tracker-store crashed with signal 5

Bug #1919255 reported by Joshua Peisach
52
This bug affects 8 people
Affects Status Importance Assigned to Milestone
tracker (Ubuntu)
Confirmed
Medium
Unassigned

Bug Description

This just happened...

ProblemType: Crash
DistroRelease: Ubuntu 21.04
Package: tracker 2.3.6-2
ProcVersionSignature: Ubuntu 5.10.0-14.15-generic 5.10.11
Uname: Linux 5.10.0-14-generic x86_64
ApportVersion: 2.20.11-0ubuntu60
Architecture: amd64
CasperMD5CheckResult: unknown
CrashCounter: 1
Date: Mon Mar 15 17:47:12 2021
ExecutablePath: /usr/libexec/tracker-store
InstallationDate: Installed on 2021-02-28 (15 days ago)
InstallationMedia: ubuntucinnamonremix "@BASECODENAME" (20210227)
ProcCmdline: /usr/libexec/tracker-store
Signal: 5
SourcePackage: tracker
StacktraceTop:
 () at /usr/lib/x86_64-linux-gnu/tracker-2.0/libtracker-data.so
 tracker_db_interface_sqlite_fts_delete_id () at /usr/lib/x86_64-linux-gnu/tracker-2.0/libtracker-data.so
 () at /usr/lib/x86_64-linux-gnu/tracker-2.0/libtracker-data.so
 () at /usr/lib/x86_64-linux-gnu/tracker-2.0/libtracker-data.so
 () at /usr/lib/x86_64-linux-gnu/tracker-2.0/libtracker-data.so
Title: tracker-store crashed with signal 5
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups: adm cdrom dip lpadmin plugdev sudo
separator:

Revision history for this message
Joshua Peisach (itzswirlz) wrote :
information type: Private → Public
Revision history for this message
Apport retracing service (apport) wrote :

StacktraceTop:
 execute_stmt (interface=0x55ebc8234250, stmt=<optimized out>, error=error@entry=0x7f5650d26370, cancellable=0x0) at ../src/libtracker-data/tracker-db-interface-sqlite.c:2348
 tracker_db_statement_execute (stmt=stmt@entry=0x7f563c1ed9c0, error=error@entry=0x7f5650d26370) at ../src/libtracker-data/tracker-db-interface-sqlite.c:3018
 tracker_db_interface_sqlite_fts_delete_id (db_interface=db_interface@entry=0x55ebc8234250, id=102000) at ../src/libtracker-data/tracker-db-interface-sqlite.c:1891
 cache_delete_resource_type_full (data=0x55ebc823c000, class=0x55ebc84980d0, graph=graph@entry=0x0, graph_id=graph_id@entry=0, single_type=single_type@entry=0) at ../src/libtracker-data/tracker-data-update.c:2246
 cache_delete_resource_type (graph_id=0, graph=0x0, class=<optimized out>, data=<optimized out>) at ../src/libtracker-data/tracker-data-update.c:2443

Revision history for this message
Apport retracing service (apport) wrote : Stacktrace.txt
Revision history for this message
Apport retracing service (apport) wrote : StacktraceSource.txt
Revision history for this message
Apport retracing service (apport) wrote : ThreadStacktrace.txt
Changed in tracker (Ubuntu):
importance: Undecided → Medium
tags: removed: need-amd64-retrace
Revision history for this message
Launchpad Janitor (janitor) wrote :

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

Changed in tracker (Ubuntu):
status: New → Confirmed
Revision history for this message
Chris Hermansen (c-hermansen) wrote :

FWIW I get this bug in Hirsute every time I boot up my machine.

Revision history for this message
Chris Hermansen (c-hermansen) wrote :

More on this problem: as an attempted workaround, I have tried removing .cache/tracker which does not seem to solve it.

I have considered removing the files in ~/.local/share/tracker/data but haven't yet tried that.

description: updated
tags: added: tracker
Revision history for this message
Merlin (merlin-skinner) wrote :

I saw this for a few days after a fresh install of Hirsute. It would happen shortly after boot, but it seems to have stopped. It could be because tracker has now scanned everything.

Revision history for this message
Chris Hermansen (c-hermansen) wrote : Re: [Bug 1919255] Re: tracker-store crashed with signal 5

Mine still happens every day about 5 minutes after starting up

On Tue, May 18, 2021, 15:10 Merlin <email address hidden> wrote:

> I saw this for a few days after a fresh install of Hirsute. It would
> happen shortly after boot, but it seems to have stopped. It could be
> because tracker has now scanned everything.
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/1919255
>
> Title:
> tracker-store crashed with signal 5
>
> To manage notifications about this bug go to:
>
> https://bugs.launchpad.net/ubuntu/+source/tracker/+bug/1919255/+subscriptions
>

Revision history for this message
Almer S. Tigelaar (almer) wrote :

Same here on a fresh Hirsute install.

There's a couple of suspect things in my syslog. Here is what happens right before the crash (anonymized):

tracker-extract[2439]: bad character 'M'
tracker-extract[2439]: bad character 'O'
tracker-extract[2439]: bad character 'D'
tracker-extract[2439]: bad character 'E'
tracker-extract[2439]: bad character 'x'
tracker-extract[2439]: bad character '/'
tracker-extract[2439]: 59975: syntax error
tracker-extract[2439]: bad character 'x'
tracker-extract[2439]: message repeated 2 times: [ bad character 'x']
tracker-extract[2439]: libpng warning: iCCP: known incorrect sRGB profile
tracker-extract[2439]: libpng warning: iCCP: known incorrect sRGB profile
tracker-extract[2439]: message repeated 2 times: [ libpng warning: iCCP: known incorrect sRGB profile]
tracker-store[2522]: SQLite error: database disk image is malformed (errno: Success)
kernel: [ 408.547231] traps: pool-tracker-st[2527] trap int3 ip:7fd050c8d3a7 sp:7fd046f9c190 error:0 in libglib-2.0.so.0.6800.0[7fd050c4e000+8d000]
tracker-store[2522]: SQLite experienced an error with file:'/home/[user]/.cache/tracker/meta.db'. It is either NOT a SQLite database or it is corrupt or there was an IO error accessing the data. This file has now been removed and will be recreated on the next start. Shutting down now.
systemd[2429]: Starting Notification regarding a crash report...
update-notifier-crash[6325]: /usr/bin/whoopsie
systemd[2429]: update-notifier-crash.service: Succeeded.
systemd[2429]: Finished Notification regarding a crash report.
systemd[2429]: tracker-store.service: Main process exited, code=killed, status=5/TRAP
systemd[2429]: tracker-store.service: Failed with result 'signal'.

So this points either to some sort of malformed input (see the bad characters messages), something with libpng and/or something with sqlite. Hope this helps.

Revision history for this message
Merlin (merlin-skinner) wrote :
Download full text (4.2 KiB)

Please ignore my earlier comment. This does still occur, I just didn't see it has I had inadvertently set apport to automatically report so there was no pop-up. After boot, tracker seems to run normally, at least initially:

merlin@Minmus:~$ tracker status
Currently indexed: 26384 files, 2120 folders
Remaining space on database partition: 735.6 GB (83.31%)
Data is still being indexed: Estimated 02m 44s left

After the crash, I see:

merlin@Minmus:~$ tracker status
Currently indexed: 0 files, 0 folders
Remaining space on database partition: 735.4 GB (83.29%)
Data is still being indexed: Estimated 36s left

Deleting the database at ~/.cache/tracker/meta.db does not fix the problem. My theory is that tracker encounters something in my files that causes a crash.

Following Almer's lead I looked at /var/log/syslog:

May 22 15:37:55 Minmus dbus-daemon[1359]: [session uid=1000 pid=1359] Activating via systemd: service name='org.freedesktop.Tracker1.Miner.Extract' unit='tracker-extract.service' requested by ':1.1' (uid=1000 pid=1356 comm="/usr/libexec/tracker-miner-fs " label="unconfined")
May 22 15:37:55 Minmus systemd[1345]: Starting Tracker metadata extractor...
May 22 15:37:55 Minmus tracker-extract[3168]: Set scheduler policy to SCHED_IDLE
May 22 15:37:55 Minmus tracker-extract[3168]: Setting priority nice level to 19
May 22 15:37:55 Minmus dbus-daemon[1359]: [session uid=1000 pid=1359] Successfully activated service 'org.freedesktop.Tracker1.Miner.Extract'
May 22 15:37:55 Minmus systemd[1345]: Started Tracker metadata extractor.
May 22 15:38:00 Minmus tracker-extract[3168]: Unknown desktop entry type 'Link'
May 22 15:38:16 Minmus systemd[1345]: Started Application launched by gnome-session-binary.
May 22 15:38:20 Minmus tracker-extract[3168]: Duplicate property or field node
May 22 15:38:20 Minmus tracker-extract[3168]: Duplicate property or field node
May 22 15:38:21 Minmus tracker-extract[3168]: Unknown desktop entry type 'Link'
May 22 15:38:21 Minmus tracker-extract[3168]: Unknown desktop entry type 'Link'
May 22 15:38:21 Minmus tracker-extract[3168]: Duplicate property or field node
May 22 15:38:21 Minmus tracker-extract[3168]: Duplicate property or field node
May 22 15:38:22 Minmus dbus-daemon[1359]: apparmor="DENIED" operation="dbus_method_call" bus="session" path="/org/gtk/Notifications" interface="org.gtk.Notifications" member="RemoveNotification" mask="send" name="org.gtk.Notifications" pid=1765 label="snap.snap-store.ubuntu-software" peer_pid=1513 peer_label="unconfined"
May 22 15:38:22 Minmus systemd[1]: systemd-timedated.service: Succeeded.
May 22 15:38:27 Minmus tracker-extract[3168]: Call to gst_discoverer_discover_uri(file:///home/ <redacted> /b00035.au) failed: The stream is in the wrong format.
May 22 15:38:28 Minmus tracker-extract[3168]: libpng warning: iCCP: known incorrect sRGB profile
May 22 15:39:16 Minmus systemd[1345]: Started Application launched by gnome-session-binary.
May 22 15:41:07 Minmus tracker-extract[3168]: Duplicate property or field node
May 22 15:41:16 Minmus tracker-extract[3168]: message repeated 6 times: [ Duplicate property or field node]
May 22 15:41:16 Minmus systemd[1345]: app-gnome-org.gnome.DejaDu...

Read more...

Revision history for this message
Merlin (merlin-skinner) wrote :

Tracker issues seem to rate pretty highly here: https://errors.ubuntu.com/?release=Ubuntu%2021.04&period=year

Revision history for this message
Matti Kerttula (mkert) wrote :

There is a fix in upstream (tracker 3.0.3) which might apply to this: https://gitlab.gnome.org/GNOME/tracker/-/merge_requests/353

Revision history for this message
Chris Hermansen (c-hermansen) wrote :

This week I moved the contents of my ~/Music directory to a different directory on a different physical drive mounted on /work, and I no longer get this behaviour.

From that I surmise, perhaps incorrectly, that there was some situation in my music files (almost all FLAC, a few MP3s) that was causing this problem.

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.