Openembedded Core Discussions
 help / color / mirror / Atom feed
* [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866
@ 2019-01-28 11:17 Marcus Cooper
  2019-01-28 11:17 ` [PATCH v2 1/3] systemd: Security fix CVE-2018-16864 Marcus Cooper
                   ` (3 more replies)
  0 siblings, 4 replies; 7+ messages in thread
From: Marcus Cooper @ 2019-01-28 11:17 UTC (permalink / raw)
  To: openembedded-core

Changed in v2:
 - Added CVE tag, Upstream-Status tag and Sign-off-by tags.
 - removed the verification of the entry length in the header
 - squashed CVE-2018-16865 patches into one
 - CVE-2018-16866 patch now taken from systemd-stable and includes
   an additional heap buffer overflow fix.

Marcus Cooper (3):
  systemd: Security fix CVE-2018-16864
  systemd: Security fix CVE-2018-16865
  systemd: Security fix CVE-2018-16866

 ...-not-store-the-iovec-entry-for-process-co.patch | 208 +++++++++++++++++++++
 ...rnald-set-a-limit-on-the-number-of-fields.patch | 139 ++++++++++++++
 ...nal-fix-out-of-bounds-read-CVE-2018-16866.patch |  49 +++++
 meta/recipes-core/systemd/systemd_239.bb           |   3 +
 4 files changed, 399 insertions(+)
 create mode 100644 meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
 create mode 100644 meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch
 create mode 100644 meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch

-- 
2.11.0



^ permalink raw reply	[flat|nested] 7+ messages in thread

* [PATCH v2 1/3] systemd: Security fix CVE-2018-16864
  2019-01-28 11:17 [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Marcus Cooper
@ 2019-01-28 11:17 ` Marcus Cooper
  2019-01-28 11:17 ` [PATCH v2 2/3] systemd: Security fix CVE-2018-16865 Marcus Cooper
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 7+ messages in thread
From: Marcus Cooper @ 2019-01-28 11:17 UTC (permalink / raw)
  To: openembedded-core

Affects < v240

Signed-off-by: Marcus Cooper <marcusc@axis.com>
---
 ...-not-store-the-iovec-entry-for-process-co.patch | 208 +++++++++++++++++++++
 meta/recipes-core/systemd/systemd_239.bb           |   1 +
 2 files changed, 209 insertions(+)
 create mode 100644 meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch

diff --git a/meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch b/meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
new file mode 100644
index 0000000000..c2f78be39e
--- /dev/null
+++ b/meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
@@ -0,0 +1,208 @@
+From 9cb07e7d82c7c4f28bbaa1478e1387e8ea3d03dd Mon Sep 17 00:00:00 2001
+From: =?UTF-8?q?Zbigniew=20J=C4=99drzejewski-Szmek?= <zbyszek@in.waw.pl>
+Date: Wed, 5 Dec 2018 18:38:39 +0100
+Subject: [PATCH] journald: do not store the iovec entry for process
+ commandline on stack
+
+This fixes a crash where we would read the commandline, whose length is under
+control of the sending program, and then crash when trying to create a stack
+allocation for it.
+
+CVE-2018-16864
+https://bugzilla.redhat.com/show_bug.cgi?id=1653855
+
+The message actually doesn't get written to disk, because
+journal_file_append_entry() returns -E2BIG.
+
+Patch backported from systemd master at
+084eeb865ca63887098e0945fb4e93c852b91b0f.
+
+CVE: CVE-2018-16864
+Upstream-Status: Backport
+Signed-off-by: Marcus Cooper <marcusc@axis.com>
+---
+ src/basic/io-util.c           | 10 ++++++++++
+ src/basic/io-util.h           |  2 ++
+ src/coredump/coredump.c       | 31 +++++++++++--------------------
+ src/journal/journald-server.c | 25 +++++++++++++++----------
+ 4 files changed, 38 insertions(+), 30 deletions(-)
+
+diff --git a/src/basic/io-util.c b/src/basic/io-util.c
+index 1f64cc933b..575398fbe6 100644
+--- a/src/basic/io-util.c
++++ b/src/basic/io-util.c
+@@ -8,6 +8,7 @@
+ #include <unistd.h>
+ 
+ #include "io-util.h"
++#include "string-util.h"
+ #include "time-util.h"
+ 
+ int flush_fd(int fd) {
+@@ -252,3 +253,12 @@ ssize_t sparse_write(int fd, const void *p, size_t sz, size_t run_length) {
+ 
+         return q - (const uint8_t*) p;
+ }
++
++char* set_iovec_string_field(struct iovec *iovec, size_t *n_iovec, const char *field, const char *value) {
++        char *x;
++
++        x = strappend(field, value);
++        if (x)
++                iovec[(*n_iovec)++] = IOVEC_MAKE_STRING(x);
++        return x;
++}
+diff --git a/src/basic/io-util.h b/src/basic/io-util.h
+index ed189b5820..792a64ad5e 100644
+--- a/src/basic/io-util.h
++++ b/src/basic/io-util.h
+@@ -71,3 +71,5 @@ static inline bool FILE_SIZE_VALID_OR_INFINITY(uint64_t l) {
+ #define IOVEC_MAKE(base, len) (struct iovec) IOVEC_INIT(base, len)
+ #define IOVEC_INIT_STRING(string) IOVEC_INIT((char*) string, strlen(string))
+ #define IOVEC_MAKE_STRING(string) (struct iovec) IOVEC_INIT_STRING(string)
++
++char* set_iovec_string_field(struct iovec *iovec, size_t *n_iovec, const char *field, const char *value);
+diff --git a/src/coredump/coredump.c b/src/coredump/coredump.c
+index 20a1cbdd45..18e4f61d72 100644
+--- a/src/coredump/coredump.c
++++ b/src/coredump/coredump.c
+@@ -1053,19 +1053,10 @@ static int send_iovec(const struct iovec iovec[], size_t n_iovec, int input_fd)
+         return 0;
+ }
+ 
+-static char* set_iovec_field(struct iovec *iovec, size_t *n_iovec, const char *field, const char *value) {
+-        char *x;
+-
+-        x = strappend(field, value);
+-        if (x)
+-                iovec[(*n_iovec)++] = IOVEC_MAKE_STRING(x);
+-        return x;
+-}
+-
+ static char* set_iovec_field_free(struct iovec *iovec, size_t *n_iovec, const char *field, char *value) {
+         char *x;
+ 
+-        x = set_iovec_field(iovec, n_iovec, field, value);
++        x = set_iovec_string_field(iovec, n_iovec, field, value);
+         free(value);
+         return x;
+ }
+@@ -1115,36 +1106,36 @@ static int gather_pid_metadata(
+                         disable_coredumps();
+                 }
+ 
+-                set_iovec_field(iovec, n_iovec, "COREDUMP_UNIT=", context[CONTEXT_UNIT]);
++                set_iovec_string_field(iovec, n_iovec, "COREDUMP_UNIT=", context[CONTEXT_UNIT]);
+         }
+ 
+         if (cg_pid_get_user_unit(pid, &t) >= 0)
+                 set_iovec_field_free(iovec, n_iovec, "COREDUMP_USER_UNIT=", t);
+ 
+         /* The next few are mandatory */
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_PID=", context[CONTEXT_PID]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_PID=", context[CONTEXT_PID]))
+                 return log_oom();
+ 
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_UID=", context[CONTEXT_UID]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_UID=", context[CONTEXT_UID]))
+                 return log_oom();
+ 
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_GID=", context[CONTEXT_GID]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_GID=", context[CONTEXT_GID]))
+                 return log_oom();
+ 
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_SIGNAL=", context[CONTEXT_SIGNAL]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_SIGNAL=", context[CONTEXT_SIGNAL]))
+                 return log_oom();
+ 
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_RLIMIT=", context[CONTEXT_RLIMIT]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_RLIMIT=", context[CONTEXT_RLIMIT]))
+                 return log_oom();
+ 
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_HOSTNAME=", context[CONTEXT_HOSTNAME]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_HOSTNAME=", context[CONTEXT_HOSTNAME]))
+                 return log_oom();
+ 
+-        if (!set_iovec_field(iovec, n_iovec, "COREDUMP_COMM=", context[CONTEXT_COMM]))
++        if (!set_iovec_string_field(iovec, n_iovec, "COREDUMP_COMM=", context[CONTEXT_COMM]))
+                 return log_oom();
+ 
+         if (context[CONTEXT_EXE] &&
+-            !set_iovec_field(iovec, n_iovec, "COREDUMP_EXE=", context[CONTEXT_EXE]))
++            !set_iovec_string_field(iovec, n_iovec, "COREDUMP_EXE=", context[CONTEXT_EXE]))
+                 return log_oom();
+ 
+         if (sd_pid_get_session(pid, &t) >= 0)
+@@ -1212,7 +1203,7 @@ static int gather_pid_metadata(
+                 iovec[(*n_iovec)++] = IOVEC_MAKE_STRING(t);
+ 
+         if (safe_atoi(context[CONTEXT_SIGNAL], &signo) >= 0 && SIGNAL_VALID(signo))
+-                set_iovec_field(iovec, n_iovec, "COREDUMP_SIGNAL_NAME=SIG", signal_to_string(signo));
++                set_iovec_string_field(iovec, n_iovec, "COREDUMP_SIGNAL_NAME=SIG", signal_to_string(signo));
+ 
+         return 0; /* we successfully acquired all metadata */
+ }
+diff --git a/src/journal/journald-server.c b/src/journal/journald-server.c
+index 4f1550ec5b..31be085c6b 100644
+--- a/src/journal/journald-server.c
++++ b/src/journal/journald-server.c
+@@ -753,6 +753,7 @@ static void dispatch_message_real(
+                 pid_t object_pid) {
+ 
+         char source_time[sizeof("_SOURCE_REALTIME_TIMESTAMP=") + DECIMAL_STR_MAX(usec_t)];
++        _cleanup_free_ char *cmdline1 = NULL, *cmdline2 = NULL;
+         uid_t journal_uid;
+         ClientContext *o;
+ 
+@@ -769,20 +770,23 @@ static void dispatch_message_real(
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, c->uid, uid_t, uid_is_valid, UID_FMT, "_UID");
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, c->gid, gid_t, gid_is_valid, GID_FMT, "_GID");
+ 
+-                IOVEC_ADD_STRING_FIELD(iovec, n, c->comm, "_COMM");
+-                IOVEC_ADD_STRING_FIELD(iovec, n, c->exe, "_EXE");
+-                IOVEC_ADD_STRING_FIELD(iovec, n, c->cmdline, "_CMDLINE");
+-                IOVEC_ADD_STRING_FIELD(iovec, n, c->capeff, "_CAP_EFFECTIVE");
++                IOVEC_ADD_STRING_FIELD(iovec, n, c->comm, "_COMM"); /* At most TASK_COMM_LENGTH (16 bytes) */
++                IOVEC_ADD_STRING_FIELD(iovec, n, c->exe, "_EXE"); /* A path, so at most PATH_MAX (4096 bytes) */
+ 
+-                IOVEC_ADD_SIZED_FIELD(iovec, n, c->label, c->label_size, "_SELINUX_CONTEXT");
++                if (c->cmdline)
++                        /* At most _SC_ARG_MAX (2MB usually), which is too much to put on stack.
++                         * Let's use a heap allocation for this one. */
++                        cmdline1 = set_iovec_string_field(iovec, &n, "_CMDLINE=", c->cmdline);
+ 
++                IOVEC_ADD_STRING_FIELD(iovec, n, c->capeff, "_CAP_EFFECTIVE"); /* Read from /proc/.../status */
++                IOVEC_ADD_SIZED_FIELD(iovec, n, c->label, c->label_size, "_SELINUX_CONTEXT");
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, c->auditid, uint32_t, audit_session_is_valid, "%" PRIu32, "_AUDIT_SESSION");
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, c->loginuid, uid_t, uid_is_valid, UID_FMT, "_AUDIT_LOGINUID");
+ 
+-                IOVEC_ADD_STRING_FIELD(iovec, n, c->cgroup, "_SYSTEMD_CGROUP");
++                IOVEC_ADD_STRING_FIELD(iovec, n, c->cgroup, "_SYSTEMD_CGROUP"); /* A path */
+                 IOVEC_ADD_STRING_FIELD(iovec, n, c->session, "_SYSTEMD_SESSION");
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, c->owner_uid, uid_t, uid_is_valid, UID_FMT, "_SYSTEMD_OWNER_UID");
+-                IOVEC_ADD_STRING_FIELD(iovec, n, c->unit, "_SYSTEMD_UNIT");
++                IOVEC_ADD_STRING_FIELD(iovec, n, c->unit, "_SYSTEMD_UNIT"); /* Unit names are bounded by UNIT_NAME_MAX */
+                 IOVEC_ADD_STRING_FIELD(iovec, n, c->user_unit, "_SYSTEMD_USER_UNIT");
+                 IOVEC_ADD_STRING_FIELD(iovec, n, c->slice, "_SYSTEMD_SLICE");
+                 IOVEC_ADD_STRING_FIELD(iovec, n, c->user_slice, "_SYSTEMD_USER_SLICE");
+@@ -803,13 +807,14 @@ static void dispatch_message_real(
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, o->uid, uid_t, uid_is_valid, UID_FMT, "OBJECT_UID");
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, o->gid, gid_t, gid_is_valid, GID_FMT, "OBJECT_GID");
+ 
++                /* See above for size limits, only ->cmdline may be large, so use a heap allocation for it. */
+                 IOVEC_ADD_STRING_FIELD(iovec, n, o->comm, "OBJECT_COMM");
+                 IOVEC_ADD_STRING_FIELD(iovec, n, o->exe, "OBJECT_EXE");
+-                IOVEC_ADD_STRING_FIELD(iovec, n, o->cmdline, "OBJECT_CMDLINE");
+-                IOVEC_ADD_STRING_FIELD(iovec, n, o->capeff, "OBJECT_CAP_EFFECTIVE");
++                if (o->cmdline)
++                        cmdline2 = set_iovec_string_field(iovec, &n, "OBJECT_CMDLINE=", o->cmdline);
+ 
++                IOVEC_ADD_STRING_FIELD(iovec, n, o->capeff, "OBJECT_CAP_EFFECTIVE");
+                 IOVEC_ADD_SIZED_FIELD(iovec, n, o->label, o->label_size, "OBJECT_SELINUX_CONTEXT");
+-
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, o->auditid, uint32_t, audit_session_is_valid, "%" PRIu32, "OBJECT_AUDIT_SESSION");
+                 IOVEC_ADD_NUMERIC_FIELD(iovec, n, o->loginuid, uid_t, uid_is_valid, UID_FMT, "OBJECT_AUDIT_LOGINUID");
+ 
+-- 
+2.11.0
+
diff --git a/meta/recipes-core/systemd/systemd_239.bb b/meta/recipes-core/systemd/systemd_239.bb
index c146897508..94729a6a39 100644
--- a/meta/recipes-core/systemd/systemd_239.bb
+++ b/meta/recipes-core/systemd/systemd_239.bb
@@ -38,6 +38,7 @@ SRC_URI += "file://touchscreen.rules \
            file://0001-sysctl-Don-t-pass-null-directive-argument-to-s.patch \
            file://0002-core-Fix-use-after-free-case-in-load_from_path.patch \
            file://0001-meson-rename-Ddebug-to-Ddebug-extra.patch \
+           file://0024-journald-do-not-store-the-iovec-entry-for-process-co.patch \
            "
 
 # patches made for musl are only applied on TCLIBC is musl
-- 
2.11.0



^ permalink raw reply related	[flat|nested] 7+ messages in thread

* [PATCH v2 2/3] systemd: Security fix CVE-2018-16865
  2019-01-28 11:17 [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Marcus Cooper
  2019-01-28 11:17 ` [PATCH v2 1/3] systemd: Security fix CVE-2018-16864 Marcus Cooper
@ 2019-01-28 11:17 ` Marcus Cooper
  2019-01-28 11:17 ` [PATCH v2 3/3] systemd: Security fix CVE-2018-16866 Marcus Cooper
  2019-01-28 11:20 ` [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Alexander Kanavin
  3 siblings, 0 replies; 7+ messages in thread
From: Marcus Cooper @ 2019-01-28 11:17 UTC (permalink / raw)
  To: openembedded-core

Affects < v240

Signed-off-by: Marcus Cooper <marcusc@axis.com>
---
 ...rnald-set-a-limit-on-the-number-of-fields.patch | 139 +++++++++++++++++++++
 meta/recipes-core/systemd/systemd_239.bb           |   1 +
 2 files changed, 140 insertions(+)
 create mode 100644 meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch

diff --git a/meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch b/meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch
new file mode 100644
index 0000000000..ae9ef5de56
--- /dev/null
+++ b/meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch
@@ -0,0 +1,139 @@
+From 7cad044b72406cbadf048da432c29afea74c3c10 Mon Sep 17 00:00:00 2001
+From: =?UTF-8?q?Zbigniew=20J=C4=99drzejewski-Szmek?= <zbyszek@in.waw.pl>
+Date: Wed, 5 Dec 2018 22:45:02 +0100
+Subject: [PATCH] journald: set a limit on the number of fields
+
+The fix for CVE-2018-16865 is plucked from two commits that have
+been pushed to systemd master.
+
+journald: set a limit on the number of fields (1k)
+
+We allocate a iovec entry for each field, so with many short entries,
+our memory usage and processing time can be large, even with a relatively
+small message size. Let's refuse overly long entries.
+
+CVE-2018-16865
+https://bugzilla.redhat.com/show_bug.cgi?id=1653861
+
+What from I can see, the problem is not from an alloca, despite what the CVE
+description says, but from the attack multiplication that comes from creating
+many very small iovecs: (void* + size_t) for each three bytes of input message.
+
+Patch backported from systemd master at
+052c57f132f04a3cf4148f87561618da1a6908b4.
+
+journal-remote: set a limit on the number of fields in a message
+
+Existing use of E2BIG is replaced with ENOBUFS (entry too long), and E2BIG is
+reused for the new error condition (too many fields).
+
+This matches the change done for systemd-journald, hence forming the second
+part of the fix for CVE-2018-16865
+(https://bugzilla.redhat.com/show_bug.cgi?id=1653861).
+
+Patch backported from systemd master at
+ef4d6abe7c7fab6cbff975b32e76b09feee56074.
+with the changes applied by 7fdb237f5473cb8fc2129e57e8a0039526dcb4fd
+removed.
+
+CVE: CVE-2018-16865
+Upstream-Status: Backport
+Signed-off-by: Marcus Cooper <marcusc@axis.com>
+---
+ src/basic/journal-importer.c             | 5 ++++-
+ src/basic/journal-importer.h             | 3 +++
+ src/journal-remote/journal-remote-main.c | 7 ++++++-
+ src/journal-remote/journal-remote.c      | 5 ++++-
+ src/journal/journald-native.c            | 5 +++++
+ 5 files changed, 22 insertions(+), 3 deletions(-)
+
+diff --git a/src/basic/journal-importer.c b/src/basic/journal-importer.c
+index ca203bbbfc..3ac55a66d9 100644
+--- a/src/basic/journal-importer.c
++++ b/src/basic/journal-importer.c
+@@ -23,6 +23,9 @@ enum {
+ };
+ 
+ static int iovw_put(struct iovec_wrapper *iovw, void* data, size_t len) {
++        if (iovw->count >= ENTRY_FIELD_COUNT_MAX)
++                return -E2BIG;
++
+         if (!GREEDY_REALLOC(iovw->iovec, iovw->size_bytes, iovw->count + 1))
+                 return log_oom();
+ 
+@@ -98,7 +101,7 @@ static int get_line(JournalImporter *imp, char **line, size_t *size) {
+                 imp->scanned = imp->filled;
+                 if (imp->scanned >= DATA_SIZE_MAX) {
+                         log_error("Entry is bigger than %u bytes.", DATA_SIZE_MAX);
+-                        return -E2BIG;
++                        return -ENOBUFS;
+                 }
+ 
+                 if (imp->passive_fd)
+diff --git a/src/basic/journal-importer.h b/src/basic/journal-importer.h
+index f49ce734a1..c4ae45d32d 100644
+--- a/src/basic/journal-importer.h
++++ b/src/basic/journal-importer.h
+@@ -16,6 +16,9 @@
+ #define DATA_SIZE_MAX (1024*1024*768u)
+ #define LINE_CHUNK 8*1024u
+ 
++/* The maximum number of fields in an entry */
++#define ENTRY_FIELD_COUNT_MAX 1024
++
+ struct iovec_wrapper {
+         struct iovec *iovec;
+         size_t size_bytes;
+diff --git a/src/journal-remote/journal-remote-main.c b/src/journal-remote/journal-remote-main.c
+index 8fda9d1499..3a01fef646 100644
+--- a/src/journal-remote/journal-remote-main.c
++++ b/src/journal-remote/journal-remote-main.c
+@@ -212,7 +212,12 @@ static int process_http_upload(
+                         break;
+                 else if (r < 0) {
+                         log_warning("Failed to process data for connection %p", connection);
+-                        if (r == -E2BIG)
++                        if (r == -ENOBUFS)
++                                return mhd_respondf(connection,
++                                                    r, MHD_HTTP_PAYLOAD_TOO_LARGE,
++                                                    "Entry is above the maximum of %u, aborting connection %p.",
++                                                    DATA_SIZE_MAX, connection);
++                        else if (r == -E2BIG)
+                                 return mhd_respondf(connection,
+                                                     r, MHD_HTTP_PAYLOAD_TOO_LARGE,
+                                                     "Entry is too large, maximum is " STRINGIFY(DATA_SIZE_MAX) " bytes.");
+diff --git a/src/journal-remote/journal-remote.c b/src/journal-remote/journal-remote.c
+index beb75a1cb4..67e3a70c06 100644
+--- a/src/journal-remote/journal-remote.c
++++ b/src/journal-remote/journal-remote.c
+@@ -408,7 +408,10 @@ int journal_remote_handle_raw_source(
+                 log_debug("%zu active sources remaining", s->active);
+                 return 0;
+         } else if (r == -E2BIG) {
+-                log_notice_errno(E2BIG, "Entry too big, skipped");
++                log_notice("Entry with too many fields, skipped");
++                return 1;
++        } else if (r == -ENOBUFS) {
++                log_notice("Entry too big, skipped");
+                 return 1;
+         } else if (r == -EAGAIN) {
+                 return 0;
+diff --git a/src/journal/journald-native.c b/src/journal/journald-native.c
+index 5ff22a10af..951d092053 100644
+--- a/src/journal/journald-native.c
++++ b/src/journal/journald-native.c
+@@ -140,6 +140,11 @@ static int server_process_entry(
+                 }
+ 
+                 /* A property follows */
++                if (n > ENTRY_FIELD_COUNT_MAX) {
++                        log_debug("Received an entry that has more than " STRINGIFY(ENTRY_FIELD_COUNT_MAX) " fields, ignoring entry.");
++                        r = 1;
++                        goto finish;
++                }
+ 
+                 /* n existing properties, 1 new, +1 for _TRANSPORT */
+                 if (!GREEDY_REALLOC(iovec, m,
+-- 
+2.11.0
+
diff --git a/meta/recipes-core/systemd/systemd_239.bb b/meta/recipes-core/systemd/systemd_239.bb
index 94729a6a39..e00d5dd914 100644
--- a/meta/recipes-core/systemd/systemd_239.bb
+++ b/meta/recipes-core/systemd/systemd_239.bb
@@ -39,6 +39,7 @@ SRC_URI += "file://touchscreen.rules \
            file://0002-core-Fix-use-after-free-case-in-load_from_path.patch \
            file://0001-meson-rename-Ddebug-to-Ddebug-extra.patch \
            file://0024-journald-do-not-store-the-iovec-entry-for-process-co.patch \
+           file://0025-journald-set-a-limit-on-the-number-of-fields.patch \
            "
 
 # patches made for musl are only applied on TCLIBC is musl
-- 
2.11.0



^ permalink raw reply related	[flat|nested] 7+ messages in thread

* [PATCH v2 3/3] systemd: Security fix CVE-2018-16866
  2019-01-28 11:17 [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Marcus Cooper
  2019-01-28 11:17 ` [PATCH v2 1/3] systemd: Security fix CVE-2018-16864 Marcus Cooper
  2019-01-28 11:17 ` [PATCH v2 2/3] systemd: Security fix CVE-2018-16865 Marcus Cooper
@ 2019-01-28 11:17 ` Marcus Cooper
  2019-01-28 11:20 ` [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Alexander Kanavin
  3 siblings, 0 replies; 7+ messages in thread
From: Marcus Cooper @ 2019-01-28 11:17 UTC (permalink / raw)
  To: openembedded-core

Affects < v240

Signed-off-by: Marcus Cooper <marcusc@axis.com>
---
 ...nal-fix-out-of-bounds-read-CVE-2018-16866.patch | 49 ++++++++++++++++++++++
 meta/recipes-core/systemd/systemd_239.bb           |  1 +
 2 files changed, 50 insertions(+)
 create mode 100644 meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch

diff --git a/meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch b/meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch
new file mode 100644
index 0000000000..3925a4abbb
--- /dev/null
+++ b/meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch
@@ -0,0 +1,49 @@
+From ebd06c37d4311db9851f4d3fdd023de3dd590de0 Mon Sep 17 00:00:00 2001
+From: Filipe Brandenburger <filbranden@google.com>
+Date: Thu, 10 Jan 2019 14:53:33 -0800
+Subject: [PATCH] journal: fix out-of-bounds read CVE-2018-16866
+
+The original code didn't account for the fact that strchr() would match on the
+'\0' character, making it read past the end of the buffer if no non-whitespace
+character was present.
+
+This bug was introduced in commit ec5ff4445cca6a which was first released in
+systemd v221 and later fixed in commit 8595102d3ddde6 which was released in
+v240, so versions in the range [v221, v240) are affected.
+
+Patch backported from systemd-stable at f005e73d3723d62a39be661931fcb6347119b52b
+also includes a change from systemd master which removes a heap buffer overflow
+a6aadf4ae0bae185dc4c414d492a4a781c80ffe5.
+
+CVE: CVE-2018-16866
+Upstream-Status: Backport
+Signed-off-by: Marcus Cooper <marcusc@axis.com>
+---
+ src/journal/journald-syslog.c | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/src/journal/journald-syslog.c b/src/journal/journald-syslog.c
+index 9dea116722..809b318c06 100644
+--- a/src/journal/journald-syslog.c
++++ b/src/journal/journald-syslog.c
+@@ -194,7 +194,7 @@ size_t syslog_parse_identifier(const char **buf, char **identifier, char **pid)
+         e = l;
+         l--;
+ 
+-        if (p[l-1] == ']') {
++        if (l > 0 && p[l-1] == ']') {
+                 size_t k = l-1;
+ 
+                 for (;;) {
+@@ -219,7 +219,7 @@ size_t syslog_parse_identifier(const char **buf, char **identifier, char **pid)
+         if (t)
+                 *identifier = t;
+ 
+-        if (strchr(WHITESPACE, p[e]))
++        if (p[e] != '\0' && strchr(WHITESPACE, p[e]))
+                 e++;
+         *buf = p + e;
+         return e;
+-- 
+2.11.0
+
diff --git a/meta/recipes-core/systemd/systemd_239.bb b/meta/recipes-core/systemd/systemd_239.bb
index e00d5dd914..f843f588bd 100644
--- a/meta/recipes-core/systemd/systemd_239.bb
+++ b/meta/recipes-core/systemd/systemd_239.bb
@@ -40,6 +40,7 @@ SRC_URI += "file://touchscreen.rules \
            file://0001-meson-rename-Ddebug-to-Ddebug-extra.patch \
            file://0024-journald-do-not-store-the-iovec-entry-for-process-co.patch \
            file://0025-journald-set-a-limit-on-the-number-of-fields.patch \
+           file://0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch \
            "
 
 # patches made for musl are only applied on TCLIBC is musl
-- 
2.11.0



^ permalink raw reply related	[flat|nested] 7+ messages in thread

* Re: [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866
  2019-01-28 11:17 [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Marcus Cooper
                   ` (2 preceding siblings ...)
  2019-01-28 11:17 ` [PATCH v2 3/3] systemd: Security fix CVE-2018-16866 Marcus Cooper
@ 2019-01-28 11:20 ` Alexander Kanavin
  2019-01-28 16:58   ` Khem Raj
  2019-01-28 17:37   ` akuster808
  3 siblings, 2 replies; 7+ messages in thread
From: Alexander Kanavin @ 2019-01-28 11:20 UTC (permalink / raw)
  To: Marcus Cooper; +Cc: OE-core

Maybe it's better to update systemd to latest upstream release?

Alex

On Mon, 28 Jan 2019 at 12:17, Marcus Cooper <marcus.cooper@axis.com> wrote:
>
> Changed in v2:
>  - Added CVE tag, Upstream-Status tag and Sign-off-by tags.
>  - removed the verification of the entry length in the header
>  - squashed CVE-2018-16865 patches into one
>  - CVE-2018-16866 patch now taken from systemd-stable and includes
>    an additional heap buffer overflow fix.
>
> Marcus Cooper (3):
>   systemd: Security fix CVE-2018-16864
>   systemd: Security fix CVE-2018-16865
>   systemd: Security fix CVE-2018-16866
>
>  ...-not-store-the-iovec-entry-for-process-co.patch | 208 +++++++++++++++++++++
>  ...rnald-set-a-limit-on-the-number-of-fields.patch | 139 ++++++++++++++
>  ...nal-fix-out-of-bounds-read-CVE-2018-16866.patch |  49 +++++
>  meta/recipes-core/systemd/systemd_239.bb           |   3 +
>  4 files changed, 399 insertions(+)
>  create mode 100644 meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
>  create mode 100644 meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch
>  create mode 100644 meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch
>
> --
> 2.11.0
>
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866
  2019-01-28 11:20 ` [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Alexander Kanavin
@ 2019-01-28 16:58   ` Khem Raj
  2019-01-28 17:37   ` akuster808
  1 sibling, 0 replies; 7+ messages in thread
From: Khem Raj @ 2019-01-28 16:58 UTC (permalink / raw)
  To: Alexander Kanavin; +Cc: OE-core

On Mon, Jan 28, 2019 at 3:20 AM Alexander Kanavin
<alex.kanavin@gmail.com> wrote:
>
> Maybe it's better to update systemd to latest upstream release?
>

Indeed but I would suggest to go to 241 which is in RC stage within
week of 240 release

> Alex
>
> On Mon, 28 Jan 2019 at 12:17, Marcus Cooper <marcus.cooper@axis.com> wrote:
> >
> > Changed in v2:
> >  - Added CVE tag, Upstream-Status tag and Sign-off-by tags.
> >  - removed the verification of the entry length in the header
> >  - squashed CVE-2018-16865 patches into one
> >  - CVE-2018-16866 patch now taken from systemd-stable and includes
> >    an additional heap buffer overflow fix.
> >
> > Marcus Cooper (3):
> >   systemd: Security fix CVE-2018-16864
> >   systemd: Security fix CVE-2018-16865
> >   systemd: Security fix CVE-2018-16866
> >
> >  ...-not-store-the-iovec-entry-for-process-co.patch | 208 +++++++++++++++++++++
> >  ...rnald-set-a-limit-on-the-number-of-fields.patch | 139 ++++++++++++++
> >  ...nal-fix-out-of-bounds-read-CVE-2018-16866.patch |  49 +++++
> >  meta/recipes-core/systemd/systemd_239.bb           |   3 +
> >  4 files changed, 399 insertions(+)
> >  create mode 100644 meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
> >  create mode 100644 meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch
> >  create mode 100644 meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch
> >
> > --
> > 2.11.0
> >
> > --
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core@lists.openembedded.org
> > http://lists.openembedded.org/mailman/listinfo/openembedded-core
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core


^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866
  2019-01-28 11:20 ` [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Alexander Kanavin
  2019-01-28 16:58   ` Khem Raj
@ 2019-01-28 17:37   ` akuster808
  1 sibling, 0 replies; 7+ messages in thread
From: akuster808 @ 2019-01-28 17:37 UTC (permalink / raw)
  To: Alexander Kanavin, Marcus Cooper; +Cc: OE-core



On 1/28/19 3:20 AM, Alexander Kanavin wrote:
> Maybe it's better to update systemd to latest upstream release?
Thud benefits from this : )

>
> Alex
>
> On Mon, 28 Jan 2019 at 12:17, Marcus Cooper <marcus.cooper@axis.com> wrote:
>> Changed in v2:
>>  - Added CVE tag, Upstream-Status tag and Sign-off-by tags.
>>  - removed the verification of the entry length in the header
>>  - squashed CVE-2018-16865 patches into one
>>  - CVE-2018-16866 patch now taken from systemd-stable and includes
>>    an additional heap buffer overflow fix.
>>
>> Marcus Cooper (3):
>>   systemd: Security fix CVE-2018-16864
>>   systemd: Security fix CVE-2018-16865
>>   systemd: Security fix CVE-2018-16866
>>
>>  ...-not-store-the-iovec-entry-for-process-co.patch | 208 +++++++++++++++++++++
>>  ...rnald-set-a-limit-on-the-number-of-fields.patch | 139 ++++++++++++++
>>  ...nal-fix-out-of-bounds-read-CVE-2018-16866.patch |  49 +++++
>>  meta/recipes-core/systemd/systemd_239.bb           |   3 +
>>  4 files changed, 399 insertions(+)
>>  create mode 100644 meta/recipes-core/systemd/systemd/0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
>>  create mode 100644 meta/recipes-core/systemd/systemd/0025-journald-set-a-limit-on-the-number-of-fields.patch
>>  create mode 100644 meta/recipes-core/systemd/systemd/0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch
>>
>> --
>> 2.11.0
>>
>> --
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.openembedded.org/mailman/listinfo/openembedded-core



^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2019-01-28 17:37 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-01-28 11:17 [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Marcus Cooper
2019-01-28 11:17 ` [PATCH v2 1/3] systemd: Security fix CVE-2018-16864 Marcus Cooper
2019-01-28 11:17 ` [PATCH v2 2/3] systemd: Security fix CVE-2018-16865 Marcus Cooper
2019-01-28 11:17 ` [PATCH v2 3/3] systemd: Security fix CVE-2018-16866 Marcus Cooper
2019-01-28 11:20 ` [PATCH v2 0/3] systemd: Fixes Security fix CVE-2018-16864 - CVE-2018-16866 Alexander Kanavin
2019-01-28 16:58   ` Khem Raj
2019-01-28 17:37   ` akuster808

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox