Valgrind errors in NesterServer.* cause subsequent tests (ServerDisconnect, ServerStartup, UnresponsiveClient) to fail

[Daniel van Vugt]

Tests always fail together: ServerDisconnect, ServerStartup, UnresponsiveClient

17:27:52 9: [ FAILED ] 5 tests, listed below:
17:27:52 9: [ FAILED ] ServerDisconnect.is_detected_by_client
17:27:52 9: [ FAILED ] ServerDisconnect.doesnt_stop_client_calling_API_functions
17:27:52 9: [ FAILED ] ServerStartup.creates_endpoint_on_filesystem
17:27:52 9: [ FAILED ] ServerStartup.after_server_sigkilled_can_start_new_instance
17:27:52 9: [ FAILED ] UnresponsiveClient.does_not_hang_server

[https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=clang,platform=mesa,release=vivid+overlay/1772/consoleFull]

This is happening regularly in CI. These failures occur together.

[Daniel van Vugt (vanvugt)]

This is happening too often. Bumping importance.

[Alexandros Frantzis (afrantzis)]

The failures are caused by valgrind errors found in previous tests. The tests fail together because the all fork processes which, when exiting, fail due to the pending valgrind errors.

[Mir CI Bot (mir-ci-bot)]

Fix committed into lp:mir at revision None, scheduled for release in mir, milestone 0.25.0

[Daniel van Vugt (vanvugt)]

Still happening (from a new root cause?)

https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=clang,platform=mesa,release=yakkety/2167/consoleFull

[Daniel van Vugt (vanvugt)]

Still happening in the 0.24 branch:

https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=i386,compiler=gcc,platform=mesa,release=xenial+overlay/2340/consoleFull

[Daniel van Vugt (vanvugt)]

Fix committed to lp:mir/0.24 at revision 3640, scheduled for release in Mir 0.24.1

[Launchpad Janitor (janitor)]

This bug was fixed in the package mir - 0.24.1+16.10.20160928-0ubuntu1

---------------
mir (0.24.1+16.10.20160928-0ubuntu1) yakkety; urgency=medium

  * New upstream release 0.24.1 (https://launchpad.net/mir/+milestone/0.24.1)
    - ABI summary:
      . All ABI numbers remain unchanged.
      . mircommon ABI at 6, but now including a symbol misplaced in 0.24.0
    - Bugs fixed:
      . mirserver ignored further keymap configuration changes after a keyboard
        is configured (LP: #1626435)
      . fix for libmircommon.so.6 ABI break: version stanza of
        mir::dispatch::ReadableFd* moved to newer version (LP: #1617865)
      . clang builds fail (again) (LP: #1609612)
      . Valgrind errors in NesterServer.* cause subsequent tests
        (ServerDisconnect, ServerStartup, UnresponsiveClient) to fail
        (LP: #1612012)
      . Break potentially infinite loop in search for next session to focus
        (LP: #1625401)

 -- Andreas Pokorny <email address hidden> Wed, 28 Sep 2016 12:09:49 +0000

[Cemil Azizoglu (cemil-azizoglu)]

Just had this happen again :

https://mir-jenkins.ubuntu.com/job/build-2-binpkg-mir/arch=amd64,compiler=clang,platform=mesa,release=yakkety/2727/consoleFull

[Daniel van Vugt (vanvugt)]

Indeed the cause is again valgrind errors in NestedServer tests. But should we be reusing the same bug? Just search for "==12777==":

20:13:30 9: [----------] 25 tests from NestedServer
20:13:30 9: [ RUN ] NestedServer.nested_platform_connects_and_disconnects
20:13:31 9: [2016-11-02 20:13:31.685261] mirserver: Starting
20:13:31 9: [2016-11-02 20:13:31.689667] mirserver: Selected driver: mir:stub-graphics (version 0.25.0)
20:13:31 9: [2016-11-02 20:13:31.733003] mirserver: Using software cursor
20:13:32 9: [2016-11-02 20:13:32.118804] mirserver: Selected input driver: mir:stub-input (version: 0.25.0)
20:13:32 9: [2016-11-02 20:13:32.128108] mirserver: Mir version 0.25.0
20:13:32 9: [2016-11-02 20:13:32.211391] mirserver: Starting
20:13:34 9: [2016-11-02 20:13:34.070068] mirserver: Using nested cursor
20:13:34 9: ==12777== Thread 9:
20:13:34 9: ==12777== Invalid read of size 4
20:13:34 9: ==12777== at 0x540688F: MirWaitHandle::wait_for_all()::$_0::operator()() const (mir_wait_handle.cpp:53)
20:13:34 9: ==12777== by 0x5406401: void std::condition_variable::wait<MirWaitHandle::wait_for_all()::$_0>(std::unique_lock<std::mutex>&, MirWaitHandle::wait_for_all()::$_0) (condition_variable:98)
20:13:34 9: ==12777== by 0x5406376: MirWaitHandle::wait_for_all() (mir_wait_handle.cpp:53)
20:13:34 9: ==12777== by 0x54774D9: mir_connection_create_buffer_stream_sync (mir_buffer_stream_api.cpp:76)
20:13:34 9: ==12777== by 0x517576D: (anonymous namespace)::MirClientHostSurface::set_cursor_image(mir::graphics::CursorImage const&) (mir_client_host_connection.cpp:168)
20:13:34 9: ==12777== by 0x5175D95: mir::graphics::nested::MirClientHostConnection::set_cursor_image(mir::graphics::CursorImage const&) (mir_client_host_connection.cpp:430)
....

[Michał Sawicz (saviq)]

Syncing task from Mir.