Overview
Request 215346 accepted
- refresh libmicrohttpd-drop-test_tls_options.patch
- add there also ignoring test_https_sni
- Update to 0.9.33
+ Fixed an issue with a missing argument in the postexample.
+ Fixed issue with bogus offset increment involving sendfile on
GNU/Linux. Adding support for SNI.
+ Fix for per-worker daemon pipes enabled with
MHD_USE_SUSPEND_RESUME that were not closed in MHD_stop_daemon.
+ Fixing warnings and build issue if --disable-https is given to
configure.
+
0.9.32:
+ Security fix: do not read past 0-terminator when unescaping
strings (thanks to Florian Weimer for reporting).
+ Signaling n times for shutdown works, but for resume we need to
wake up the correct daemon. Even if we signal n times in
that case also, there's no guarantee that some daemon can't
run through its select loop more than once before the
daemon we want to wake up gets a chance to read. Thus we
need a signal pipe per thread in the thread pool IF
MHD_suspend_connection is used. This introduces a new flag
MHD_USE_SUSPEND_RESUME to add those additional pipes and
only allow MHD_suspend_connection to be used in conjunction
with this flag.
Also, as MHD_resume_connection() will be called on a
non-daemon thread, but none of the queue insert/delete
calls are thread safe, we need to be concerned about (a)
corrupting the queue, and (b) having to add mutex
protection around every access to the queues, including
loops through timer queues, etc. This wasn't a problem (forwarded request 215320 from sleep_walker)
Request History
msmeissn created request
- refresh libmicrohttpd-drop-test_tls_options.patch
- add there also ignoring test_https_sni
- Update to 0.9.33
+ Fixed an issue with a missing argument in the postexample.
+ Fixed issue with bogus offset increment involving sendfile on
GNU/Linux. Adding support for SNI.
+ Fix for per-worker daemon pipes enabled with
MHD_USE_SUSPEND_RESUME that were not closed in MHD_stop_daemon.
+ Fixing warnings and build issue if --disable-https is given to
configure.
+
0.9.32:
+ Security fix: do not read past 0-terminator when unescaping
strings (thanks to Florian Weimer for reporting).
+ Signaling n times for shutdown works, but for resume we need to
wake up the correct daemon. Even if we signal n times in
that case also, there's no guarantee that some daemon can't
run through its select loop more than once before the
daemon we want to wake up gets a chance to read. Thus we
need a signal pipe per thread in the thread pool IF
MHD_suspend_connection is used. This introduces a new flag
MHD_USE_SUSPEND_RESUME to add those additional pipes and
only allow MHD_suspend_connection to be used in conjunction
with this flag.
Also, as MHD_resume_connection() will be called on a
non-daemon thread, but none of the queue insert/delete
calls are thread safe, we need to be concerned about (a)
corrupting the queue, and (b) having to add mutex
protection around every access to the queues, including
loops through timer queues, etc. This wasn't a problem (forwarded request 215320 from sleep_walker)
factory-auto added a reviewer
Please review sources
factory-auto added a reviewer
Please review build success
factory-auto accepted review
Check script succeeded
factory-auto added a reviewer
Check Staging Project
licensedigger accepted review
{"approve": "preliminary, version number changed"}
factory-repo-checker accepted review
Builds for repo openSUSE_Factory
coolo declined request
3 builds of this update all show the same problem: a deadlock in the test suite
It all ends the same:
[ 105s] make[4]: Leaving directory `/home/abuild/rpmbuild/BUILD/libmicrohttpd-0.9.33/src/testcurl'
[ 105s] /usr/bin/make check-TESTS
[ 105s] make[4]: Entering directory `/home/abuild/rpmbuild/BUILD/libmicrohttpd-0.9.33/src/testcurl'
Then nothing for hours
msmeissn added a reviewer
cannot reproduce build failure. openSUSE:Factory:Staging:D looks good. openSUSE 12.3 does not build as its gnutls is too old
coolo declined request
staging:D doesn't have this update - and when it had it has shown the problem I declined it for.
I'm aware it's not reproducible, but it's real. See mail
coolo added a reviewer
I just noticed in Staging:C that the old version has the same problem. So possibly you want to fix the problem - but indepdendent of this update
coolo accepted review
dimstar approved review
ok
dimstar accepted review
ok
coolo accepted request
checkin