[digikam] [Bug 339154] New: Digikam goes into infinite loop when working with tags/captions information tab templates

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

            Bug ID: 339154
           Summary: Digikam goes into infinite loop when working with
                    tags/captions information tab templates
           Product: digikam
           Version: 4.3.0
          Platform: openSUSE RPMs
                OS: Linux
            Status: UNCONFIRMED
          Severity: critical
          Priority: NOR
         Component: Tags
          Assignee: [hidden email]
          Reporter: [hidden email]

Created attachment 88731
  --> https://bugs.kde.org/attachment.cgi?id=88731&action=edit
Stack backtrace

Reproducible: Always

Steps to Reproduce:
1. Start Digikam
2. Select a photo in any album
3. Go to the Captions/tags sidebar
4. Go to the information tab
5. Try using a template (that has worked plenty of times in the not too distant
past).
Actual Results:  
CPU utilization of one core goes to 100%.  Digikam is unresponsive.

This problem occurs when "Apply"ing a title to a photo also.  

Expected Results:  
Add the copyright information from the template into the photo metadata.

The attachment has a gdb stack backtrace from attaching to a digikam in the
infinite loop.

Relevant packages installed (from OpenSUSE repositories):

$ rpm -qa | egrep -i 'digikam|qt'
yast2-control-center-qt-3.0.0-2.1.3.x86_64
qt4-qtscript-0.2.0-8.1.3.x86_64
libQtGStreamer-0_10-0-debuginfo-0.10.3-1.1.x86_64
libQtGStreamer-0_10-0-0.10.3-1.1.x86_64
libntrack-qt4-1-016-7.1.3.x86_64
gstreamer-0_10-plugins-qt-0.10.3-1.1.x86_64
libqt4-sql-4.8.5-5.9.2.x86_64
libqt4-qt3support-32bit-4.8.5-5.9.2.x86_64
texlive-qpxqtx-2013.72.svn15878-11.2.1.noarch
libyui-qt5-2.43.5-2.1.4.x86_64
libqt4-sql-mysql-32bit-4.8.5-5.9.2.x86_64
libsvnqt7-1.6.0-5.1.3.x86_64
libqt4-qt3support-4.8.5-5.9.2.x86_64
libmygpo-qt1-1.0.7-2.1.2.x86_64
libpolkit-qt-1-1-0.103.0-2.1.3.x86_64
libQtWebKit4-32bit-4.8.5+2.3.3-2.6.1.x86_64
qt3-3.3.8c-128.1.2.x86_64
libpoppler-qt4-4-0.24.3-12.1.x86_64
python-qt4-devel-4.10.3-3.4.1.x86_64
libQtGLib-2_0-0-0.10.3-1.1.x86_64
libyui-qt-graph5-2.42.7-2.2.1.x86_64
libqt4-x11-32bit-4.8.5-5.9.2.x86_64
libyui-qt-pkg5-2.42.13-2.4.1.x86_64
yast2-qt-branding-openSUSE-13.1-10.4.13.noarch
libqt4-sql-32bit-4.8.5-5.9.2.x86_64
libqt4-4.8.5-5.9.2.x86_64
libqt4-sql-mysql-4.8.5-5.9.2.x86_64
digikam-doc-3.5.0-7.3.noarch
libpackagekit-qt2-6-0.8.8-2.2.2.x86_64
python-qt4-utils-4.10.3-3.4.1.x86_64
libqt4-sql-sqlite-32bit-4.8.5-5.9.2.x86_64
pinentry-qt4-0.8.3-2.1.3.x86_64
libqt4-devel-4.8.5-5.9.2.x86_64
libqt4-32bit-4.8.5-5.9.2.x86_64
digikam-debuginfo-4.3.0-19.1.x86_64
libQtWebKit4-4.8.5+2.3.3-2.6.1.x86_64
appmenu-qt-0.2.6-6.1.2.x86_64
libqt4-linguist-4.8.5-5.9.2.x86_64
texlive-qpxqtx-doc-2013.72.svn15878-11.2.1.noarch
ibus-qt-1.3.2-2.1.2.x86_64
libQtGLib-2_0-0-debuginfo-0.10.3-1.1.x86_64
digikam-4.3.0-19.1.x86_64
libqt4-sql-sqlite-4.8.5-5.9.2.x86_64
python-qt4-4.10.3-3.4.1.x86_64
libdbusmenu-qt2-0.9.2-7.1.3.x86_64
libqt4-x11-4.8.5-5.9.2.x86_64


Problem also exists with
libqt4-4.8.5-5.6.1.x86_64

Problem also exists with Digikam 4.2.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

Gilles Caulier <[hidden email]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[hidden email]
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |BACKTRACE

--- Comment #1 from Gilles Caulier <[hidden email]> ---
There is nothing to see in your backtrace.

Showing debug statements in the console will be more informative about
operation performed in background.

Look how to turn on debug statements in this page at "For hangs out or
dysfunctions" section :

https://www.digikam.org/contrib

Gilles Caulier

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #2 from Kenneth Ingham <[hidden email]> ---
Created attachment 88739
  --> https://bugs.kde.org/attachment.cgi?id=88739&action=edit
Debug statements from Digikam

All Info and warnings for Digikam, core, kio-slave, showfoto, editor plugins,
and database server.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #3 from Kenneth Ingham <[hidden email]> ---
For the attached info and warnings, the following appeared in my terminal
window:
$ digikam
Object::connect: No such signal
org::freedesktop::UPower::DeviceAdded(QDBusObjectPath)
Object::connect: No such signal
org::freedesktop::UPower::DeviceRemoved(QDBusObjectPath)
QSqlDatabasePrivate::removeDatabase: connection 'ConnectionTest' is still in
use, all queries will cease to work.
Killed

I killed Digikam after it was in the infinite loop.

I can also give you a full backtrace rather than just the first page, if that
would be more useful.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #4 from Gilles Caulier <[hidden email]> ---
After this line :

digikam(8733)/digikam (core) Digikam::DNotificationWrapper: Event is dispatched
to KDE desktop notifier

... there is nothing. If you try to reproduce the problem (the infinite loop
stuff), something must be printed on the console, especially if tags are
changed...

Sound like you have started digiKam and that all...

Did you enable KExiv2 and KDcraw debug spaces ?

Gilles Caulier

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #5 from Gilles Caulier <[hidden email]> ---
Object::connect: No such signal
org::freedesktop::UPower::DeviceAdded(QDBusObjectPath)
Object::connect: No such signal
org::freedesktop::UPower::DeviceRemoved(QDBusObjectPath)
QSqlDatabasePrivate::removeDatabase: connection 'ConnectionTest' is still in
use, all queries will cease to work.
Killed

==> This messages are not the origin of the problem. This just warnings from
Qt.

2 first one are about dbus. The last one is fixed in current implementation

Gilles Caulier

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

Kenneth Ingham <[hidden email]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #88739|0                           |1
        is obsolete|                            |

--- Comment #6 from Kenneth Ingham <[hidden email]> ---
Created attachment 88741
  --> https://bugs.kde.org/attachment.cgi?id=88741&action=edit
Debug statements from Digikam with KExiv2 and KDcraw also

I replaced the old debug statements.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #7 from Gilles Caulier <[hidden email]> ---
Well there is nothing special to see...

Just to be sure.

- create a new temp account
- start digiKam with this account and create a new empty database in this
account
- add some new images in collection
- try to repeat the dysfunction to attribute Tags

infinite loop is reproducible ?

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

Kenneth Ingham <[hidden email]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |VERIFIED

--- Comment #8 from Kenneth Ingham <[hidden email]> ---
I did this and Digikam worked normally.

How (short of deleting my database and starting over) can we track this next?

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

Gilles Caulier <[hidden email]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|BACKTRACE                   |---
             Status|VERIFIED                    |UNCONFIRMED

--- Comment #9 from Gilles Caulier <[hidden email]> ---
So, i suspect a problem with your SQLite database file (digikam4.db).

2 solutions :

- if you have a backup of this file restore it.
- if you store metadata to image or sidecar, drop DB file and rescan your
collection. DB will be recreated using metadata.

Q: do you store your DB file in local file system ?

Note : it miss a tool to maintain/consolidate  database. There is already a
report about this subject.

Gilles Caulier

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

Gilles Caulier <[hidden email]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Digikam goes into infinite  |digiKam goes into infinite
                   |loop when working with      |loop when working with
                   |tags/captions information   |tags/captions information
                   |tab templates               |tab templates
          Component|Tags                        |Database

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #10 from Kenneth Ingham <[hidden email]> ---
I do store the DB on a local filesystem.  The only unexpected shutdowns for
Digikam are if it becomes unresponsive, which, till recently, has been rare.

Unfortunately, my backups rotate weekly, and this problem has been plaguing me
for over a week now.

I moved the db files and it has been rebuilding for a while.  I have 1.5TB of
photos, so this will probably take a bit longer.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #11 from Kenneth Ingham <[hidden email]> ---
I started Digikam rebuilding both thumbnail and metadata databases.  It ran for
a while and then died.  I restarted it, and it again ran for a while and then
died.  Now, when I start it, it immediately dies.  These are the last two lines
in the debug message file:
digikam(17579)/KEXIV2 KExiv2Iface::KExiv2::Private::printExiv2nnot load
metadata using Exiv2   (Error # 11 :  /home/ingham/P.3gp: The file contains
data of an unknown image type
digikam(17579)/KEXIV2 KExiv2Iface::KExiv2::Private::printExiv2nnot load
metadata using Exiv2   (Error # 11 :  /home/ingham/P.3gp: The file contains
data of an unknown image type

I can upload the whole message file if needed.

Oddly, there is no file /home/ingham/P.3gp,   There is a file:
/home/ingham/Photos/HawaiiTurtle.3gp: ISO Media, MPEG v4 system, 3GPP
That is the only file that ends in 3gp.

If I start digikam in gdb, I get the following:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fff367fc700 (LWP 22473)]
0x00007fffed455072 in
Exiv2::PreviewManager::getPreviewImage(Exiv2::PreviewProperties const&) const
() from /usr/lib64/libexiv2.so.12
(gdb) bt
#0  0x00007fffed455072 in
Exiv2::PreviewManager::getPreviewImage(Exiv2::PreviewProperties const&) const
() from /usr/lib64/libexiv2.so.12
#1  0x00007ffff5ca6512 in KExiv2Iface::KExiv2Previews::data(int) ()
   from /usr/lib64/libkexiv2.so.11
#2  0x00007ffff5ca676b in KExiv2Iface::KExiv2Previews::image(int) ()
   from /usr/lib64/libkexiv2.so.11
#3  0x00007ffff529f50d in Digikam::ThumbnailCreator::createThumbnail (
    this=this@entry=0x1000a70, info=..., detailRect=...)
    at
/usr/src/debug/digikam-4.3.0/core/libs/threadimageio/thumbnailcreator.cpp:528
#4  0x00007ffff52a21a6 in Digikam::ThumbnailCreator::load (this=0x1000a70,
    path=..., rect=..., pregenerate=pregenerate@entry=true)
    at
/usr/src/debug/digikam-4.3.0/core/libs/threadimageio/thumbnailcreator.cpp:261
#5  0x00007ffff52a2af0 in Digikam::ThumbnailCreator::pregenerate (
    this=<optimized out>, path=...)
    at
/usr/src/debug/digikam-4.3.0/core/libs/threadimageio/thumbnailcreator.cpp:184
#6  0x00007ffff52b14a0 in Digikam::ThumbnailLoadingTask::execute (
    this=0x6606a10)
    at
/usr/src/debug/digikam-4.3.0/core/libs/threadimageio/thumbnailtask.cpp:82
#7  0x00007ffff52896c6 in Digikam::LoadSaveThread::run (this=0x11a6950)
    at
/usr/src/debug/digikam-4.3.0/core/libs/threadimageio/loadsavethread.cpp:136
#8  0x00007ffff52b76be in Digikam::DynamicThread::DynamicThreadPriv::run (
    this=0x1304ba0)
    at /usr/src/debug/digikam-4.3.0/core/libs/threads/dynamicthread.cpp:186
#9  0x00007ffff14c913e in ?? () from /usr/lib64/libQtCore.so.4
#10 0x00007ffff14d568f in ?? () from /usr/lib64/libQtCore.so.4
#11 0x00007fffee63e0db in start_thread () from /lib64/libpthread.so.0
#12 0x00007ffff06ee58d in clone () from /lib64/libc.so.6
(gdb)

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #12 from Gilles Caulier <[hidden email]> ---
See this file about your crash :

https://bugs.kde.org/show_bug.cgi?id=339144

Gilles Caulier

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #13 from Kenneth Ingham <[hidden email]> ---
Based on that bug, I went back to Digikam 4.2, which does start up without
crashing.

The infinite loop problem appears to still be there when I try to add metadata
to a photo such as copyright information.  I removed the previous metadata and
thumbnail databases, just in case they were corrupt, and I am rebuilding the
databases now.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #14 from Kenneth Ingham <[hidden email]> ---
4.2 dies the same way as 4.3.  Unless there is another suggestion about how to
resolve this, I am about to start the download and build from source process to
see if the problem still occurs in the current version.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #15 from Kenneth Ingham <[hidden email]> ---
Created attachment 88888
  --> https://bugs.kde.org/attachment.cgi?id=88888&action=edit
Digikam log info trace

Debug trace from OpenSUSE digikam digikam-4.3.0-24.1.x86_64

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #16 from Kenneth Ingham <[hidden email]> ---
Digikam is still going into the infinite loop when I try to tag a photo.  I
just uploaded the debug log from a test start this morning.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Kenneth Ingham]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #17 from Kenneth Ingham <[hidden email]> ---
The infinite loop also occurs when trying to search.

How can I get more useful information for tracking this down?  It makes Digikam
unusable.

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel

[Gilles Caulier-4]

https://bugs.kde.org/show_bug.cgi?id=339154

--- Comment #18 from Gilles Caulier <[hidden email]> ---
There is nothing special to see in your backtrace. All sound fine for me.

If i'm not too wrong, it look like you turn on Color Management in digiKam. Try
to turn off this feature to see if this fix the problem.

Gilles Caulier

--
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Digikam-devel mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-devel