Re: [Tracker] Updated tracker cvs - ready for testing

["Edward Duffy" <eduffy gmail com>]

On 8/1/06, Laurent Aguerreche <laurent aguerreche free fr> wrote:
> Le lundi 31 juillet 2006 Ã 10:42 +0100, Jamie McCracken a Ãcrit :
> > Edward Duffy wrote:
> > > So, I ran trackerd through gdb and I got a stack trace this
> > > time...before I thought something other than trackerd was seg faulting
> > > since trackerd  was still up.  Anyway ... here the relevant top of the
> > > stack trace.
> > >
> > > #0  0xb7aa1d18 in strcmp () from /lib/tls/i686/cmov/libc.so.6
> > > #1  0x08062125 in alias_entry_cmp (v1=0xb78e71cc,
> > >     v2=0xb5f097c0) at xdgmimealias.c:95
> > > #2  0xb7a624a9 in bsearch () from /lib/tls/i686/cmov/libc.so.6
> > > #3  0x08062174 in _xdg_mime_alias_list_lookup (
> > >     list=0xb5f09e70, alias=0x6f <Address 0x6f out of bounds>)
> > >     at xdgmimealias.c:110
> > > #4  0x08061c6b in xdg_mime_unalias_mime_type (
> > >     mime_type=0x6f <Address 0x6f out of bounds>)
> > >     at xdgmime.c:614
> > > #5  0x08061cab in xdg_mime_mime_type_equal (
> > >     mime_a=0xb5f08988 "o",
> > >     mime_b=0x6f <Address 0x6f out of bounds>) at xdgmime.c:629
> > > #6  0x0806589c in _xdg_mime_magic_lookup_data (
> > >     mime_magic=0xb5f08548, data=0xb5f29310, len=2141,
> > >     mime_types=0xb78e72f8, n_mime_types=5)
> > >     at xdgmimemagic.c:692
> > > #7  0x08061a73 in xdg_mime_get_mime_type_for_file (
> > >     file_name=0xb5f32808 "/home/eduffy/media/music/John Williams/Star
> > > Wars III: Revenge of the Sith/", statbuf=0xb78e72a0) at xdgmime.c:516
> > > #8  0x0805fd1a in tracker_get_mime_type (
> > >     uri=0xb5f32808 "/home/eduffy/media/music/John Williams/Star Wars
> > > III: Revenge of the Sith/10 - Anakin's Dark Deeds.ogg")
> > >     at tracker-utils.c:1082
> > > #9  0x0804e73a in index_file (db_con=0xb78e73ec,
> > >     info=0xb5f327b8) at trackerd.c:588
> > > #10 0x0804fa6d in process_files_thread () at trackerd.c:1238
> >
> >
> > Thanks yes. I had updated to the latest xdgmime (which is causing the
> > crash) in tracker from the freedesktop cvs. If its buggy I should
> > probably revert. Any chance sending me the offending file off list (if
> > its small) so I know when its sorted?
>
> Sorry, but it seems that the bug is still there...
>
>
> #0  0xa7b1af48 in strcmp () from /lib/tls/i686/cmov/libc.so.6
> No symbol table info available.
> #1  0x08062096 in alias_entry_cmp (v1=0xa795919c, v2=0x877fed0)
>     at xdgmimealias.c:95
> No locals.
> #2  0xa7ada22d in bsearch () from /lib/tls/i686/cmov/libc.so.6
> No symbol table info available.
> #3  0x080620e5 in _xdg_mime_alias_list_lookup (list=0x877c6f8,
>     alias=0x2d782f6f <Address 0x2d782f6f out of bounds>) at
> xdgmimealias.c:110
>         entry = (XdgAlias *) 0xa7e54420
>         key = {alias = 0x2d782f6f <Address 0x2d782f6f out of bounds>,
>   mime_type = 0x0}
> #4  0x08061b9d in _xdg_mime_unalias_mime_type (
>     mime_type=0x2d782f6f <Address 0x2d782f6f out of bounds>) at
> xdgmime.c:612
>         lookup = 0x0
> #5  0x08061cd4 in _xdg_mime_mime_type_subclass (
>     mime=0x2d782f6f <Address 0x2d782f6f out of bounds>,
>     base=0x877a458 "PK\003\004/x-chdr") at xdgmime.c:693
>         umime = 0x80616e5 "\205ït\005ïn\003"
>         ubase = 0x877a458 "PK\003\004/x-chdr"
>         parents = (const char **) 0x0
> #6  0x08061ddc in xdg_mime_mime_type_subclass (
>     mime=0x2d782f6f <Address 0x2d782f6f out of bounds>,
>     base=0x877a458 "PK\003\004/x-chdr") at xdgmime.c:730
> No locals.
> #7  0x08061d93 in _xdg_mime_mime_type_subclass (mime=0x8779f88 "ïïw
> \b/x-csrc",
>     base=0x877a458 "PK\003\004/x-chdr") at xdgmime.c:717
>         umime = 0x8779f88 "ïïw\b/x-csrc"
>         ubase = 0x877a458 "PK\003\004/x-chdr"
>         parents = (const char **) 0x877bacc
> #8  0x0806569a in _xdg_mime_magic_lookup_data (mime_magic=0x86288f0,
>     data=0x878aa00, len=2141, mime_types=0xa7959308, n_mime_types=0)
>     at xdgmimemagic.c:675
>         match = (XdgMimeMagicMatch *) 0x878f818
>         mime_type = 0x877a458 "PK\003\004/x-chdr"
>         n = 0
>         priority = 50
>         had_match = 1
> #9  0x080619aa in xdg_mime_get_mime_type_for_file (
>     file_name=0x8795a10
> "/home/laurent/Desktop/mysql-5.0-3/mysql-dfsg-5.0-5.0.22.orig/bdb/dist/win_config.in", statbuf=0xa79592b0) at 
> xdgmime.c:516
>         mime_type = 0xa7f63684 "`e\b"
>         mime_types = {0x2c <Address 0x2c out of bounds>, 0xa7f63684 "`e
> \b",
>   0x0, 0x2c <Address 0x2c out of bounds>, 0xa7959348 "h\223\225ïï_ïï,"}
>         file = (FILE *) 0x880a3a0
>         data = (
>     unsigned char *) 0x878aa00 "/* Define to 1 if you want to build a
> version for running the test suite. */\n/* #undef CONFIG_TEST */\n\n/*
> Define to 1 if you want a debugging version. */\n/* #undef DEBUG */\n#if
> defined(_DEBUG)\n#if !de"...
>         max_extent = 2141
>         bytes_read = 2141
>         buf = {st_dev = 834, __pad1 = 0, st_ino = 11747413, st_mode =
> 33188,
>   st_nlink = 1, st_uid = 1002, st_gid = 1002, st_rdev = 0, __pad2 = 0,
>   st_size = 12745, st_blksize = 4096, st_blocks = 32, st_atim = {
>     tv_sec = 1154458969, tv_nsec = 0}, st_mtim = {tv_sec = 1148547406,
>     tv_nsec = 0}, st_ctim = {tv_sec = 1154458973, tv_nsec = 0},
> __unused4 = 0,
>   __unused5 = 0}
>         base_name = 0x8795a56 "win_config.in"
>         n = 0
> #10 0x0805fceb in tracker_get_mime_type (
>     uri=0x8795a10
> "/home/laurent/Desktop/mysql-5.0-3/mysql-dfsg-5.0-5.0.22.orig/bdb/dist/win_config.in") at tracker-utils.c:1082
>         result = 0xa7e54420 "U\211ï\213U\b\213B\f\203ï\001t\"\203ï
> \002t1ïB\b"
> #11 0x0804e79a in index_file (db_con=0xa79593fc, info=0x87959c0)
>     at trackerd.c:588
>         str_dir = 0x0
>         str_link = 0x4475714e <Address 0x4475714e out of bounds>
>         str_link_uri = 0x0
>         str_mtime = 0x44cfa55d <Address 0x44cfa55d out of bounds>
>         str_file_id = 0x0
>         meta_table = (GHashTable *) 0x8644880
>         str_atime = 0x0
>         name = 0x87cd4f0 "-rw-r--r--"
>         path = 0x87910c0 "ïåïå\001"
>         __PRETTY_FUNCTION__ = "index_file"
> #12 0x0804fac4 in process_files_thread () at trackerd.c:1238
>         info = (FileInfo *) 0x87959c0
>         db_con = {db = 0x8598d60, insert_contents_stmt = 0x1}
>         moved_from_list = (GSList *) 0x0
>         need_index = 1
>         has_pending = 1
>         res = (MYSQL_RES *) 0x877ffb0
>         row = (MYSQL_ROW) 0x880e63c
> #13 0xa7f1f5df in g_thread_create_full () from /usr/lib/libglib-2.0.so.0
> No symbol table info available.
> #14 0xa7e52260 in start_thread ()
> from /lib/tls/i686/cmov/libpthread.so.0
> No symbol table info available.
> #15 0xa7b7b3fe in clone () from /lib/tls/i686/cmov/libc.so.6
> No symbol table info available.
> (gdb)
>
>
> File responsible of that is attached.
>
>
> Laurent.
If you were to remove your ~/.Tracker directory and restart trackerd,
does it crash on a different file?  When I had the problem, I did this
about 6 or 7 times and it never crashed on the same file.

> _______________________________________________
> tracker-list mailing list
> tracker-list gnome org
> http://mail.gnome.org/mailman/listinfo/tracker-list