tracker-miner-fs crashed with SIGSEGV in magazine_chain_pop_head()

[Cristian Aravena Romero]

I'm working with Ubuntu and Crash.

ProblemType: Crash
DistroRelease: Ubuntu 13.10
Package: tracker-miner-fs 0.16.1-1ubuntu2
ProcVersionSignature: Ubuntu 3.11.0-4.9-generic 3.11.0-rc7
Uname: Linux 3.11.0-4-generic x86_64
ApportVersion: 2.12.1-0ubuntu3
Architecture: amd64
Date: Sun Sep 1 20:34:28 2013
ExecutablePath: /usr/lib/tracker/tracker-miner-fs
InstallationDate: Installed on 2013-07-31 (32 days ago)
InstallationMedia: Ubuntu-GNOME 13.10 "Saucy Salamander" - Alpha amd64 (20130724)
MarkForUpload: True
ProcCmdline: /usr/lib/tracker/tracker-miner-fs
ProcEnviron:
 LANGUAGE=es_CL:es
 PATH=(custom, no user)
 XDG_RUNTIME_DIR=<set>
 LANG=es_CL.UTF-8
 SHELL=/bin/bash
SegvAnalysis:
 Segfault happened at: 0x7f7c1da89987 <g_slice_alloc+167>: mov (%rbx),%rax
 PC (0x7f7c1da89987) ok
 source "(%rbx)" (0x00bfaad4) not located in a known VMA region (needed readable region)!
 destination "%rax" ok
 Stack memory exhausted (SP below stack segment)
SegvReason: reading unknown VMA
Signal: 11
SourcePackage: tracker
StacktraceTop:
 g_slice_alloc () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
 g_slice_alloc0 () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
 g_type_create_instance () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
 ?? () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
 g_object_newv () from /usr/lib/x86_64-linux-gnu/libgobject-2.0.so.0
Title: tracker-miner-fs crashed with SIGSEGV in g_slice_alloc()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sambashare sudo

[Apport retracing service (apport)]

StacktraceTop:
 magazine_chain_pop_head (magazine_chunks=0x7f7c08000990) at /build/buildd/glib2.0-2.37.6/./glib/gslice.c:545
 thread_memory_magazine1_alloc (tmem=<optimized out>, ix=1) at /build/buildd/glib2.0-2.37.6/./glib/gslice.c:848
 g_slice_alloc (mem_size=mem_size@entry=32) at /build/buildd/glib2.0-2.37.6/./glib/gslice.c:1007
 g_slice_alloc0 (mem_size=mem_size@entry=32) at /build/buildd/glib2.0-2.37.6/./glib/gslice.c:1042
 g_type_create_instance (type=<optimized out>) at /build/buildd/glib2.0-2.37.6/./gobject/gtype.c:1850

[Sasa Paporovic (melchiaros)]

Hi Christian,

I have just lokked in the stacktrace. There is a notice of which file is causing this crash/properly).

I do not past the filename here. This mails here are indexed by websearch engines.

You can find the affected file in the stacktrace

https://bugs.launchpad.net/ubuntu/+source/tracker/+bug/1219606/+attachment/3798516/+files/Stacktrace.txt

search for the line with the following beginning:

#8 0x00007f7c1e271a5a in _g_local_file_new (filename=0x7f7c08020f20

There is the directory path. Well, the normal procedure is to attach the file here for reproducing...

May be you send Martyn Russell(tracker developer at GNOME) a private message about the crash the report who he can find the official informations and which file could be used to recreate(URL and such). May be you whant to set this report back to privat?!

Martyn has a account on launchpad, but I could not say if he react here. Please search for his email in other tracker reports on GNOME bugtracker. You should be able to find it very quickly.

[Sasa Paporovic (melchiaros)]

@Martyn Russell

Hi Martyn,

here is a tracker-miner problem that is based on a very special file.Details in the comment just before. This has made me to subscribe you here directly. Could you handle this with the needed silence?

Thank you

melchiaros

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Fabio Duran Verdugo (fabioduran)]

I get this bug, but the stracktrace is insufficient for this report, is a memory issue (malloc) and we need a valgrind log.

I will try to generate valgrind log.

@Cristian, I think is a bad idea send uptreams without confirm if actually this tracker issue affects to GNOME.
This memory problem (not all), and maybe just only ubuntu affect.

Chears