[karmic] in KDE apport does not open the browser to report a bug

Bug #405378 reported by Katsudon
82
This bug affects 12 people
Affects Status Importance Assigned to Milestone
apport (Ubuntu)
Fix Released
High
Yuriy Kozlov
Karmic
Fix Released
High
Yuriy Kozlov

Bug Description

Kubuntu Karmic alpha 3.

Apport crash handler opens a window asking if I want to send the report to the developers, I click "yes" then nothing happens, no browser window is opened, etc.

Hint: the outcome of ps aux | grep apport shows that there is quite a lot of processes related to apport running in the memory:

piotr 4807 0.0 1.4 88096 29060 ? S 16:11 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4813 0.0 1.4 88084 28944 ? S 16:11 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4828 0.0 1.4 88100 29064 ? S 16:11 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4834 0.0 1.4 88100 29064 ? S 16:11 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4929 0.0 1.4 88100 29060 ? S 16:12 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4987 0.0 1.4 88096 29060 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4993 0.0 1.4 88100 29064 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 4998 0.0 1.4 88096 29060 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5003 0.0 1.4 88100 29064 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5008 0.0 1.4 88096 29060 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5013 0.0 1.4 88100 29064 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5018 0.0 4.2 159604 86956 ? S 16:15 0:07 /usr/bin/python /usr/share/apport/apport-kde
piotr 5028 0.0 1.4 88092 29056 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5105 0.0 1.4 88100 29064 ? S 16:15 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5140 0.0 3.6 146556 74432 ? S 16:16 0:02 /usr/bin/python /usr/share/apport/apport-kde
piotr 5160 0.0 1.4 88088 28948 ? S 16:16 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5234 0.0 1.4 88096 28948 ? S 16:16 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 5239 0.0 1.4 88096 29060 ? S 16:16 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 7086 0.1 0.8 65744 17028 ? Sl 18:32 0:00 /usr/bin/kdesudo /usr/share/apport/apport-kde
root 7093 0.1 1.3 87468 28452 ? S 18:32 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 7164 0.1 0.8 65748 16912 ? Sl 18:32 0:00 /usr/bin/kdesudo /usr/share/apport/apport-kde
root 7169 0.5 3.0 135696 61480 ? S 18:32 0:02 /usr/bin/python /usr/share/apport/apport-kde
piotr 7195 0.1 1.4 88096 29056 ? S 18:33 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 7272 0.1 1.4 88096 29056 ? S 18:33 0:00 /usr/bin/python /usr/share/apport/apport-kde
piotr 7327 0.0 0.0 3156 816 pts/1 S+ 18:39 0:00 grep apport

This bug seems to be similar to Bug #81207, nevertheless I understand that official policy is to file a new bug if the old one is "fix commited".

Revision history for this message
databubble (phil-linttell) wrote :

I was wondering why my system was starting to swap.... and discovered that I have 112 of these processes running,
each consuming between 1MB and 25MB of memory (over 2GB in total.) No on screen notifications, and nothing in the task bar.

Also running Karmic alpha 3. Not sure what bugs is causing the apports to be generated, but apport itself seems like a bigger bug at the moment.

Revision history for this message
Luka Renko (lure) wrote :

apport-kde as it is today in Karmic is useless, as it is not possible to report a bug and we are loosing bug reports due to this.

I had to install apport-gtk in order to be able to report bugs again. This really needs to be addressed before Karmic beta if we want to collect feedback from our users.

Changed in apport (Ubuntu):
status: New → Confirmed
importance: Undecided → High
Revision history for this message
Katsudon (katsudon) wrote :

I am tempted to close this bug because the bad behavior of apport has changed. The situation described in this bug happens no more, instead I get Bug #417688.

Revision history for this message
Katsudon (katsudon) wrote :

Bug #417688 is gone. Now the situation is back the same as in the description of this bug.

Revision history for this message
Katsudon (katsudon) wrote :

Now bug reporting seems to be perfectly fine. I am closing this bug.

Changed in apport (Ubuntu):
status: Confirmed → Fix Released
Revision history for this message
Katsudon (katsudon) wrote :

Unfortunately, apport in KDE again fails to open the browser.

Changed in apport (Ubuntu):
status: Fix Released → New
Revision history for this message
Matthew Darwin (bugs-mdarwin) wrote :

I am seeing this problem today.

Revision history for this message
Gareth Hart (tghe-retford) wrote :

This problem seems to occur if I use ubuntu-bug in the command line or with the majority of instances where apport catches a crash. However, one time when VLC crashed, it did work, but as I mentioned, in the majority of cases, it fails to open in a browser to report a bug.

Revision history for this message
Yuriy Kozlov (yuriy-kozlov) wrote :

I can confirm this on Karmic -- running ubuntu-bug to start ubuntu-kde and clicking send, apport just hangs.

Changed in apport (Ubuntu):
status: New → Won't Fix
status: Won't Fix → Confirmed
Revision history for this message
Sjors Gielen (sgielen) wrote :

Tried to find out a little more regarding this.

ubuntu-bug just spawns an ubuntu-kde process, as it should; when killing ubuntu-bug ubuntu-kde stays in the system, so that's not the problem.

The problem is that ubuntu-kde, after trying to send, seems to freeze/halt in a poll():

poll([{fd=3, events=POLLIN}, {fd=9, events=POLLIN}, {fd=10, events=POLLIN}, {fd=8, events=POLLIN}], 4, 18779) = 1 ([{fd=8, revents=POLLIN}])
read(8, "\34\257\311 \247\1\0\0\325\1\0\0\237\337\340\0\0\0\0\0\0\0\0\0P\317)\1\0\0\0\0"..., 4096) = 64
read(8, 0x29f25e4, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(8, 0x29f25e4, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(8, 0x29f25e4, 4096) = -1 EAGAIN (Resource temporarily unavailable)
poll([{fd=3, events=POLLIN}, {fd=9, events=POLLIN}, {fd=10, events=POLLIN}, {fd=8, events=POLLIN}], 4, 18726) = 1 ([{fd=8, revents=POLLIN}])
read(8, "\34K\311 \247\1\0\0\333\1\0\0\271\337\340\0\0\320)\1\0\0\0\0P\317)\1\0\0\0\0"..., 4096) = 64
read(8, 0x29f25e4, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(8, 0x29f25e4, 4096) = -1 EAGAIN (Resource temporarily unavailable)
read(8, 0x29f25e4, 4096) = -1 EAGAIN (Resource temporarily unavailable)
poll([{fd=3, events=POLLIN}, {fd=9, events=POLLIN}, {fd=10, events=POLLIN}, {fd=8, events=POLLIN}], 4, 18699^C <unfinished ...>
Process 2869 detached

This last poll() never finishes, probably because it waits for GUI input when all windows are gone. During this, apport-bug just stays around wait()ing for the apport-kde process.

This happens after apport-kde just sent the bug report; the next step would be opening the browser, which apport-cli perfectly succeeds to do.

Revision history for this message
Yuriy Kozlov (yuriy-kozlov) wrote :

Do you get the sending progress dialog? The problem I am seeing is that it never gets that far. In part this is because an incorrect result is returned from apport-kde when submitting a non-crash bug -- fixing that makes things work (patch attached) but it shouldn't be hanging as is anyway.

Revision history for this message
Daniel Hahler (blueyed) wrote :

I can confirm this.

To reproduce:
/usr/share/apport/apport-kde apport-kde

Nothing happens after clicking "Send", only the process stays around:
user 21714 8.1 2.0 138584 64264 pts/2 S+ 17:28 0:03 /usr/bin/python /usr/share/apport/apport-kde apport-kde

Changed in apport (Ubuntu):
status: Confirmed → Triaged
Changed in apport (Ubuntu):
assignee: nobody → Yuriy Kozlov (yuriy-kozlov)
milestone: none → karmic-alpha-6
status: Triaged → In Progress
Revision history for this message
Yuriy Kozlov (yuriy-kozlov) wrote :

I think I have a fix for this, but now I'm running into bug 403361.

Martin Pitt (pitti)
tags: added: regression-potential
Revision history for this message
Martin Pitt (pitti) wrote :

Committed Yuriy's fix to apport trunk r1608, thank you! Works fine here.

Changed in apport (Ubuntu):
status: In Progress → Fix Committed
Changed in apport (Ubuntu Karmic):
milestone: karmic-alpha-6 → ubuntu-9.10-beta
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package apport - 1.9.1-0ubuntu2

---------------
apport (1.9.1-0ubuntu2) karmic; urgency=low

  * kde/bugreport.ui: Select "complete report" option by default, to actually
    be able to file bugs if the options are not shown. Thanks to Yuriy Kozlov!
    Fix cherrypicked from trunk. (LP: #405378)

 -- Martin Pitt <email address hidden> Wed, 23 Sep 2009 10:18:28 +0200

Changed in apport (Ubuntu Karmic):
status: Fix Committed → Fix Released
Revision history for this message
Luka Renko (lure) wrote :

Just submitted my first bug with apport-kde and it works! Yuriy, thanks for fixing this important bug!

Revision history for this message
Santiago (santiago-osella) wrote :

I have succeeded in reporting bugs with apport (with Karmic beta) but the process itself keeps spawning!
Also tried killing all instances but one, but some seconds later it started running again (several times).
Any clues?

Revision history for this message
Yuriy Kozlov (yuriy-kozlov) wrote :

The hang is not fixed, I'm a bit stumped on how to make it neither hang or crash without rewriting a lot and possibly changing the UI. So that's why you get the leftover process. Why it respawns when you kill it, I don't know. Apport gets disabled post-release so it shouldn't be *too* big a deal and we'll fix for Lucid.

Revision history for this message
LucidStrike (lucidstrike) wrote :

Please do. :D

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

Duplicates of this bug

Other bug subscribers

Remote bug watches

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