| Summary: | Crash - mdb_txn_begin (m_env=0) | ||
|---|---|---|---|
| Product: | [Frameworks and Libraries] frameworks-baloo | Reporter: | Wyatt Childers <kdebugs.81do7> |
| Component: | general | Assignee: | Pinak Ahuja <pinak.ahuja> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | crash | CC: | alexey.min, alexminder, andysem, arthur.marsh, aseques, aspotashev, asturm, bhush94, borden_c, bugs.kde.org, christoph, cobexer, corsaroangelo, crissi99, dariovmartine, devguy.ca, gauvain, gojunkmailx, grantipak, hebert.soares, hemlock-solmes, iganem, kde, kenaaker, lluis, mrl586, netphilos, oded, peter.pille, peterchibunna, pinak.ahuja, plasma-bugs, pstrong, rdieter, rushpl, slawek, stupor_scurvy343, wacha, yzhernand, zhaixiang |
| Priority: | NOR | Keywords: | drkonqi |
| Version: | 5.17.0 | ||
| Target Milestone: | --- | ||
| Platform: | openSUSE | ||
| OS: | Linux | ||
| Latest Commit: | Version Fixed In: | ||
| Sentry Crash Report: | |||
| Attachments: |
plasmashell-20151214-070254.kcrash.txt
attachment-31661-0.html plasmashell-20160621-192127.kcrash.txt Full crash stack trace |
||
|
Description
Wyatt Childers
2015-09-29 22:07:24 UTC
Happens to me as well. *** Bug 353329 has been marked as a duplicate of this bug. *** *** Bug 353860 has been marked as a duplicate of this bug. *** *** Bug 354184 has been marked as a duplicate of this bug. *** *** Bug 354231 has been marked as a duplicate of this bug. *** Is it possible to get a slightly more descriptive bug title to reduce the number of duplicate reports? I never would have guessed before filing my bug report that this was the duplicate. https://github.com/LMDB/lmdb/issues/5 liblmdb issue mdb_txn_begin(MDB_env *env, MDB_txn *parent, unsigned int flags, MDB_txn **ret) does not check parameter MDB_env *env is NULL! and in void mdb_txn_abort(MDB_txn *txn), it also forget to check txn->mt_env *** Bug 355162 has been marked as a duplicate of this bug. *** *** Bug 355000 has been marked as a duplicate of this bug. *** *** Bug 354606 has been marked as a duplicate of this bug. *** *** Bug 354084 has been marked as a duplicate of this bug. *** *** Bug 353238 has been marked as a duplicate of this bug. *** So, is this really an upstream bug and to be solved there as seen in comment #7? (In reply to andreas.sturmlechner from comment #13) > So, is this really an upstream bug and to be solved there as seen in comment > #7? We have worked around it for 5.17. We really should have had more error checking on our part. I'm leaving this open for now. I want to see if we still get any crashes as it was not easily reproducible. Created attachment 96073 [details]
plasmashell-20151214-070254.kcrash.txt
In that case I'll attach my backtrace from after upgrade to kde-frameworks-5.17 (I think it's related).
Please also note that I have baloo file search disabled.
@Andreas: Could you possible send me your Baloo database? It should be present at ~/.local/share/baloo/index. This would really help be diagnose what is going wrong. Created attachment 96086 [details] attachment-31661-0.html Hi Vishesh: Sorry. I don't longer have OpenSuse Tumbleweed + KDE installed because of stability issues. I changed to Leap + KDE and everything is working fine. Thanks a lot. El mar., 15 de diciembre de 2015 12:35, Vishesh Handa via KDE Bugzilla < bugzilla_noreply@kde.org> escribió: > https://bugs.kde.org/show_bug.cgi?id=353342 > > --- Comment #16 from Vishesh Handa <me@vhanda.in> --- > @Andreas: Could you possible send me your Baloo database? > > It should be present at ~/.local/share/baloo/index. This would really help > be > diagnose what is going wrong. > > -- > You are receiving this mail because: > You are on the CC list for the bug. > (In reply to Vishesh Handa from comment #16) > @Andreas: Could you possible send me your Baloo database? > > It should be present at ~/.local/share/baloo/index. This would really help > be diagnose what is going wrong. Btw you should have received it some time ago, unless it ended up in your spam filter. *** Bug 357169 has been marked as a duplicate of this bug. *** *** Bug 358016 has been marked as a duplicate of this bug. *** *** Bug 360154 has been marked as a duplicate of this bug. *** Anything I can do to help debug this? I hit this daily. @Michael could you get me the output of $ balooctl checkDb and it would also be helpful if you could upload your baloo database (~/.local/share/baloo/index) so that I can try looking at what's wrong. (In reply to Pinak Ahuja from comment #23) > @Michael could you get me the output of > $ balooctl checkDb > and it would also be helpful if you could upload your baloo database > (~/.local/share/baloo/index) > so that I can try looking at what's wrong. $ balooctl checkDb Baloo Index could not be opened I have baloo disabled and unfortunately recently deleted the database to see if it would help with the crashes (it seems to). A corrupted/incomplete database could indeed be the cause - if I recall correctly, I did have to manually kill baloo_file_extractor in the past. Created attachment 99645 [details]
plasmashell-20160621-192127.kcrash.txt
Still happens in 16.04 LTS : Backtrace of the crash I experienced: Application: Baloo File Indexing Daemon (baloo_file), signal: Segmentation fault Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". [Current thread is 1 (Thread 0x7f2bf80438c0 (LWP 4015))] Thread 2 (Thread 0x7f2bed297700 (LWP 4030)): [KCrash Handler] #6 mdb_txn_begin (env=0x0, parent=0x0, flags=131072, ret=0x7f2bed296c38) at mdb.c:2773 #7 0x000000000041b5fd in Baloo::FirstRunIndexer::run (this=0x222c7f0) at /build/baloo-kf5-wIK3t6/baloo-kf5-5.18.0/src/file/firstrunindexer.cpp:46 #8 0x00007f2bf6b51343 in QThreadPoolThread::run (this=0x22557c0) at thread/qthreadpool.cpp:93 #9 0x00007f2bf6b5484e in QThreadPrivate::start (arg=0x22557c0) at thread/qthread_unix.cpp:331 #10 0x00007f2bf57b86fa in start_thread () from /lib/x86_64-linux-gnu/libpthread.so.0 #11 0x00007f2bf6161b5d in clone () from /lib/x86_64-linux-gnu/libc.so.6 Thread 1 (Thread 0x7f2bf80438c0 (LWP 4015)): #0 0x00007f2bf61519cd in read () from /lib/x86_64-linux-gnu/libc.so.6 #1 0x00007f2bf3369740 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #2 0x00007f2bf3325e84 in g_main_context_check () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #3 0x00007f2bf3326340 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #4 0x00007f2bf33264ac in g_main_context_iteration () from /lib/x86_64-linux-gnu/libglib-2.0.so.0 #5 0x00007f2bf6d8ba7f in QEventDispatcherGlib::processEvents (this=0x216e570, flags=...) at kernel/qeventdispatcher_glib.cpp:418 #6 0x00007f2bf6d32dea in QEventLoop::exec (this=this@entry=0x7ffd9610d6a0, flags=..., flags@entry=...) at kernel/qeventloop.cpp:204 #7 0x00007f2bf6d3ae8c in QCoreApplication::exec () at kernel/qcoreapplication.cpp:1229 #8 0x000000000041698c in main (argc=1, argv=<optimized out>) at /build/baloo-kf5-wIK3t6/baloo-kf5-5.18.0/src/file/main.cpp:88 falken@wopr:~$ balooctl checkDb Baloo Index could not be opened (In reply to Tom Chiverton from comment #27) > falken@wopr:~$ balooctl checkDb > > Baloo Index could not be opened Can you try moving the baloo database somewhere safe (please keep it as it could be useful for debugging if the issue is resolved) and restarting plasma to see if the crashes stop? The only thing that worked was balooctl disable; balooctl enable and now balooctl status reports Baloo File Indexer is running Indexer state: Idle Indexed 15624 / 118006 files Current size of index is 74.84 MiB Given the well defined exception, and recovery steps, why doesn't Baloo do disable /enable automatically when it occurs? Overhead of a rescan less bad than no indexing *** Bug 366807 has been marked as a duplicate of this bug. *** *** Bug 363476 has been marked as a duplicate of this bug. *** *** Bug 367504 has been marked as a duplicate of this bug. *** *** Bug 353995 has been marked as a duplicate of this bug. *** *** Bug 366824 has been marked as a duplicate of this bug. *** *** Bug 366525 has been marked as a duplicate of this bug. *** *** Bug 366354 has been marked as a duplicate of this bug. *** *** Bug 366189 has been marked as a duplicate of this bug. *** *** Bug 365373 has been marked as a duplicate of this bug. *** *** Bug 361562 has been marked as a duplicate of this bug. *** *** Bug 362966 has been marked as a duplicate of this bug. *** *** Bug 359420 has been marked as a duplicate of this bug. *** *** Bug 363563 has been marked as a duplicate of this bug. *** *** Bug 363725 has been marked as a duplicate of this bug. *** *** Bug 363182 has been marked as a duplicate of this bug. *** *** Bug 363282 has been marked as a duplicate of this bug. *** *** Bug 360838 has been marked as a duplicate of this bug. *** *** Bug 364993 has been marked as a duplicate of this bug. *** *** Bug 360696 has been marked as a duplicate of this bug. *** *** Bug 360758 has been marked as a duplicate of this bug. *** *** Bug 363552 has been marked as a duplicate of this bug. *** *** Bug 358797 has been marked as a duplicate of this bug. *** *** Bug 361020 has been marked as a duplicate of this bug. *** *** Bug 356301 has been marked as a duplicate of this bug. *** *** Bug 361285 has been marked as a duplicate of this bug. *** *** Bug 360694 has been marked as a duplicate of this bug. *** *** Bug 355543 has been marked as a duplicate of this bug. *** *** Bug 356426 has been marked as a duplicate of this bug. *** Created attachment 100894 [details]
Full crash stack trace
Another possibly useful stack trace of this crash on KDE frameworks 5.24.0, SUSE tumbleweed.
*** Bug 368238 has been marked as a duplicate of this bug. *** *** Bug 362088 has been marked as a duplicate of this bug. *** I "fixed" that baloo_file now detects if the db can't be opened and recreates it, therefore the baloo_file crashs here are solved, but yeah, I am not sure if the Query handles corrupted databases very well :/ https://quickgit.kde.org/?p=baloo.git&a=commit&h=a03b0caa4ca1fbfc249bfc0c2730aac340bbf929 Use other bug for query problem if db is corrupt. *** This bug has been marked as a duplicate of bug 360946 *** |