rpms/libqb
5
0
Fork 0
Code Issues Pull Requests Releases Activity
1799fc0be2
libqb/04-Med-add-extra-run-time-client-libqb-checks-that-logg.patch
Jan Pokorný 1799fc0be2
1.0.2-14 - Evolution of the previous (rhbz#1478089) 
- Make -devel package dependency on the main package arch-qualified

Signed-off-by: Jan Pokorný <jpokorny@redhat.com>
2017-11-15 18:32:21 +01:00

632 lines
29 KiB
Diff
Raw Blame History

From 260768d94017023771af9e7f1bddcff91883617a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jan=20Pokorn=C3=BD?= <jpokorny@redhat.com>
Date: Fri, 6 Oct 2017 17:17:26 +0200
Subject: [PATCH 4/6] Med: add extra run-time (client, libqb) checks that
logging will work
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Now that the previous commit provides a foundation for what exactly can
go wrong with ld.bfd = 2.29+ linker, let's start reconciling that with
some reasonable assurance that logging is not silently severed, because
realizing the logs are missing is otherwise bound to happen when the
logs are suddently pretty crucial analytical resource :-)
We'll proceed in two steps as detailed.
* * *
As a first step, the table below concludes how the test matrix overview
introduced with a message for the preceding commit (also introducing
log_test_mock.sh runner which got reused here) looks as of this
refreshed sanity check, once QB_LOG_INIT_DATA macro at hand gets applied
(meaning "for non-libqb logging participants" so as not complicate the
matrix further). That macro is nothing triggered directly, it will just
plant a constructor-like function (to be invoked automatically early in
the execution) that will run through the checks (one original and couple
of new ones as of this changeset).
Note that for libqb users, this implies a new link dependency on libdl,
because they may opt-in for refreshed QB_LOG_INIT_DATA sanity check that
calls out to dlopen/dlsym/dladdr directly in case of "attribute section"
being available for the particular platform, and hence immediately needs
those symbols resolved in link time. Hence, add this conditional link
dependency to libqb.pc pkg-config file under Libs variable -- we
actually restore the occurrence of "-ldl" there as it used to be present
until commit 56754d0. While doing so, also move immediate link
dependencies of libqb (if any, currently not but that may be
a regression arising from the cleanup related to the mentioned commit)
represented with the LIBS autoconf variable under Libs.private variable
in libqb.pc, where it belongs per pkg-config documentation.
The promised table follows, but first as a recap, "X(Y)" denotes
"X linked with linker Y":
X(a) .. ld.bfd < 2.29
X(b) .. ld.bfd = 2.29 (and only 2.29)
goes like this (values in <angle brackets> denote non-trivial change
[not mere rewording] introduced as of this commit, in comparison to
the table stated in the preceding commit):
+=========+=========+=========+=========+=========+=========+=========+
#client(x)# libqb(a) usage # libqb(b) usage #
# vvv #---------+---------+---------+---------+---------+---------+
# V # direct | libX(a) : libX(b) # direct | libX(a) : libX(b) #
+=========+=========+=========+=========+=========+=========+=========+
# x = a # OK | OK : BAD[*2] #<BAD[*E]>|<BAD[*F]>:<BAD[*G]>#
# x = b # BAD[*A] | BAD[*B] : BAD[*C] #<BAD[*E]>|<BAD[*F]>:<BAD[*G]>#
+=========+=========+=========+=========+=========+=========+=========+
Woefully, nothing changes if we swap binutils 2.29 for 2.29.1, i.e.,
X(b) .. ld.bfd = 2.29.1, compared to previous state, i.e., the second
table from the previous commit is still applicable for that situation.
The added sanity checks are useful nonetheless, consider for example,
that attribute-section-less libqb is what gets run-time linked to an
attribute-section-full target. The most precise check we could use
-- a custom logger function applied in a self-test scheme -- is not
available at the point the macro-defined function gets invoked, simply
because qb_log_init hasn't been invoked by the time that constructor
gets triggered. However, what we can do is to add a non-trapping
libqb-residing reverse-testing of the client space that (and once it)
voluntarily initiates qb_log_init (delivering abruption all of a sudden
at some unanticipated, as opposed to a well-timed like with
constructors, execution point, seems pretty bad idea + libqb as
a library is a mere helper, not an undertaker :) -- this check then
only announces, via syslog (the only pre-enabled logging target),
the target's logging may be severed.
* * *
Hence, as a promised second step, after incorporating the syslog
change (and extending log_test_mock.sh so as to capture syslog
stream within the container), not much changes with the table above,
i.e., X(b) .. ld.bfd = 2.29:
[*A] in addition, unless QB_LOG_INIT_DATA used on client side,
syslog carries this warning:
"(LOG@QB) target chain supplied section not observed by libqb,
target's and/or libqb's linkage at fault and logging of the target
will not work reliably (unless qb_log_init function used unexpectedly
in no-logging context, or target chain built purposefully without
callsite section)"
logged by libqb proper
[*C] in addition, unless QB_LOG_INIT_DATA used on interlib side,
syslog carries this warning:
"(LOG@QB) target chain supplied section not observed by libqb,
target's and/or libqb's linkage at fault and logging of the target
will not work reliably (unless qb_log_init function used unexpectedly
in no-logging context, or target chain built purposefully without
callsite section)"
logged by libqb proper
[*E] in addition, unless QB_LOG_INIT_DATA used on client side,
syslog carries this warning:
"(LOG@QB) target chain supplied section empty, target's and/or
libqb's linkage at fault and logging of the target will not
work reliably (unless qb_log_init function used unexpectedly
in no-logging context)"
logged by libqb proper
[*F] in addition, unless QB_LOG_INIT_DATA used on interlib side,
syslog carries this warning:
"(LOG@QB) target chain supplied section empty, target's and/or
libqb's linkage at fault and logging of the target will not
work reliably (unless qb_log_init function used unexpectedly
in no-logging context)"
logged by libqb proper
[*G] in addition, unless QB_LOG_INIT_DATA used on interlib side,
syslog carries this warning:
"(LOG@QB) target chain supplied section empty, target's and/or
libqb's linkage at fault and logging of the target will not
work reliably (unless qb_log_init function used unexpectedly
in no-logging context)"
logged by libqb proper
but desirably changes with "X(b) .. ld.bfd = 2.29.1" one
(DEP ~ "depends"):
+=========+=========+=========+=========+=========+=========+=========+
#client(x)# libqb(a) usage # libqb(b) usage #
# vvv #---------+---------+---------+---------+---------+---------+
# V # direct | libX(a) : libX(b) # direct | libX(a) : libX(b) #
+=========+=========+=========+=========+=========+=========+=========+
# x = a # OK | OK : DEP[*J] #<BAD[*M]>|<BAD[*M]>:<BAD[*L]>#
# x = b #<DEP[*N]>| DEP[*I] :<DEP[*O]>#<BAD[*M]>|<BAD[*M]>:<BAD[*L]>#
+=========+=========+=========+=========+=========+=========+=========+
* * *
[*1] client logging not working
[*2] interlib logging not working
[*3] both client and interlib logging not working
[*A] boils down to [*1], unless QB_LOG_INIT_DATA used on client side,
which then fails on
"non-empty implicit callsite section, otherwise target's linkage at
fault and logging would not work reliably"
assertion
[*B] boils down to [*1], unless QB_LOG_INIT_DATA used on interlib side,
which then fails on
"non-empty implicit callsite section, otherwise target's linkage at
fault and logging would not work reliably"
assertion
[*C] boils down to [*3], unless QB_LOG_INIT_DATA used on interlib side,
which then fails on
"non-empty implicit callsite section, otherwise target's linkage at
fault and logging would not work reliably"
assertion
[*E] boils down to [*1], unless QB_LOG_INIT_DATA used on client side,
which then fails on
"implicit callsite section self-observable, otherwise target's and/or
libqb's linkage at fault and logging would not work reliably"
assertion
[*F] boils down to [*3], unless QB_LOG_INIT_DATA used on interlib side,
which then fails on
"libqb's callsite section non-empty, otherwise libqb's linkage at
fault and logging would not work reliably"
assertion
[*G] boils down to [*3], unless QB_LOG_INIT_DATA used on interlib side,
which then fails on
"implicit callsite section self-observable, otherwise target's and/or
libqb's linkage at fault and logging would not work reliably"
assertion
[*I] boils down to [*1], unless QB_LOG_INIT_DATA used on client side,
which makes it, likely through self-reference keepalive (see
below), work OK
[*J] boils down to [*2], unless QB_LOG_INIT_DATA used on interlib side,
which makes it, likely through self-reference keepalive (see
below), work OK
[*K] boils down to [*3]
in addition, syslog carries this warning:
"(LOG@QB) target chain supplied section not observed by libqb,
target's and/or libqb's linkage at fault and logging of the target
will not work reliably (unless qb_log_init function used unexpectedly
in no-logging context, or target chain built purposefully without
callsite section)"
logged by libqb proper
[*L] boils down to [*3], unless QB_LOG_INIT_DATA used on interlib side
(sufficient?), which makes it, likely through self-reference
keepalive (see below), boil down just to [*1];
in addition, syslog carries this warning:
"(LOG@QB) target chain supplied section not observed by libqb,
target's and/or libqb's linkage at fault and logging of the target
will not work reliably (unless qb_log_init function used unexpectedly
in no-logging context, or target chain built purposefully without
callsite section)"
logged by libqb proper
[*M] boils down to [*1];
in addition, syslog carries this warning:
"(LOG@QB) target chain supplied section not observed by libqb,
target's and/or libqb's linkage at fault and logging of the target
will not work reliably (unless qb_log_init function used unexpectedly
in no-logging context, or target chain built purposefully without
callsite section)"
logged by libqb proper
[*N] boils down to [*M], unless QB_LOG_INIT_DATA used on client side,
which makes it, likely through self-reference keepalive (see
below), work OK
[*O] boils down to [*K], unless QB_LOG_INIT_DATA used on both client
and interlib side, which makes it, likely through self-reference
keepalive (see below), work OK (it's expected that this a mere
composite of situations [*I] and [*J] with consequences as stated)
* * *
Note: the only problematic (i.e. not captured automatically by the
QB_LOG_INIT_DATA macro presumably utilized at every non-libqb logging
system participant in the form of a discrete compilation unit)
combination with 2.29 is the one intersecting at "BAD[*2]" pertaining
"everything but interlib compiled with ld.bfd < 2.29". It would, of
course, be solvable as well, but presumably not in an easy way, and
that use case should not be as frequent.
Takeway: whenever your target (library or client program) actively
utilizes logging (meaning it emits at least a single log message,
otherwise there's an imminent danger of possibly even run-terminating
false positive in the self-check mechanism!),
YOU ARE strongly ENCOURAGED TO USE QB_LOG_INIT_DATA macro function
at (exactly) one of the source code files (presumably the main one)
per respective target's compilation unit.
It will alleviate the hassles possibly caused by downgrading libqb
to the linker-vs-libqb incompatibly compiled one or in similar
circumstances arising merely from the linker behaviour change,
which the current build system/code shake is all about.
Signed-off-by: Jan Pokorný <jpokorny@redhat.com>
---
configure.ac | 1 +
include/qb/qblog.h | 111 ++++++++++++++++++++++--------
lib/libqb.pc.in | 3 +-
lib/log.c | 53 ++++++++++++++
tests/functional/Makefile.am | 3 +-
tests/functional/log.am | 2 +-
tests/functional/log_external/Makefile.am | 8 ++-
tests/functional/log_test_mock.sh | 13 +++-
tests/functional/syslog-stdout.py | 51 ++++++++++++++
9 files changed, 209 insertions(+), 36 deletions(-)
create mode 100755 tests/functional/syslog-stdout.py
diff --git a/configure.ac b/configure.ac
index 8c588cb..fdcd93c 100644
--- a/configure.ac
+++ b/configure.ac
@@ -655,6 +655,7 @@ if test "x${GCC}" = xyes; then
AC_DEFINE([QB_HAVE_ATTRIBUTE_SECTION], 1,
[Enabling code using __attribute__((section))])
PACKAGE_FEATURES="$PACKAGE_FEATURES attribute-section"
+ AC_SUBST([client_dlopen_LIBS],[$dlopen_LIBS])
fi
fi
diff --git a/include/qb/qblog.h b/include/qb/qblog.h
index 4954932..a9a9ef2 100644
--- a/include/qb/qblog.h
+++ b/include/qb/qblog.h
@@ -1,9 +1,10 @@
/*
- * Copyright (C) 2010 Red Hat, Inc.
+ * Copyright 2017 Red Hat, Inc.
*
* All rights reserved.
*
* Author: Angus Salkeld <asalkeld@redhat.com>
+ * Jan Pokorny <jpokorny@redhat.com>
*
* libqb is free software: you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published by
@@ -44,6 +45,7 @@ extern "C" {
#ifdef QB_HAVE_ATTRIBUTE_SECTION
#include <assert.h> /* possibly needed for QB_LOG_INIT_DATA */
+#include <dlfcn.h> /* dynamic linking: dlopen, dlsym, dladdr, ... */
#endif
/**
@@ -70,14 +72,14 @@ extern "C" {
*
* @note
* In practice, such a minimalistic approach hardly caters real use cases.
- * Following section discusses the customization. Moreover, it's quite
- * vital to instrument the target user of this logging subsystem with
- * @c QB_LOG_INIT_DATA() macro placed in the top file scope in an exactly
- * one source file (preferably the main one) to be mixed into the resulting
- * compilation unit. This is a self-defensive measure for when the
- * linker-assisted collection of callsite data silently fails, which
- * could otherwise go unnoticed, causing troubles down the road.
- *
+ * Following section discusses the customization. Moreover when employing
+ * the log module is bound to its active use (some log messages are assuredly
+ * emitted within the target compilation unit), it's quite vital to instrument
+ * the target side with @c QB_LOG_INIT_DATA() macro placed in the top file
+ * scope in exactly one source file (preferably the main one) to be mixed into
+ * the resulting compilation unit. This is a self-defensive measure for when
+ * the linker-assisted collection of callsite data silently fails, which could
+ * otherwise go unnoticed, causing troubles down the road.
*
* @par Configuring log targets.
* A log target can be syslog, stderr, the blackbox, stdout, or a text file.
@@ -283,35 +285,86 @@ typedef void (*qb_log_filter_fn)(struct qb_log_callsite * cs);
extern struct qb_log_callsite QB_ATTR_SECTION_START[];
extern struct qb_log_callsite QB_ATTR_SECTION_STOP[];
-/* optional on-demand self-check of 1/ toolchain sanity (prerequisite for
- the logging subsystem to work properly) and 2/ non-void active use of
- logging (satisfied with a justifying existence of a logging callsite as
- defined with a @c qb_logt invocation) at the target (but see below), which
- is supposedly assured by it's author(!) as of relying on this very macro
- [technically, the symbols that happen to be resolved under the respective
- identifiers do not necessarily originate in the same compilation unit as
- when it's not the end executable (or by induction, a library positioned
- earlier in the symbol lookup order) but a shared library, the former takes
- a precedence unless that site comes short of exercising the logging,
- making its callsite section empty and, in turn, without such boundary
- symbols, hence making the resolution continue further in the lookup order
- -- despite fuzzily targeted attestation, the check remains reasonable];
- only effective when link-time ("run-time amortizing") callsite collection
- is; as a side effect, it can ensure the boundary-denoting symbols for the
- target collection area are kept alive with some otherwise unkind linkers;
- may be extended in future for more in-depth self-validation */
+/* Related to the next macro that is -- unlike this one -- a public API */
+#ifndef _GNU_SOURCE
+#define QB_NONAPI_LOG_INIT_DATA_EXTRA_ \
+ _Pragma(QB_PP_STRINGIFY(GCC warning QB_PP_STRINGIFY( \
+ without "_GNU_SOURCE" defined (directly or not) \
+ QB_LOG_INIT_DATA cannot check sanity of libqb proper)))
+#else
+#define QB_NONAPI_LOG_INIT_DATA_EXTRA_ \
+ { Dl_info work_dli; \
+ /* libqb sanity (locating libqb by it's relatively unique \
+ -- and currently only such per-linkage global one -- \
+ non-functional symbol, due to possible confusion otherwise) */ \
+ if (dladdr(dlsym(RTLD_DEFAULT, "facilitynames"), &work_dli) \
+ && (work_handle = dlopen(work_dli.dli_fname, \
+ RTLD_LOCAL|RTLD_LAZY)) != NULL) { \
+ work_s1 = (struct qb_log_callsite *) \
+ dlsym(work_handle, QB_ATTR_SECTION_START_STR); \
+ work_s2 = (struct qb_log_callsite *) \
+ dlsym(work_handle, QB_ATTR_SECTION_STOP_STR); \
+ assert("libqb's callsite section observable, otherwise libqb's \
+linkage at fault and logging would not work reliably" \
+ && work_s1 != NULL && work_s2 != NULL); \
+ assert("libqb's callsite section non-empty, otherwise libqb's \
+linkage at fault and logging would not work reliably" \
+ && work_s1 != work_s2); \
+ dlclose(work_handle); } }
+#endif /* _GNU_SOURCE */
+
+/**
+ * Optional on-demand self-check of 1/ toolchain sanity (prerequisite for
+ * the logging subsystem to work properly) and 2/ non-void active use of
+ * logging (satisfied with a justifying existence of a logging callsite as
+ * defined with a @c qb_logt invocation) at the target (but see below), which
+ * is supposedly assured by it's author(!) as of relying on this very macro
+ * [technically, the symbols that happen to be resolved under the respective
+ * identifiers do not necessarily originate in the same compilation unit as
+ * when it's not the end executable (or by induction, a library positioned
+ * earlier in the symbol lookup order) but a shared library, the former takes
+ * a precedence unless that site comes short of exercising the logging,
+ * making its callsite section empty and, in turn, without such boundary
+ * symbols, hence making the resolution continue further in the lookup order
+ * -- despite fuzzily targeted attestation, the check remains reasonable];
+ * only effective when link-time ("run-time amortizing") callsite collection
+ * is; as a side effect, it can ensure the boundary-denoting symbols for the
+ * target collection area are kept alive with some otherwise unkind linkers.
+ *
+ * Applying this macro in the target program/library is strongly recommended
+ * whenever the logging as framed by this header file is in use.
+ * Moreover, it's important to state that using this check while not ensuring
+ * @c _GNU_SOURCE macro definition is present at compile-time means only half
+ * of the available sanity checking will be performed, possibly resulting
+ * in libqb's own internally logged messages being lost without warning.
+ */
#define QB_LOG_INIT_DATA(name) \
void name(void); \
void name(void) { \
+ void *work_handle; struct qb_log_callsite *work_s1, *work_s2; \
/* our own (target's) sanity, or possibly that of higher priority \
symbol resolution site (unless target equals end executable) \
or even the lower one if no such predecessor defines these */ \
- assert("non-empty implicit callsite section" \
- && QB_ATTR_SECTION_START != QB_ATTR_SECTION_STOP); } \
+ if ((work_handle = dlopen(NULL, RTLD_LOCAL|RTLD_LAZY)) != NULL) { \
+ work_s1 = (struct qb_log_callsite *) \
+ dlsym(work_handle, QB_ATTR_SECTION_START_STR); \
+ work_s2 = (struct qb_log_callsite *) \
+ dlsym(work_handle, QB_ATTR_SECTION_STOP_STR); \
+ assert("implicit callsite section observable, otherwise \
+target's and/or libqb's linkage at fault and logging would not work \
+reliably" \
+ && work_s1 != NULL && work_s2 != NULL); \
+ dlclose(work_handle); /* perhaps overly eager thing to do */ } \
+ /* better targeted attestations when available */ \
+ QB_NONAPI_LOG_INIT_DATA_EXTRA_; \
+ /* finally, original, straightforward check */ \
+ assert("non-empty implicit callsite section, otherwise target's \
+linkage at fault and logging would not work reliably" \
+ && QB_ATTR_SECTION_START != QB_ATTR_SECTION_STOP); } \
void __attribute__ ((constructor)) name(void);
#else
#define QB_LOG_INIT_DATA(name)
-#endif
+#endif /* QB_HAVE_ATTRIBUTE_SECTION */
/**
* Internal function: use qb_log() or qb_logt()
diff --git a/lib/libqb.pc.in b/lib/libqb.pc.in
index 8a8d0ba..37d27b7 100644
--- a/lib/libqb.pc.in
+++ b/lib/libqb.pc.in
@@ -7,5 +7,6 @@ Name: libqb
Version: @PACKAGE_VERSION@
Description: libqb
Requires:
-Libs: -L${libdir} -lqb @LIBS@
+Libs: -L${libdir} -lqb @client_dlopen_LIBS@
+Libs.private: @LIBS@
Cflags: -I${includedir}
diff --git a/lib/log.c b/lib/log.c
index bfd218f..eb35d6f 100644
--- a/lib/log.c
+++ b/lib/log.c
@@ -853,6 +853,18 @@ qb_log_init(const char *name, int32_t facility, uint8_t priority)
{
int32_t l;
enum qb_log_target_slot i;
+#ifdef QB_HAVE_ATTRIBUTE_SECTION
+ void *work_handle; struct qb_log_callsite *work_s1, *work_s2;
+ Dl_info work_dli;
+#endif /* QB_HAVE_ATTRIBUTE_SECTION */
+ /* cannot reuse single qb_log invocation in various contexts
+ through the variables (when section attribute in use),
+ hence this indirection */
+ enum {
+ preinit_err_none,
+ preinit_err_target_sec,
+ preinit_err_target_empty,
+ } preinit_err = preinit_err_none;
l = pthread_rwlock_init(&_listlock, NULL);
assert(l == 0);
@@ -871,6 +883,26 @@ qb_log_init(const char *name, int32_t facility, uint8_t priority)
qb_log_dcs_init();
#ifdef QB_HAVE_ATTRIBUTE_SECTION
+ /* sanity check that target chain supplied QB_ATTR_SECTION_ST{ART,OP}
+ symbols and hence the local references to them are not referencing
+ the proper libqb's ones (locating libqb-self by it's relatively
+ unique -- and currently only such per-linkage global one --
+ non-functional symbol, due to possible confusion otherwise) */
+ if (dladdr(dlsym(RTLD_DEFAULT, "facilitynames"), &work_dli)
+ && (work_handle = dlopen(work_dli.dli_fname,
+ RTLD_LOCAL|RTLD_LAZY)) != NULL) {
+ work_s1 = (struct qb_log_callsite *)
+ dlsym(work_handle, QB_ATTR_SECTION_START_STR);
+ work_s2 = (struct qb_log_callsite *)
+ dlsym(work_handle, QB_ATTR_SECTION_STOP_STR);
+ if (work_s1 == QB_ATTR_SECTION_START
+ || work_s2 == QB_ATTR_SECTION_STOP) {
+ preinit_err = preinit_err_target_sec;
+ } else if (work_s1 == work_s2) {
+ preinit_err = preinit_err_target_empty;
+ }
+ dlclose(work_handle); /* perhaps overly eager thing to do */
+ }
qb_log_callsites_register(QB_ATTR_SECTION_START, QB_ATTR_SECTION_STOP);
dl_iterate_phdr(_log_so_walk_callback, NULL);
_log_so_walk_dlnames();
@@ -884,6 +916,27 @@ qb_log_init(const char *name, int32_t facility, uint8_t priority)
_log_target_state_set(&conf[QB_LOG_SYSLOG], QB_LOG_STATE_ENABLED);
(void)qb_log_filter_ctl(QB_LOG_SYSLOG, QB_LOG_FILTER_ADD,
QB_LOG_FILTER_FILE, "*", priority);
+
+ if (preinit_err == preinit_err_target_sec)
+ qb_util_log(LOG_WARNING, "(LOG@QB) target chain supplied"
+ " section not observed by libqb,"
+ " target's and/or libqb's linkage"
+ " at fault and logging of the target"
+ " will not work reliably (unless"
+ " qb_log_init function used"
+ " unexpectedly in no-logging"
+ " context, or target chain built"
+ " purposefully without callsite"
+ " sections)");
+ else if (preinit_err == preinit_err_target_empty) {
+ qb_util_log(LOG_WARNING, "(LOG@QB) target chain supplied"
+ " section empty, target's and/or"
+ " libqb's linkage at fault and"
+ " logging of the target will not"
+ " work reliably (unless qb_log_init"
+ " function used unexpectedly in"
+ " no-logging context)");
+ }
}
void
diff --git a/tests/functional/Makefile.am b/tests/functional/Makefile.am
index 0d34ae8..522fb61 100644
--- a/tests/functional/Makefile.am
+++ b/tests/functional/Makefile.am
@@ -19,5 +19,6 @@
MAINTAINERCLEANFILES = Makefile.in
EXTRA_DIST = GNUmakefile log_test_client.err log_test_interlib_client.err \
- log_test_client.sh log_test_interlib_client.sh log_test_mock.sh
+ log_test_client.sh log_test_interlib_client.sh \
+ log_test_mock.sh syslog-stdout.py
SUBDIRS = log_external log_internal
diff --git a/tests/functional/log.am b/tests/functional/log.am
index e8e4740..80bccb0 100644
--- a/tests/functional/log.am
+++ b/tests/functional/log.am
@@ -20,7 +20,7 @@
MAINTAINERCLEANFILES = Makefile.in
CLEANFILES = log_test_client.err.real log_test_interlib_client.err.real
-AM_CPPFLAGS = -I$(top_builddir)/include -I$(top_srcdir)/include
+AM_CPPFLAGS = -D_GNU_SOURCE -I$(top_builddir)/include -I$(top_srcdir)/include
noinst_PROGRAMS = log_client log_interlib_client
# cannot use {check,noinst}_LTLIBRARIES because it leads to solely static lib
diff --git a/tests/functional/log_external/Makefile.am b/tests/functional/log_external/Makefile.am
index 36aa0af..ca1c8a5 100644
--- a/tests/functional/log_external/Makefile.am
+++ b/tests/functional/log_external/Makefile.am
@@ -19,5 +19,9 @@
include ../log.am
-log_client_LDFLAGS = -lqb
-liblog_inter_la_LIBADD = -lqb
+# while linking with system-wide version of libqb, we are still pursuing
+# local in-tree header file, hence we need to link with dynamic linking
+# library (which is a prerequisite for using QB_LOG_INIT_DATA defined
+# in qblog.h) explicitly
+log_client_LDFLAGS = -lqb @client_dlopen_LIBS@
+liblog_inter_la_LIBADD = -lqb @client_dlopen_LIBS@
diff --git a/tests/functional/log_test_mock.sh b/tests/functional/log_test_mock.sh
index 145b9f4..cdfce1f 100755
--- a/tests/functional/log_test_mock.sh
+++ b/tests/functional/log_test_mock.sh
@@ -166,10 +166,19 @@ do_compile_and_test_client () {
-exec rm -- {} \;"
;;
esac
+ mock ${mock_args} --copyin "syslog-stdout.py" "builddir"
mock ${mock_args} --shell "( cd \"builddir/build/BUILD/$1\"; ./configure )"
mock ${mock_args} --shell \
- "make -C \"builddir/build/BUILD/$1/tests/functional/log_external\" \
- check-TESTS \"TESTS=../${_logfile}.sh\" $5" \
+ "python3 builddir/syslog-stdout.py \
+ >\"builddir/build/BUILD/$1/tests/functional/log_external/.syslog\" & \
+ { sleep 2; make -C \"builddir/build/BUILD/$1/tests/functional/log_external\" \
+ check-TESTS \"TESTS=../${_logfile}.sh\" $5; } \
+ && ! test -s \"builddir/build/BUILD/$1/tests/functional/log_external/.syslog\"; \
+ ret_ec=\$?; \
+ ( cd \"builddir/build/BUILD/$1/tests/functional/log_external\"; \
+ cat .syslog >> test-suite.log; \
+ echo SYSLOG-begin; cat .syslog; echo SYSLOG-end ); \
+ ret () { return \$1; }; ret \${ret_ec}" \
&& _result="${_result}_good" \
|| _result="${_result}_bad"
mock ${mock_args} --copyout \
diff --git a/tests/functional/syslog-stdout.py b/tests/functional/syslog-stdout.py
new file mode 100755
index 0000000..64baf4c
--- /dev/null
+++ b/tests/functional/syslog-stdout.py
@@ -0,0 +1,51 @@
+#!/usr/bin/python3
+# Copyright 2017 Red Hat, Inc.
+#
+# Author: Jan Pokorny <jpokorny@redhat.com>
+#
+# This file is part of libqb.
+#
+# libqb is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Lesser General Public License as published by
+# the Free Software Foundation, either version 2.1 of the License, or
+# (at your option) any later version.
+#
+# libqb is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public License
+# along with libqb. If not, see <http://www.gnu.org/licenses/>.
+
+"""Simple /dev/log to stdout forwarding"""
+
+import socket
+from atexit import register
+from os import remove
+from sys import argv
+
+# no locking, but anyway
+try:
+ remove("/dev/log")
+except FileNotFoundError:
+ pass
+sock = socket.socket(socket.AF_UNIX, socket.SOCK_DGRAM)
+sock.bind("/dev/log")
+
+def shutdown():
+ sock.close()
+ remove("/dev/log")
+
+def main(*argv):
+ register(shutdown)
+ while True:
+ try:
+ b = sock.recv(4096)
+ # flushing is crucial here
+ print(">>> " + str(b, 'ascii').split(' ', 3)[-1], flush=True)
+ except IOError:
+ pass
+
+if __name__ == '__main__':
+ main(*argv)
--
2.15.0
Reference in New Issue View Git Blame Copy Permalink