admin/systemd/haproxy.service.in | 2 1 + 1 - 0 !
1 file changed, 1 insertion(+), 1 deletion(-)

 start after rsyslog.service

As HAProxy is running chrooted by default, we rely on an additional syslog
socket created by rsyslog inside the chroot for logging. As this socket cannot
trigger syslog activation, we explicitly order HAProxy after rsyslog.service.
Note that we are not using syslog.service here, since the additional socket is
rsyslog-specific.

admin/systemd/haproxy.service.in | 2 2 + 0 - 0 !
1 file changed, 2 insertions(+)

 add documentation field to the systemd unit

admin/systemd/haproxy.service.in | 1 1 + 0 - 0 !
1 file changed, 1 insertion(+)

 haproxy.service: make systemd bind /dev/log inside chroot

This enables logging to work without rsyslog being present.

Makefile | 2 1 + 1 - 0 !
1 file changed, 1 insertion(+), 1 deletion(-)

---

include/haproxy/http.h | 18 18 + 0 - 0 !
src/h2.c | 23 3 + 20 - 0 !
2 files changed, 21 insertions(+), 20 deletions(-)

 reorg: http: move has_forbidden_char() from h2.c to http.h

src/h3.c | 38 38 + 0 - 0 !
1 file changed, 38 insertions(+)

 bug/major: h3: reject header values containing invalid chars

reg-tests/http-messaging/h1_to_h1.vtc | 26 26 + 0 - 0 !
reg-tests/http-messaging/h2_to_h1.vtc | 60 60 + 0 - 0 !
src/h1.c | 20 17 + 3 - 0 !
src/http.c | 20 17 + 3 - 0 !
4 files changed, 120 insertions(+), 6 deletions(-)

 bug/major: http: reject any empty content-length header value

include/import/ist.h | 47 47 + 0 - 0 !
1 file changed, 47 insertions(+)

 minor: ist: add new function ist_find_range() to find a character
 range

include/haproxy/http.h | 19 19 + 0 - 0 !
1 file changed, 19 insertions(+)

 minor: http: add new function http_path_has_forbidden_char()

include/haproxy/h2.h | 2 1 + 1 - 0 !
src/h2.c | 6 5 + 1 - 0 !
src/mux_h2.c | 3 2 + 1 - 0 !
3 files changed, 8 insertions(+), 3 deletions(-)

 minor: h2: pass accept-invalid-http-request down the request parser

reg-tests/http-rules/normalize_uri.vtc | 2 2 + 0 - 0 !
1 file changed, 2 insertions(+)

 regtests: http-rules: add accept-invalid-http-request for
 normalize-uri tests

src/h1.c | 15 11 + 4 - 0 !
1 file changed, 11 insertions(+), 4 deletions(-)

 bug/minor: h1: do not accept '#' as part of the uri component

src/h2.c | 15 11 + 4 - 0 !
1 file changed, 11 insertions(+), 4 deletions(-)

 bug/minor: h2: reject more chars from the :path pseudo header

src/h3.c | 14 14 + 0 - 0 !
1 file changed, 14 insertions(+)

 bug/minor: h3: reject more chars from the :path pseudo header

reg-tests/http-rules/normalize_uri.vtc | 11 11 + 0 - 0 !
1 file changed, 11 insertions(+)

 regtests: http-rules: verify that we block '#' by default for
 normalize-uri

doc/configuration.txt | 20 17 + 3 - 0 !
1 file changed, 17 insertions(+), 3 deletions(-)

 doc: clarify the handling of url fragments in requests

src/sample.c | 2 1 + 1 - 0 !
1 file changed, 1 insertion(+), 1 deletion(-)

 bug/medium: sample: fix risk of overflow when replacing multiple
 regex back-refs

Aleandro Prudenzano of Doyensec and Edoardo Geraci of Codean Labs
reported a bug in sample_conv_regsub(), which can cause replacements
of multiple back-references to overflow the temporary trash buffer.

The problem happens when doing "regsub(match,replacement,g)": we're
replacing every occurrence of "match" with "replacement" in the input
sample, which requires a length check. For this, a max is applied, so
that a replacement may not use more than the remaining length in the
buffer. However, the length check is made on the replaced pattern and
not on the temporary buffer used to carry the new string. This results
in the remaining size to be usable for each input match, which can go
beyond the temporary buffer size if more than one occurrence has to be
replaced with something that's larger than the remaining room.

The fix proposed by Aleandro and Edoardo is the correct one (check on
"trash" not "output"), and is the one implemented in this patch.

While it is very unlikely that a config will replace multiple short
patterns each with a larger one in a request, this possibility cannot
be entirely ruled out (e.g. mask a known, short IP address using
"XXX.XXX.XXX.XXX").  However when this happens, the replacement pattern
will be static, and not be user-controlled, which is why this patch is
marked as medium.

The bug was introduced in 2.2 with commit 07e1e3c93e ("MINOR: sample:
regsub now supports backreferences"), so it must be backported to all
versions.

Special thanks go to Aleandro and Edoardo for reporting this bug with
a simple reproducer and a fix.

(cherry picked from commit 3e3b9eebf871510aee36c3a3336faac2f38c9559)
Signed-off-by: Aurelien DARRAGON <adarragon@haproxy.com>
(cherry picked from commit db87c8d9fe621539531f6f915ba9e1755a2a26cb)
Signed-off-by: Aurelien DARRAGON <adarragon@haproxy.com>

src/mjson.c | 14 12 + 2 - 0 !
1 file changed, 12 insertions(+), 2 deletions(-)

 bug/critical: mjson: fix possible dos when parsing numbers

Mjson comes with its own strtod() implementation for portability
reasons and probably also because many generic strtod() versions as
provided by operating systems do not focus on resource preservation
and may call malloc(), which is not welcome in a parser.

The strtod() implementation used here apparently originally comes from
https://gist.github.com/mattn/1890186 and seems to have purposely
omitted a few parts that were considered as not needed in this context
(e.g. skipping white spaces, or setting errno). But when subject to the
relevant test cases of the designated file above, the current function
provides the same results.

The aforementioned implementation uses pow() to calculate exponents,
but mjson authors visibly preferred not to introduce a libm dependency
and replaced it with an iterative loop in O(exp) time. The problem is
that the exponent is not bounded and that this loop can take a huge
amount of time. There's even an issue already opened on mjson about
this: https://github.com/cesanta/mjson/issues/59. In the case of
haproxy, fortunately, the watchdog will quickly stop a runaway process
but this remains a possible denial of service.

A first approach would consist in reintroducing pow() like in the
original implementation, but if haproxy is built without Lua nor
51Degrees, -lm is not used so this will not work everywhere.

Anyway here we're dealing with integer exponents, so an easy alternate
approach consists in simply using shifts and squares, to compute the
exponent in O(log(exp)) time. Not only it doesn't introduce any new
dependency, but it turns out to be even faster than the generic pow()
(85k req/s per core vs 83.5k on the same machine).

This must be backported as far as 2.4, where mjson was introduced.

Many thanks to Oula Kivalo for reporting this issue.