Bug 754585

Created attachment 534105 [details]
File: maps

Created attachment 534106 [details]
File: backtrace

Does it crash every time your computer wakes up from sleep?

No, it was the first time. I didn't retry to reproduce this bug since it happens.

Package: xfce4-power-manager-1.0.10-2.fc16
Architecture: x86_64
OS Release: Fedora release 16 (Verne)

Comment
-----
When i plugged in the power plug

Seems you are the only one to get this crash. Can you tell us a little about your hardware or even better give us a link to it's smolt profile?

Created attachment 536037 [details]
Smolt profile

*** Bug 758379 has been marked as a duplicate of this bug. ***

*** Bug 758572 has been marked as a duplicate of this bug. ***

Package: xfce4-power-manager-1.0.10-2.fc16
Architecture: x86_64
OS Release: Fedora release 16 (Verne)

Comment
-----
When unplugging the power plug

*** Bug 759464 has been marked as a duplicate of this bug. ***

*** Bug 765762 has been marked as a duplicate of this bug. ***

*** Bug 766496 has been marked as a duplicate of this bug. ***

Plug / Unplug power supply on netbook.

rating: 4
Package: xfce4-power-manager-1.0.10-2.fc16
Architecture: i686
OS Release: Fedora release 16 (Verne)

I have just do it - power manager crashed. If it significant the netbook battery is being charged now.

Seems like the problem is coming directly from xfce4 and not Fedora itself.

Maybe we should submit a ticket about this to xfce4 dev team.

Under Xubuntu (using xfce4), we have the same problem as follow :

https://bugs.launchpad.net/ubuntu/+source/xfce4-power-manager/+bug/910776

B.D.

Plug and unplug AC cable. Wakeup after going to suspen and hibernate mode.

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-2.fc16
OS Release: Fedora release 16 (Verne)

Created attachment 564756 [details]
File: backtrace

It just crashed after suspending.

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-2.fc16
Architecture: i686
OS Release: Fedora release 16 (Verne)

Created attachment 573025 [details]
File: backtrace

Just log in

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
OS Release: Fedora release 17 (Beefy Miracle)

changing battery to AC

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

just power on after standby

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

battery to AC change

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

switch to battery

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

switch to AC

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

plugged AC then crash

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

Back from standby mode. 

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
OS Release: Fedora release 17 (Beefy Miracle)

I plugged the power cord, in fact, each time I plug it, I get this error

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

wake up from hibernate in battery mode

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
Architecture: i686
OS Release: Fedora release 17 (Beefy Miracle)

Unplugged AC

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
OS Release: Fedora release 17 (Beefy Miracle)

Not sure what I did, but here's the stack trace.  This is with
a completely up to date Fedora 18 system.

xfce4-power-manager-1.2.0-2.fc18.x86_64

(gdb) t a a bt

Thread 2 (Thread 0x7f90a008c700 (LWP 2825)):
#0  0x0000003a834e96ad in poll () at ../sysdeps/unix/syscall-template.S:81
#1  0x0000003a86448074 in g_main_context_iterate.isra.24 ()
   from /lib64/libglib-2.0.so.0
#2  0x0000003a864484d2 in g_main_loop_run () from /lib64/libglib-2.0.so.0
#3  0x0000003a8a0d5ed6 in gdbus_shared_thread_func ()
   from /lib64/libgio-2.0.so.0
#4  0x0000003a8646b0c5 in g_thread_proxy () from /lib64/libglib-2.0.so.0
#5  0x0000003a83807d15 in start_thread (arg=0x7f90a008c700)
    at pthread_create.c:308
#6  0x0000003a834f218d in clone ()
    at ../sysdeps/unix/sysv/linux/x86_64/clone.S:114

Thread 1 (Thread 0x7f90a66cc9c0 (LWP 2811)):
#0  0x0000003a83447ea8 in _IO_vfprintf_internal (s=s@entry=0x7fff4d478050, 
    format=<optimized out>, format@entry=
    0x7f90a6714874 <Address 0x7f90a6714874 out of bounds>, ap=ap@entry=
    0x7fff4d4781b8) at vfprintf.c:1615
#1  0x0000003a83509e01 in __GI___vasprintf_chk (result_ptr=0x7fff4d478198, 
    flags=1, format=0x7f90a6714874 <Address 0x7f90a6714874 out of bounds>, 
    args=0x7fff4d4781b8) at vasprintf_chk.c:66
#2  0x0000003a86483e2b in g_vasprintf () from /lib64/libglib-2.0.so.0
#3  0x0000003a86463c2d in g_strdup_vprintf () from /lib64/libglib-2.0.so.0
#4  0x0000003a86463ccc in g_strdup_printf () from /lib64/libglib-2.0.so.0
#5  0x000000000040dfc0 in xfpm_battery_notify ()
#6  0x000000000040e67f in xfpm_battery_notify_idle ()
#7  0x0000003a86447da5 in g_main_context_dispatch ()
   from /lib64/libglib-2.0.so.0
#8  0x0000003a864480d8 in g_main_context_iterate.isra.24 ()
   from /lib64/libglib-2.0.so.0
#9  0x0000003a864484d2 in g_main_loop_run () from /lib64/libglib-2.0.so.0
#10 0x0000003a9814ac87 in gtk_main () from /lib64/libgtk-x11-2.0.so.0
#11 0x00000000004086ee in xfpm_start ()
#12 0x0000000000408c76 in main ()

Looking at the code, I don't even see where that function calls
{g_,}vasprintf.  Must be in a glib macro ...

Here's a better stack trace with more debuginfo packages
installed.  This also confirms that it's happening in the
same place as for the original bug reporter.

#0  0x0000003a83447ea8 in _IO_vfprintf_internal (s=s@entry=0x7fff6bdc04f0, 
    format=<optimized out>, format@entry=
    0x7f7408b538ad <Address 0x7f7408b538ad out of bounds>, ap=ap@entry=
    0x7fff6bdc0658) at vfprintf.c:1615
#1  0x0000003a83509e01 in __GI___vasprintf_chk (result_ptr=0x7fff6bdc0638, 
    flags=1, format=0x7f7408b538ad <Address 0x7f7408b538ad out of bounds>, 
    args=0x7fff6bdc0658) at vasprintf_chk.c:66
#2  0x0000003a86483e2b in g_vasprintf () from /lib64/libglib-2.0.so.0
#3  0x0000003a86463c2d in g_strdup_vprintf () from /lib64/libglib-2.0.so.0
#4  0x0000003a86463ccc in g_strdup_printf () from /lib64/libglib-2.0.so.0
#5  0x000000000040dfc0 in xfpm_battery_get_message_from_battery_state (battery=
    0x12c1830) at xfpm-battery.c:244
#6  xfpm_battery_notify (battery=0x12c1830) at xfpm-battery.c:345
#7  0x000000000040e67f in xfpm_battery_notify_idle (data=<optimized out>)
    at xfpm-battery.c:367
#8  0x0000003a86447da5 in g_main_context_dispatch ()
   from /lib64/libglib-2.0.so.0
#9  0x0000003a864480d8 in g_main_context_iterate.isra.24 ()
   from /lib64/libglib-2.0.so.0
#10 0x0000003a864484d2 in g_main_loop_run () from /lib64/libglib-2.0.so.0
#11 0x0000003a9814ac87 in IA__gtk_main () at gtkmain.c:1257
#12 0x00000000004086ee in xfpm_start (bus=bus@entry=0x1293c08, client_id=0x0, 
    dump=<optimized out>) at xfpm-main.c:238
#13 0x0000000000408c76 in main (argc=1, argv=0x7fff6bdc0c38) at xfpm-main.c:426


The code which is failing is:

                    if ( battery->priv->time_to_empty != 0 )
                    {
                        gchar *tmp, *est_time_str;
                        tmp = g_strdup (msg);
                        g_free (msg);
                        
                        est_time_str = xfpm_battery_get_time_string (battery->pr
iv->time_to_empty);
                        
        HERE>>>         msg = g_strdup_printf (_("%s (%i%%)\nEstimated time left
 is %s."), tmp, battery->priv->percentage, est_time_str);
                        g_free (tmp);
                        g_free (est_time_str);
                    }

Unfortunately 'tmp' and 'est_time_str' are <optimized out> and
*battery looks OK.  I will compile a local copy with -g -O0 and
see if I can reproduce the bug from that.

Without optimization it still crashes.  Here is some more
data from gdb:

(gdb) frame 5
#5  0x000000000040f919 in xfpm_battery_get_message_from_battery_state (battery=
    0x698030) at xfpm-battery.c:244
244				msg = g_strdup_printf (_("%s (%i%%)\nEstimated time left is %s."), tmp, battery->priv->percentage, est_time_str);
(gdb) print tmp
$1 = (gchar *) 0x7c3ec0 "System is running on battery power"
(gdb) print est_time_str
$2 = (gchar *) 0x6e6520 "1 hour 3 minutes"
(gdb) print *battery
$3 = {
  parent = {
    parent_instance = {
      g_type_instance = {
        g_class = 0x6972f0
      }, 
      ref_count = 1, 
      qdata = 0x0
    }, 
    priv = 0x698060
  }, 
  priv = 0x6980a0
}
(gdb) print *battery->priv
$5 = {
  conf = 0x68a700, 
  notify = 0x6515e0, 
  button = 0x673740, 
  proxy = 0x67c070, 
  proxy_prop = 0x7fffec003510, 
  icon_prefix = 0x6d0830 "xfpm-primary-", 
  charge = XFPM_BATTERY_CHARGE_OK, 
  state = XFPM_DEVICE_STATE_DISCHARGING, 
  type = XFPM_DEVICE_TYPE_BATTERY, 
  ac_online = 0, 
  present = 1, 
  percentage = 64, 
  time_to_full = 0, 
  time_to_empty = 3768, 
  battery_name = 0x7ffff7ffc94c "battery", 
  sig = 34, 
  sig_bt = 35, 
  notify_idle = 89
}

I must be missing the bleedin' obvious here, but why would that crash ...?

Plugged in the powersupply of my laptop.
Reproducible crash every time I go from battery to power or back.


backtrace_rating: 4
Package: xfce4-power-manager-1.2.0-2.fc18
OS Release: Fedora release 18 (Spherical Cow)

In the Ubuntu forums it is suggested to do 

rm -rf ~/.cache/sessions

while xfce is not running.

I did not work for me though...

When i'm pluging my ac powermanagement stops working. Thank you

backtrace_rating: 4
Package: xfce4-power-manager-1.0.10-3.fc17
OS Release: Fedora release 17 (Beefy Miracle)

Found the solution (in my case at least) at the xfce bugzilla.

It has to do with translation and language packages.
Set your language to en_US (was en_UK in my case).
(yum install system-config-language)
and reboot.


Can someone at redhat check that Christoph Wickert has not evaporated?

I haven't, I'm still alive and kicking.

However there is not much I can do here. I am using xfce4-power-manager heavily and I never hit that bug. That's why I am very surprised about the workaround you describe. Where do you have it from?

see 

https://bugzilla.xfce.org/show_bug.cgi?id=9404

sorry , that should have been

https://bugzilla.xfce.org/show_bug.cgi?id=8992

FYI, I manually patched the xfce4-power-manager-1.2.0-2.fc18.src.rpm using with the changes in https://bugzilla.xfce.org/attachment.cgi?id=4639 (from https://bugzilla.xfce.org/show_bug.cgi?id=8992) and after force-installing it (didn't make a patch or change the changelog) and starting xfce4-power-manager, plugging/unplugging the power no longer crashes it. My locale is en_GB.utf8

Seems like a pretty trivial fix to port to the Fedora package, and this must be frustrating other UK Xfce users.

Thanks, I will look at this ASAP.

I left the computer for a while and I think the screen switched off to save power, and I couldn't wake it up again.

backtrace_rating: 4
Package: xfce4-power-manager-1.2.0-2.fc18
Architecture: armv7l
OS Release: Fedora release 18 (Spherical Cow)

Still not fixed in xfce4-power-manager-1.2.0-5.fc18

Can we please get this fixed, there is a known, upstream, patch which is completely trivial. I have manually applied the patch again in 1.2.0-5 and confirmed it fixes the issue.

Sorry, it took so long. It affected more locales than just en_GB and and not all patches wee applying cleanly. Plus, I was just terribly busy. Update is currently building, stay tuned.

(In reply to comment #47)
> Sorry, it took so long. It affected more locales than just en_GB and and not
> all patches wee applying cleanly. Plus, I was just terribly busy. Update is
> currently building, stay tuned.

Thanks for the update. I had not realised other locales were affected, obviously not seeing the bigger picture.

xfce4-power-manager-1.2.0-7.fc18 has been submitted as an update for Fedora 18.
https://admin.fedoraproject.org/updates/xfce4-power-manager-1.2.0-7.fc18

xfce4-power-manager-1.0.10-4.fc17 has been submitted as an update for Fedora 17.
https://admin.fedoraproject.org/updates/xfce4-power-manager-1.0.10-4.fc17

Package xfce4-power-manager-1.2.0-7.fc18:
* should fix your issue,
* was pushed to the Fedora 18 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing xfce4-power-manager-1.2.0-7.fc18'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2013-4203/xfce4-power-manager-1.2.0-7.fc18
then log in and leave karma (feedback).

xfce4-power-manager-1.2.0-7.fc18 has been pushed to the Fedora 18 stable repository.  If problems still persist, please make note of it in this bug report.

xfce4-power-manager-1.0.10-4.fc17 has been pushed to the Fedora 17 stable repository.  If problems still persist, please make note of it in this bug report.