Bug 287966

Summary: Akonadi out-of-memory-crash
Product: [Frameworks and Libraries] Akonadi Reporter: Bernd Oliver Sünderhauf <pancho.s>
Component: Nepomuk Feeder AgentsAssignee: kdepim bugs <kdepim-bugs>
Status: RESOLVED DUPLICATE    
Severity: crash CC: boboudrescu, myriam, vkrause
Priority: NOR    
Version: 4.7   
Target Milestone: ---   
Platform: Ubuntu   
OS: Linux   
Latest Commit: Version Fixed In:
Sentry Crash Report:
Attachments: Memory and CPU use of Akonadi right before the crash

Description Bernd Oliver Sünderhauf 2011-12-01 08:33:26 UTC 
Application: akonadi_nepomuk_email_feeder (4.7)
KDE Platform Version: 4.7.3 (4.7.3)
Qt Version: 4.7.4
Operating System: Linux 3.0.0-14-generic x86_64
Distribution: Ubuntu 11.10

-- Information about the crash:
- What I was doing when the application crashed:
Just started the system and launched Thunderbird, when Akonadi crashed.

- Unusual behavior I noticed:
For some days, the system's responsivity has been very bad right after starting the system. Whenever I wait some 10 minutes until things calmed down, I can continue working.
However, if I start any application without waiting long enough, the system regularly gets out of memory. Sometimes it's just the Desktop Effects being disabled, sometimes Akonadi crashes (such as here) and sometimes the whole system goes down with a black screen.

The crash can be reproduced some of the time.

-- Backtrace:
Application: Akonadi Agent (akonadi_nepomuk_email_feeder), signal: Aborted
[KCrash Handler]
#6  0x00007fbd670f53a5 in __GI_raise (sig=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:64
#7  0x00007fbd670f8b0b in __GI_abort () at abort.c:92
#8  0x00007fbd6772fd7d in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#9  0x00007fbd6772df26 in ?? () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#10 0x00007fbd6772df53 in std::terminate() () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#11 0x00007fbd6772e096 in __cxa_rethrow () from /usr/lib/x86_64-linux-gnu/libstdc++.so.6
#12 0x00007fbd6a0de15b in QList<Akonadi::NotificationMessage>::append (this=0x7fff9f8682e0, t=...) at /usr/include/qt4/QtCore/qlist.h:520
#13 0x00007fbd6a0df6cd in operator<< (t=..., this=0x7fff9f8682e0) at /usr/include/qt4/QtCore/qlist.h:321
#14 loadFrom (device=0x7fff9f868140, this=0x24ec530) at ../../akonadi/changerecorder_p.h:160
#15 Akonadi::ChangeRecorderPrivate::loadNotifications (this=0x24ec530) at ../../akonadi/changerecorder_p.h:118
#16 0x00007fbd6a0dbae7 in Akonadi::ChangeRecorder::setConfig (this=<optimized out>, settings=<optimized out>) at ../../akonadi/changerecorder.cpp:50
#17 0x00007fbd6a0bcad7 in Akonadi::AgentBasePrivate::init (this=0x24d4dc0) at ../../akonadi/agentbase.cpp:216
#18 0x00007fbd6a0bd717 in Akonadi::AgentBase::AgentBase (this=0x24d55d0, id=...) at ../../akonadi/agentbase.cpp:510
#19 0x000000000041b566 in NepomukFeederAgentBase::NepomukFeederAgentBase (this=0x24d55d0, id=<optimized out>) at ../../nepomuk_email_feeder/shared/nepomukfeederagentbase.cpp:98
#20 0x000000000040e7b7 in NepomukFeederAgent (id=<optimized out>, this=0x24d55d0) at ../../nepomuk_email_feeder/shared/nepomukfeederagent.h:37
#21 Akonadi::NepomukEMailFeeder::NepomukEMailFeeder (this=0x24d55d0, id=<optimized out>) at ../../nepomuk_email_feeder/nepomukemailfeeder.cpp:43
#22 0x000000000041041d in Akonadi::AgentBase::init<Akonadi::NepomukEMailFeeder> (argc=<optimized out>, argv=<optimized out>) at /usr/include/akonadi/agentbase.h:335
#23 0x00007fbd670e030d in __libc_start_main (main=0x40e2e0 <main(int, char**)>, argc=3, ubp_av=0x7fff9f868bf8, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7fff9f868be8) at libc-start.c:226
#24 0x000000000040e375 in _start ()

Possible duplicates by query: bug 286786, bug 263758.

Reported using DrKonqi
Comment 1 Bernd Oliver Sünderhauf 2011-12-01 08:37:09 UTC 
Created attachment 66247 [details]
Memory and CPU use of Akonadi right before the crash
Comment 2 Bogdan Udrescu 2011-12-02 13:02:16 UTC 
related or duplicate of bug 283922
Comment 3 Bernd Oliver Sünderhauf 2011-12-04 00:12:12 UTC 
Here's a syslog snippet with the trace of a situation where Akonadi brought the system down with a black screen. Is an upstream kernel bug the culprit? Or is this just a coincidence?

---

Dec  3 23:55:16 Luca kernel: [53391.460953] kernel BUG at /build/buildd/linux-3.0.0/fs/btrfs/extent_io.c:739!
Dec  3 23:55:16 Luca kernel: [53391.461041] invalid opcode: 0000 [#1] SMP 
Dec  3 23:55:16 Luca kernel: [53391.461095] CPU 0 
Dec  3 23:55:16 Luca kernel: [53391.461119] Modules linked in: dm_crypt snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event ppdev snd_seq snd_timer snd_seq_device k10temp snd soundcore snd_page_alloc sp5100_tco edac_core i2c_piix4 edac_mce_amd parport_pc shpchp lp parport btrfs zlib_deflate libcrc32c radeon ttm drm_kms_helper drm usbhid usb_storage hid uas i2c_algo_bit pata_atiixp wmi r8169 ahci libahci
Dec  3 23:55:16 Luca kernel: [53391.461658] 
Dec  3 23:55:16 Luca kernel: [53391.461677] Pid: 4543, comm: akonadi_imap_re Tainted: G        W   3.0.0-14-generic #23-Ubuntu MICRO-STAR INTERNATIONAL CO.,LTD MS-7596/785GM-E65 (MS-7596)
Dec  3 23:55:16 Luca kernel: [53391.461854] RIP: 0010:[<ffffffffa023a09d>]  [<ffffffffa023a09d>] set_extent_bit+0x29d/0x510 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.461979] RSP: 0000:ffff8801140d7888  EFLAGS: 00010246
Dec  3 23:55:16 Luca kernel: [53391.462046] RAX: 0000000000000000 RBX: 000000000001f000 RCX: 0000000000000001
Dec  3 23:55:16 Luca kernel: [53391.462133] RDX: 0000000000000001 RSI: 0000000000000040 RDI: 0000000000000206
Dec  3 23:55:16 Luca kernel: [53391.462221] RBP: ffff8801140d7928 R08: ffffffff81ab6110 R09: 0000000000000040
Dec  3 23:55:16 Luca kernel: [53391.462309] R10: 0000000000000000 R11: 0000000000000000 R12: 000000000001f000
Dec  3 23:55:16 Luca kernel: [53391.462397] R13: ffff880118944420 R14: 0000000000000050 R15: 0000000000000010
Dec  3 23:55:16 Luca kernel: [53391.462485] FS:  00007fa080e08780(0000) GS:ffff88011fc00000(0000) knlGS:00000000f6738880
Dec  3 23:55:16 Luca kernel: [53391.462586] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
Dec  3 23:55:16 Luca kernel: [53391.462656] CR2: 00007f64223da640 CR3: 0000000118ffe000 CR4: 00000000000006f0
Dec  3 23:55:16 Luca kernel: [53391.462744] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Dec  3 23:55:16 Luca kernel: [53391.462831] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Dec  3 23:55:16 Luca kernel: [53391.462918] Process akonadi_imap_re (pid: 4543, threadinfo ffff8801140d6000, task ffff880093061720)
Dec  3 23:55:16 Luca kernel: [53391.463029] Stack:
Dec  3 23:55:16 Luca kernel: [53391.463054]  0000000000000000 ffff8801140d7920 ffff880100000000 ffff8801140d7960
Dec  3 23:55:16 Luca kernel: [53391.463149]  000000081427c0f8 000000000001ffff ffff880118944420 ffff8801140d7fd8
Dec  3 23:55:16 Luca kernel: [53391.463244]  0000000000001000 0000000000000000 000000000001e000 0000100800000010
Dec  3 23:55:16 Luca kernel: [53391.463339] Call Trace:
Dec  3 23:55:16 Luca kernel: [53391.463377]  [<ffffffffa023a4e8>] lock_extent_bits+0x88/0xb0 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.463464]  [<ffffffffa023a526>] lock_extent+0x16/0x20 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.463544]  [<ffffffffa023b6ae>] __extent_read_full_page+0xce/0x6e0 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.463633]  [<ffffffff8115fe68>] ? __mem_cgroup_commit_charge_lrucare+0x78/0xf0
Dec  3 23:55:16 Luca kernel: [53391.463727]  [<ffffffff8110af04>] ? add_to_page_cache_locked.part.29+0x94/0x100
Dec  3 23:55:16 Luca kernel: [53391.463825]  [<ffffffffa021df10>] ? btrfs_read_locked_inode+0x410/0x410 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.463923]  [<ffffffffa023ca90>] extent_readpages+0xc0/0x100 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.464022]  [<ffffffffa021df10>] ? btrfs_read_locked_inode+0x410/0x410 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.464119]  [<ffffffffa021c80f>] btrfs_readpages+0x1f/0x30 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.464198]  [<ffffffff81115378>] read_pages+0x48/0x100
Dec  3 23:55:16 Luca kernel: [53391.464263]  [<ffffffff81115593>] __do_page_cache_readahead+0x163/0x180
Dec  3 23:55:16 Luca kernel: [53391.468624]  [<ffffffff81115901>] ra_submit+0x21/0x30
Dec  3 23:55:16 Luca kernel: [53391.472958]  [<ffffffff815d7bb8>] do_sync_mmap_readahead.isra.26+0x94/0x9c
Dec  3 23:55:16 Luca kernel: [53391.477338]  [<ffffffff8110cc11>] filemap_fault+0x361/0x3e0
Dec  3 23:55:16 Luca kernel: [53391.481690]  [<ffffffff8112c6b4>] __do_fault+0x54/0x510
Dec  3 23:55:16 Luca kernel: [53391.486024]  [<ffffffff8110c543>] ? generic_file_aio_read+0x153/0x290
Dec  3 23:55:16 Luca kernel: [53391.490372]  [<ffffffff8112fdca>] handle_pte_fault+0xfa/0x210
Dec  3 23:55:16 Luca kernel: [53391.494707]  [<ffffffff81130298>] handle_mm_fault+0x1f8/0x350
Dec  3 23:55:16 Luca kernel: [53391.498997]  [<ffffffff81103a90>] ? __perf_event_task_sched_out+0x30/0x60
Dec  3 23:55:16 Luca kernel: [53391.503321]  [<ffffffff815eedd3>] do_page_fault+0x153/0x530
Dec  3 23:55:16 Luca kernel: [53391.507655]  [<ffffffff81009702>] ? __switch_to+0x132/0x310
Dec  3 23:55:16 Luca kernel: [53391.511998]  [<ffffffff8104e94d>] ? set_next_entity+0x9d/0xb0
Dec  3 23:55:16 Luca kernel: [53391.516325]  [<ffffffff8104e5e9>] ? finish_task_switch+0x49/0xf0
Dec  3 23:55:16 Luca kernel: [53391.520660]  [<ffffffff815e8db4>] ? __schedule+0x3d4/0x700
Dec  3 23:55:16 Luca kernel: [53391.525011]  [<ffffffff815ebad5>] page_fault+0x25/0x30
Dec  3 23:55:16 Luca kernel: [53391.529317] Code: 4c 89 e3 0f 85 e5 fd ff ff 44 8b 7d b8 45 85 ff 0f 84 d8 fd ff ff 8b 7d 18 e8 d0 e8 ff ff 48 85 c0 48 89 45 a8 0f 85 c3 fd ff ff <0f> 0b 90 49 83 7d 28 00 4d 8d 7d 10 0f 84 e2 fd ff ff 48 39 c3 
Dec  3 23:55:16 Luca kernel: [53391.538454] RIP  [<ffffffffa023a09d>] set_extent_bit+0x29d/0x510 [btrfs]
Dec  3 23:55:16 Luca kernel: [53391.543076]  RSP <ffff8801140d7888>
Dec  3 23:55:16 Luca kernel: [53391.583020] ---[ end trace 0d896fee0a7c08f1 ]---
Comment 4 Myriam Schweingruber 2011-12-08 09:28:28 UTC 
I think this is just a coincidence.

*** This bug has been marked as a duplicate of bug 283922 ***