Bug#855933: mod-gnutls: FTBFS: Test failures
Hi Daniel, hi Lucas, I have pushed a suggested version 0.8.2-3 to the "for-debian" branch at commit 266c7750aa4c1a25089d3ad79e4b5359b9342214 in my Github repository [1]. I tried to keep the differences reasonably small, I hope the result is acceptable during the freeze. The set of changes assumes that (patch numbers refer to the numbered file names in debian/patches, 0001 is already in 0.8.2-2): * Verbose logging around the fixed bugs is desirable. That's why it includes patches 0002 to 0004, just increasing the timeouts could be one smaller patch. * Not breaking compatibility for jessie-backports is desirable (patch 0005). Ironically, this could be dropped without the verbose logging part above. * Fixing the tests for hurd-i386 is desirable, even though it's not a release architecture (patch 0006). * The update will go through unstable (patch 0007 for compatibility with GnuTLS 3.5.9). Regards, Thomas [1] https://github.com/airtower-luna/mod_gnutls/commits/for-debian
Bug#855933: GnuTLS 3.5.9 compatibility
If a patched upload goes through unstable, I suggest also including commit 61353933986460ea2d7bcbc57531ecd066795f0d (patch attached). It's just one line, but without it we'll be back to FTBFS when GnuTLS 3.5.9 hits unstable. From 61353933986460ea2d7bcbc57531ecd066795f0d Mon Sep 17 00:00:00 2001 From: Thomas Klute <thomas2.kl...@uni-dortmund.de> Date: Sun, 19 Feb 2017 18:57:56 +0100 Subject: [PATCH] Do not treat warnings about deprecated declarations as errors GnuTLS has declared OpenPGP support as deprecated in version 3.5.9. Treating deprecation warnings as errors causes the build to fail with this version, so exempt them from "-Werror" until OpenPGP support is removed from mod_gnutls. --- configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index cc3e8ae..3335773 100644 --- a/configure.ac +++ b/configure.ac @@ -64,7 +64,7 @@ AC_ARG_ENABLE(strict, STRICT_CFLAGS="" if test "$use_strict" != "no"; then -STRICT_CFLAGS="-Wall -Werror -Wextra" + STRICT_CFLAGS="-Wall -Werror -Wextra -Wno-error=deprecated-declarations" fi AC_MSG_CHECKING([whether to enable SRP functionality]) -- 2.9.3
Bug#855933: mod-gnutls: FTBFS: Test failures
These look like the timeout issues I discovered in the build logs for 0.8.2-2, see [1] for Daniel's report and [2] for my analysis, plus my two follow up mails if jessie-backports and hurd-i386 matter. I'm going to quote that mail below. Just a little clarification ahead: I've since confirmed that the issue was definitely the combination of parallel=32 and a pretty short timeout for the flock calls. All test cases got started simultaneously but had to wait their turn for the port(s) used in the tests (that's what the flock calls around Apache are for), effectively making the flock timeout (30 seconds by default) a timeout for the whole test suite except the last test case to complete. [1] https://lists.gnupg.org/pipermail/mod_gnutls-devel/2017-February/000185.html [2] https://lists.gnupg.org/pipermail/mod_gnutls-devel/2017-February/000186.html Looking at both build logs my best guess is that the tests are hitting timeouts, both when getting the server instance locks (used to prevent port conflicts) and running HTTPS requests against them. The log for test-15_basic_msva.bash on sparc64 even looks like the timeout hit in the middle of request handling. I've pushed three related commits to master (the attached patch combines all three), in order: 8184ad0eda43102d24d7cf158710e0109c7e293b Test suite: Run flock with "--verbose" to log timeouts bbfcbb57c2cb89790316b66bdd475c3a64ca4080 Test suite: Log if a process to be stopped by PID file is not running 6c030c14da928da3e05f232e3fb8dd5aa98c659a Test suite: Make timeouts for server locks and HTTPS requests configurable Commits 8184ad0eda43102d24d7cf158710e0109c7e293b and bbfcbb57c2cb89790316b66bdd475c3a64ca4080 just make the logging more detailed to show if the tests are really hitting timeouts. To change the timeouts, you can either cherry-pick 6c030c14da928da3e05f232e3fb8dd5aa98c659a and pass TEST_LOCK_WAIT and TEST_QUERY_TIMEOUT with large values (e.g. 300 or even 600, default is 30 [seconds]) to ./configure, or patch the values of lock_wait and TEST_QUERY_DELAY in test/Makefile.am (I've replaced/renamed the variables in the patch). > I can make force the entire build to not be parallel if you think it > would be useful to test. While i see "parallel=32" in the sparc64 > transcript, i don't see a "parallel=" note at all in ppc64. Running in parallel definitely makes hitting the timeouts more likely, but shouldn't cause any additional problems. You can force just the test part of the build to run serially by passing --disable-flock to ./configure, which will add the special target .NOTPARALLEL to test/Makefile. I'd prefer if you try the patches first, though.
Bug#851384: Fixed upstream
This is a bug in the program which generates the OCSP database used in the failed test. I believe I have fixed this issue upstream in version 0.8.2. A build in a qemubuilder mips environment produced a correct test database.
Bug#848743: Build problem fixed in mod_gnutls 0.8.1
I have confirmed that the patch in my previous mail works on i386, and released mod_gnutls 0.8.1 to fix the build failures on 32 bit architectures.
Bug#848743: mod_gnutls 0.8.0-1 build failures
It looks like the test failures were cause by bug #848339, which was fixed in libunbound2 1.6.0-2. Relevant log excerpts: Setting up the build environment (libunbound2 version): > Selecting previously unselected package libunbound2:amd64. > Preparing to unpack .../088-libunbound2_1.6.0-1_amd64.deb ... > Unpacking libunbound2:amd64 (1.6.0-1) ... Test log (look at any of the failed tests): > gnutls-cli: symbol lookup error: /usr/lib/x86_64-linux-gnu/libunbound.so.2: > undefined symbol: fake_dsa I believe a rebuild with libunbound2 1.6.0-2 will fix the test failures. However, there were also build failures on the 32 bit architectures, which are my fault for not using the right (portable) format modifier with apr_size_t in a few calls to ap_log_error() or apr_psprintf(). I didn't get to test a 32 bit build yet, but I believe the attached patch will fix the problem. diff --git a/src/gnutls_cache.c b/src/gnutls_cache.c index 3ebbd08..2b615f9 100644 --- a/src/gnutls_cache.c +++ b/src/gnutls_cache.c @@ -494,7 +494,7 @@ static gnutls_datum_t dbm_cache_fetch(mgs_handle_t *ctxt, gnutls_datum_t key) } ap_log_cerror(APLOG_MARK, APLOG_TRACE1, rv, ctxt->c, - "fetched %ld bytes from cache", + "fetched %" APR_SIZE_T_FMT " bytes from cache", dbmval.dsize); memcpy(data.data, dbmval.dptr + sizeof (apr_time_t), data.size); @@ -587,7 +587,8 @@ static int dbm_cache_store(server_rec *s, gnutls_datum_t key, apr_global_mutex_unlock(sc->cache->mutex); ap_log_error(APLOG_MARK, APLOG_TRACE1, rv, s, - "stored %ld bytes of data (%ld byte key) in cache '%s'", + "stored %" APR_SIZE_T_FMT " bytes of data (%" + APR_SIZE_T_FMT " byte key) in cache '%s'", dbmval.dsize, dbmkey.dsize, sc->cache_config); apr_pool_destroy(spool); diff --git a/src/gnutls_util.c b/src/gnutls_util.c index 9905bad..5e98bd4 100644 --- a/src/gnutls_util.c +++ b/src/gnutls_util.c @@ -28,7 +28,7 @@ const char* http_post_header(apr_pool_t *p, apr_uri_t *uri, "Host: %s\r\n" "Content-Type: %s\r\n" "Accept: %s\r\n" -"Content-Length: %ld\r\n\r\n", +"Content-Length: %" APR_SIZE_T_FMT "\r\n\r\n", apr_uri_unparse(p, uri, APR_URI_UNP_OMITSITEPART), uri->hostname, content_type, accept != NULL ? accept : "*/*",
Bug#635711: Still broken in monkeysphere 0.37-2 on sid
I still see this bug in monkeysphere 0.37-2 on sid (fresh stable install, upgrade through testing to unstable). Aptitude installation: Setting up monkeysphere (0.37-2) ... adding monkeysphere user... ms: setting up Monkeysphere authentication trust core... Failed running transition script /usr/share/monkeysphere/transitions/0.23 dpkg: error processing package monkeysphere (--configure): subprocess installed post-installation script returned error exit status 141 [... other packages ...] Errors were encountered while processing: monkeysphere E: Sub-process /usr/bin/dpkg returned an error code (1) Failed to perform requested operation on package. Trying to recover: Setting up monkeysphere (0.37-2) ... ms: setting up Monkeysphere authentication trust core... /usr/share/monkeysphere/ma/setup: line 73: printf: write error: Broken pipe Failed running transition script /usr/share/monkeysphere/transitions/0.23 dpkg: error processing package monkeysphere (--configure): subprocess installed post-installation script returned error exit status 1 Errors were encountered while processing: monkeysphere Running the transition script manually doesn't produce any output that tells me what is wrong, but the return value is the same: # /usr/share/monkeysphere/transitions/0.23; echo $? ms: setting up Monkeysphere authentication trust core... 141 -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
