Skip to contentSkip to conversation

add clock_adjtime to certification rulesets

PATCH In review Richard Guy Briggs 2 participants 3 messages Nov 4, 2021

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 add clock_adjtime to certification rulesets

The clock_adjtime syscall is missing from several certification rulesets that monitor changes to the system clock. Add it.

Please see https://bugzilla.redhat.com/show_bug.cgi?id=1991919

---
 rules/30-nispom.rules      | 4 ++--
 rules/30-pci-dss-v31.rules | 4 ++--
 rules/30-stig.rules        | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)
rules/30-nispom.rules +2 -2
index e3873ef95069..ecac01a0b4e1 100644
@@ -10,8 +10,8 @@
 ## Things that could affect time
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F key=time-change
 -a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=time-change
--a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=time-change
--a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=time-change
+-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change
+-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change
 # Introduced in 2.6.39, commented out because it can make false positives
 #-a always,exit -F arch=b32 -S clock_adjtime -F key=time-change
 #-a always,exit -F arch=b64 -S clock_adjtime -F key=time-change
rules/30-pci-dss-v31.rules +2 -2
index 7062b35f165c..0251bcafcc03 100644
@@ -77,8 +77,8 @@
 ## We will place rules to check time synchronization
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F key=10.4.2b-time-change
 -a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=10.4.2b-time-change
--a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=10.4.2b-time-change
--a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=10.4.2b-time-change
+-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=10.4.2b-time-change
+-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=10.4.2b-time-change
 # Introduced in 2.6.39, commented out because it can make false positives
 #-a always,exit -F arch=b32 -S clock_adjtime -F key=10.4.2b-time-change
 #-a always,exit -F arch=b64 -S clock_adjtime -F key=10.4.2b-time-change
rules/30-stig.rules +2 -2
index 234f239cac06..60384f6b247d 100644
@@ -26,8 +26,8 @@
 ## Things that could affect time
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F key=time-change
 -a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=time-change
--a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=time-change
--a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=time-change
+-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change
+-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change
 # Introduced in 2.6.39, commented out because it can make false positives
 #-a always,exit -F arch=b32 -S clock_adjtime -F key=time-change
 #-a always,exit -F arch=b64 -S clock_adjtime -F key=time-change
Signature · 6 lines
2.27.0

--
Linux-audit mailing list
[email protected]
https://listman.redhat.com/mailman/listinfo/linux-audit
#1 lore raw

Hello Richard,

Quoted 24 lines · EDT Richard Guy Briggs

The clock_adjtime syscall is missing from several certification rulesets that monitor changes to the system clock. Add it.

Please see https://bugzilla.redhat.com/show_bug.cgi?id=1991919

---
 rules/30-nispom.rules      | 4 ++--
 rules/30-pci-dss-v31.rules | 4 ++--
 rules/30-stig.rules        | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)
rules/30-nispom.rules +1 -3
index e3873ef95069..ecac01a0b4e1 100644
@@ -10,8 +10,8 @@
 ## Things that could affect time
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F key=time-
 change
-a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=time-change
--a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=time-change
--a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=time-change
+-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F
Quoted 6 lines

key=time-change
+-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change
# Introduced in 2.6.39, commented out because it can make false positives #-a always,exit -F arch=b32 -S clock_adjtime -F key=time-change #-a always,exit -F arch=b64 -S clock_adjtime -F key=time-change

Not sure if you saw this, but directly under the change is a rule for that syscall, but it's commented out. This is because it is a very noisy event. Anyone that wants it can enable it.

-Steve

Quoted 26 lines
rules/30-pci-dss-v31.rules +0 -0
index 7062b35f165c..0251bcafcc03 100644
@@ -77,8 +77,8 @@
 ## We will place rules to check time synchronization
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F

key=10.4.2b-time-change -a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=10.4.2b-time-change --a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=10.4.2b-time-change --a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=10.4.2b-time-change +-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=10.4.2b-time-change +-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=10.4.2b-time-change # Introduced in 2.6.39, commented out because it can make false positives #-a always,exit -F arch=b32 -S clock_adjtime -F key=10.4.2b-time-change #-a always,exit -F arch=b64 -S clock_adjtime -F key=10.4.2b-time-change diff --git a/rules/30-stig.rules b/rules/30-stig.rules
index 234f239cac06..60384f6b247d 100644

rules/30-stig.rules +0 -0
@@ -26,8 +26,8 @@
 ## Things that could affect time
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F

key=time-change -a always,exit -F arch=b64 -S adjtimex,settimeofday -F

Quoted 8 lines

key=time-change --a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=time-change --a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=time-change +-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change +-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change # Introduced in 2.6.39, commented out because it can make false positives #-a always,exit -F arch=b32 -S clock_adjtime -F key=time-change
#-a always,exit -F arch=b64 -S clock_adjtime -F key=time-change

--
Linux-audit mailing list
[email protected]
https://listman.redhat.com/mailman/listinfo/linux-audit

#2 lore raw
Quoted 29 lines · Steve Grubb

Hello Richard,

Quoted 26 lines · EDT Richard Guy Briggs

The clock_adjtime syscall is missing from several certification rulesets that monitor changes to the system clock. Add it.

Please see https://bugzilla.redhat.com/show_bug.cgi?id=1991919

---
 rules/30-nispom.rules      | 4 ++--
 rules/30-pci-dss-v31.rules | 4 ++--
 rules/30-stig.rules        | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)
rules/30-nispom.rules +1 -3
index e3873ef95069..ecac01a0b4e1 100644
@@ -10,8 +10,8 @@
 ## Things that could affect time
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F key=time-
 change
-a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=time-change
--a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=time-change
--a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=time-change
+-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F

key=time-change
+-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F

Quoted 8 lines
Quoted 4 lines

key=time-change
# Introduced in 2.6.39, commented out because it can make false positives #-a always,exit -F arch=b32 -S clock_adjtime -F key=time-change #-a always,exit -F arch=b64 -S clock_adjtime -F key=time-change

Not sure if you saw this, but directly under the change is a rule for that syscall, but it's commented out. This is because it is a very noisy event. Anyone that wants it can enable it.

Well, how about that! No, I didn't see it, not even when I did a git grep for that symbol. Not sure how I missed that. Yes, that rule *is* noisy. f35 uses clock_adjtime while rhel8 uses adjtimex for the same use, which is why I noticed the missing syscall.

Sorry for my noise, please ignore this patch. <blush>

Quoted 28 lines

-Steve

Quoted 26 lines
rules/30-pci-dss-v31.rules +0 -0
index 7062b35f165c..0251bcafcc03 100644
@@ -77,8 +77,8 @@
 ## We will place rules to check time synchronization
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F

key=10.4.2b-time-change -a always,exit -F arch=b64 -S adjtimex,settimeofday -F key=10.4.2b-time-change --a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=10.4.2b-time-change --a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=10.4.2b-time-change +-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=10.4.2b-time-change +-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=10.4.2b-time-change # Introduced in 2.6.39, commented out because it can make false positives #-a always,exit -F arch=b32 -S clock_adjtime -F key=10.4.2b-time-change #-a always,exit -F arch=b64 -S clock_adjtime -F key=10.4.2b-time-change diff --git a/rules/30-stig.rules b/rules/30-stig.rules
index 234f239cac06..60384f6b247d 100644

rules/30-stig.rules +0 -0
@@ -26,8 +26,8 @@
 ## Things that could affect time
 -a always,exit -F arch=b32 -S adjtimex,settimeofday,stime -F

key=time-change -a always,exit -F arch=b64 -S adjtimex,settimeofday -F

Quoted 8 lines
Quoted 8 lines

key=time-change --a always,exit -F arch=b32 -S clock_settime -F a0=0x0 -F key=time-change --a always,exit -F arch=b64 -S clock_settime -F a0=0x0 -F key=time-change +-a always,exit -F arch=b32 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change +-a always,exit -F arch=b64 -S clock_settime,clock_adjtime -F a0=0x0 -F key=time-change # Introduced in 2.6.39, commented out because it can make false positives #-a always,exit -F arch=b32 -S clock_adjtime -F key=time-change
#-a always,exit -F arch=b64 -S clock_adjtime -F key=time-change

- RGB

--
Richard Guy Briggs <[email protected]>
Sr. S/W Engineer, Kernel Security, Base Operating Systems Remote, Ottawa, Red Hat Canada
IRC: rgb, SunRaycer
Voice: +1.647.777.2635, Internal: (81) 32635

--
Linux-audit mailing list
[email protected]
https://listman.redhat.com/mailman/listinfo/linux-audit