weren't building the kernel and running the tests at the same, I doubt it would be the problem.

Understand, just something on openSUSE Tumbleweed system.

Kind regards,
Petr

Quoted 12 lines
Quoted 2 lines

I don't expect you'd have time to look into it, in case you're interested and have time sending a links to rpm binary and src package.

Ok.

Quoted 2 lines

https://download.opensuse.org/repositories/home:/pevik:/ima-limit-open-writers-ToMToU/standard/x86_64/kernel-default-6.14~rc3-1.1.gb6b4102.x86_64.rpm
https://download.opensuse.org/repositories/home:/pevik:/ima-limit-open-writers-ToMToU/standard/src/kernel-source-6.14~rc3-1.1.gb6b4102.src.rpm

thanks,

Mimi

author maintainer
#14 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 80 lines · Petr Vorel
Quoted 67 lines · Petr Vorel
Quoted 60 lines
Quoted 54 lines · Mimi Zohar
Quoted 43 lines · Petr Vorel
Quoted 34 lines
Quoted 30 lines · Petr Vorel
Quoted 7 lines

Hi Mimi,

Quoted 4 lines

Kernel patch "ima: limit the number of ToMToU integrity violations"
prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Quoted 6 lines
Quoted 3 lines

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel
patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with
the
IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA
policy?
e.g. cat /sys/kernel/security/ima/policy

Quoted 3 lines

thanks,

Mimi

Quoted 7 lines

Now testing on kernel *with* your patches. First run always fails, regardless
whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

Quoted 9 lines

I'm not seeing that on my test machine.  Could there be other things running on your
system causing violations.  In anycase, your original test was less exacting.
Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent
new
"gt" test.

Quoted 1 line

-> "-ge"

Sure, changing to -ge fixes the problem:
if [ $(($num_violations_new - $num_violations)) -ge $expected_violations ]; then

Quoted 5 lines

I guess we need "-ge" for older kernels (unless "fix" for stable).  Should we
accept "$expected_violations || $expected_violations + 1" for new kernels to
avoid problems like the one on my system.

Quoted 6 lines

The problem is that we don't control what else is running on the system.  So there could
be other violations independent of these tests.  I'll have to think about it some more and
get back to you.  (There's no rush to do anything with these LTP IMA violation tests.)

OK, thank you. The worse scenario would be to use less precise variant "-ge".

Quoted 3 lines
Quoted 3 lines

I wonder if the problem was somehow caused by the fact that I built kernel. OTOH
it's build by OBS (official openSUSE build service).

Quoted 5 lines
Quoted 3 lines

As long as you weren't building the kernel and running the tests at the same, I doubt it
would be the problem.

Understand, just something on openSUSE Tumbleweed system.

Peter, thank you for the tumbleweed image.

The default IMA tcb policy results is measuring $LOG (/var/log/audit/audit.log) on the first call to validate(). To prevent that from interfering with test1, I would add the following line or something similar in setup() to force measuring $LOG to happen earlier.

exec 3< $LOG || exit 1

Assuming that works, I'll update the kernel and LTP tests.

thanks,

Mimi

#15 lore raw

Re: [RFC PATCH 3/3] ima: additional ToMToU violation tests

Quoted 81 lines · Petr Vorel
Quoted 79 lines
Quoted 67 lines · Petr Vorel
Quoted 60 lines
Quoted 54 lines · Mimi Zohar
Quoted 43 lines · Petr Vorel
Quoted 34 lines
Quoted 30 lines · Petr Vorel
Quoted 7 lines

Hi Mimi,

Quoted 4 lines

Kernel patch "ima: limit the number of ToMToU integrity violations"
prevents superfluous ToMToU violations.  Add corresponding LTP tests.

Quoted 6 lines
Quoted 3 lines

Link:
https://lore.kernel.org/linux-integrity/[email protected]/

Unfortunately tests fail on both mainline kernel and kernel with your patches.

The new LTP IMA violations patches should fail without the associated kernel
patches.

Quoted 1 line

Any hint what could be wrong?

Of course it's dependent on the IMA policy.  The tests assume being booted with
the
IMA
TCB measurement policy or similar policy being loaded.  Can you share the IMA
policy?
e.g. cat /sys/kernel/security/ima/policy

Quoted 3 lines

thanks,

Mimi

Quoted 7 lines

Now testing on kernel *with* your patches. First run always fails, regardless
whether using ima_policy=tcb or /opt/ltp/testcases/data/ima_violations/violations.policy).

Kind regards,
Petr

Quoted 9 lines

I'm not seeing that on my test machine.  Could there be other things running on your
system causing violations.  In anycase, your original test was less exacting.
Similarly,
instead of "-eq", try using "-qe" in the following test and removing the subsequent
new
"gt" test.

Quoted 1 line

-> "-ge"

Sure, changing to -ge fixes the problem:
if [ $(($num_violations_new - $num_violations)) -ge $expected_violations ]; then

Quoted 5 lines

I guess we need "-ge" for older kernels (unless "fix" for stable).  Should we
accept "$expected_violations || $expected_violations + 1" for new kernels to
avoid problems like the one on my system.

Quoted 6 lines

The problem is that we don't control what else is running on the system.  So there could
be other violations independent of these tests.  I'll have to think about it some more and
get back to you.  (There's no rush to do anything with these LTP IMA violation tests.)

OK, thank you. The worse scenario would be to use less precise variant "-ge".

Quoted 3 lines
Quoted 3 lines

I wonder if the problem was somehow caused by the fact that I built kernel. OTOH
it's build by OBS (official openSUSE build service).

Quoted 5 lines
Quoted 5 lines
Quoted 3 lines

As long as you weren't building the kernel and running the tests at the same, I doubt it
would be the problem.

Understand, just something on openSUSE Tumbleweed system.

Hi Mimi,

Quoted 1 line

Peter, thank you for the tumbleweed image.

Thanks for debugging on the image!

Quoted 4 lines

The default IMA tcb policy results is measuring $LOG (/var/log/audit/audit.log) on the first call to validate(). To prevent that from interfering with test1, I would add the following line or something similar in setup() to force measuring $LOG to happen earlier.

+1

Quoted 1 line

exec 3< $LOG || exit 1

Ideally use:
exec 3< $LOG || tst_brk TBROK "some explanation..."

Quoted 1 line

Assuming that works, I'll update the kernel and LTP tests.

+1 (patch from you is preferred)

Kind regards,
Petr

Quoted 3 lines

thanks,

Mimi

#16 lore raw
In reply to Mimi Zohar

Hi Mimi,

Quoted 13 lines

Add support for the number of expected violations. Include the expected number of violations in the output.

---
 .../security/integrity/ima/tests/ima_violations.sh     | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +0 -0
index 37d8d473c..7f0382fb8 100755

testcases/kernel/security/integrity/ima/tests/ima_violations.sh validate() · line 74

Quoted 4 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +1 -0
@@ -71,20 +71,26 @@ validate()
 	local num_violations="$1"
 	local count="$2"
 	local search="$3"
+	local expected_violations=$4

nit: safer to quote as much as possible (="$4") to avoid errors.

testcases/kernel/security/integrity/ima/tests/ima_violations.sh validate() · line 78

Quoted 4 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +1 -0
 	local max_attempt=3
 	local count2 i num_violations_new
 
+	[ -z $expected_violations ] && expected_violations=1

Also here: -z "$expected_violations"

I can add quotes before merge if you don't want to bother (I would send you a diff to ack it before merging).

Quoted 6 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +2 -1
+
 	for i in $(seq 1 $max_attempt); do
 		read num_violations_new < $IMA_VIOLATIONS
 		count2="$(get_count $search)"
-		if [ $(($num_violations_new - $num_violations)) -gt 0 ]; then
+		if [ $(($num_violations_new - $num_violations)) -eq $expected_violations ]; then

testcases/kernel/security/integrity/ima/tests/ima_violations.sh validate() · line 92

Quoted 10 lines
testcases/kernel/security/integrity/ima/tests/ima_violations.sh +3 -1
 			if [ $count2 -gt $count ]; then
-				tst_res TPASS "$search violation added"
+				tst_res TPASS "$expected_violations $search violation(s) added"
 				return
 			else
 				tst_res TINFO "$search not found in $LOG ($i/$max_attempt attempt)..."
 				tst_sleep 1s
 			fi
+		elif [ $(($num_violations_new - $num_violations)) -gt 0 ]; then
+			tst_res $IMA_FAIL "$search too many violations added"

nit: maybe print values for debugging?

tst_res $IMA_FAIL "$search too many violations added: $num_violations_new - $num_violations"

FYI failing tests has 2 or 3 higher:

ima_violations 1 TINFO: SUT has required policy content
ima_violations 1 TINFO: using log /var/log/audit/audit.log
ima_violations 1 TINFO: verify open writers violation
ima_violations 1 TFAIL: open_writers too many violations added: 106 - 104
ima_violations 2 TINFO: verify ToMToU violation
ima_violations 2 TFAIL: ToMToU too many violations added: 109 - 107
ima_violations 3 TINFO: verify open_writers using mmapped files
tst_test.c:1900: TINFO: LTP version: 20250130-22-gcd2215702f
tst_test.c:1904: TINFO: Tested kernel: 6.13.0-2.g0127a37-default #1 SMP PREEMPT_DYNAMIC Thu Jan 23 11:21:55 UTC 2025 (0127a37) x86_64 tst_kconfig.c:88: TINFO: Parsing kernel config '/proc/config.gz'
tst_kconfig.c:676: TINFO: CONFIG_FAULT_INJECTION kernel option detected which might slow the execution tst_test.c:1722: TINFO: Overall timeout per run is 0h 02m 00s
ima_mmap.c:38: TINFO: sleep 3s
ima_violations 3 TFAIL: open_writers too many violations added: 112 - 110
ima_mmap.c:41: TPASS: test completed

Summary:

passed   1
failed   0
broken   0
skipped  0

warnings 0
ima_violations 4 TINFO: verify limiting single open writer violation ima_violations 4 TFAIL: open_writers too many violations added: 116 - 113 ima_violations 5 TINFO: verify limiting multiple open writers violations ima_violations 5 TFAIL: open_writers too many violations added: 121 - 117 ima_violations 6 TINFO: verify new open writer causes additional violation ima_violations 6 TFAIL: open_writers too many violations added: 126 - 122 ima_violations 7 TINFO: verify limiting single open reader ToMToU violations ima_violations 7 TFAIL: ToMToU too many violations added: 130 - 127 ima_violations 8 TINFO: verify new open reader causes additional ToMToU violation ima_violations 8 TFAIL: ToMToU too many violations added: 134 - 131

As I noted in previous mail, either has of a backport (can be added later, we don't have to wait for merging) or skip on older kernels (tst_kvcmp -lt ...).

Kind regards,
Petr