Bug 757641

Summary: [abrt] tumbler-0.1.22-4.fc16: g_type_check_instance_is_a: Process /usr/lib64/tumbler-1/tumblerd was killed by signal 11 (SIGSEGV)
Product: [Fedora] Fedora Reporter: Ralf Corsepius <rc040203>
Component: tumblerAssignee: Christoph Wickert <christoph.wickert>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 16CC: christoph.wickert, kevin, maxamillion
Target Milestone: ---Keywords: Reopened
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
Whiteboard: abrt_hash:b846cf57a2d992493763cd5742b9a260e0d90a71
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-12-04 10:03:29 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
File: build_ids
none
File: dso_list
none
File: backtrace
none
File: maps none

Description Ralf Corsepius 2011-11-28 07:55:59 UTC
libreport version: 2.0.7
abrt_version:   2.0.6
backtrace_rating: 4
cmdline:        /usr/lib64/tumbler-1/tumblerd
crash_function: g_type_check_instance_is_a
executable:     /usr/lib64/tumbler-1/tumblerd
kernel:         3.1.2-1.fc16.x86_64
pid:            13378
pwd:            /
reason:         Process /usr/lib64/tumbler-1/tumblerd was killed by signal 11 (SIGSEGV)
time:           Mon 28 Nov 2011 08:53:21 AM CET
uid:            8690
username:       corsepiu

backtrace:      Text file, 31772 bytes
build_ids:      Text file, 4100 bytes
dso_list:       Text file, 9462 bytes
maps:           Text file, 41895 bytes

environ:
:SHELL=/bin/bash
:DBUS_STARTER_ADDRESS=unix:abstract=/tmp/dbus-7AdMhBVsGa,guid=f6bfa464bec0a96fc2a3839900000035
:XDG_SESSION_COOKIE=fa8ca0e3ac19b584c263547100000006-1322416118.230815-1063403350
:XDG_RUNTIME_DIR=/run/user/corsepiu
:DISPLAY=:0
:DESKTOP_SESSION=xfce
:WINDOWPATH=1
:PATH=/usr/local/bin:/usr/bin:/bin
:GDMSESSION=xfce
:XDG_VTNR=1
:USERNAME=corsepiu
:XDG_SESSION_ID=2
:DBUS_SESSION_BUS_ADDRESS=unix:abstract=/tmp/dbus-7AdMhBVsGa,guid=f6bfa464bec0a96fc2a3839900000035
:XDG_SEAT=seat0
:XAUTHORITY=/var/run/gdm/auth-for-corsepiu-3qWkBe/database
:USER=corsepiu
:DBUS_STARTER_BUS_TYPE=session
:GNOME_KEYRING_PID=1519
:SHLVL=1
:GDM_LANG=en_US.UTF-8
:PWD=/users/corsepiu
:EDITOR=joe
:'PS1=\\u@\\h:\\w > '
:GNOME_KEYRING_CONTROL=/tmp/keyring-1LyAij
:LANG=en_US.UTF-8
:_=/usr/bin/dbus-launch
:LOGNAME=corsepiu
:HOME=/users/corsepiu

var_log_messages:
:Nov 28 08:53:21 beck kernel: [50736.424505] tumblerd[13379] general protection ip:33706329e0 sp:7ffcbeebcae8 error:0 in libgobject-2.0.so.0.3000.1[3370600000+4c000]
:Nov 28 08:53:21 beck abrt[13386]: Saved core dump of pid 13378 (/usr/lib64/tumbler-1/tumblerd) to /var/spool/abrt/ccpp-2011-11-28-08:53:21-13378 (83181568 bytes)

Comment 1 Ralf Corsepius 2011-11-28 07:56:02 UTC
Created attachment 537378 [details]
File: build_ids

Comment 2 Ralf Corsepius 2011-11-28 07:56:04 UTC
Created attachment 537379 [details]
File: dso_list

Comment 3 Ralf Corsepius 2011-11-28 07:56:06 UTC
Created attachment 537380 [details]
File: backtrace

Comment 4 Ralf Corsepius 2011-11-28 07:56:08 UTC
Created attachment 537381 [details]
File: maps

Comment 5 Christoph Wickert 2011-11-28 09:24:31 UTC
Please always include a comment in the bug report, even if it's just that you don't know how to reproduce the crash.

*** This bug has been marked as a duplicate of bug 754961 ***

Comment 6 Ralf Corsepius 2011-11-28 10:42:37 UTC
(In reply to comment #5)
> Please always include a comment in the bug report, even if it's just that you
> don't know how to reproduce the crash.

Then talk to RH's abrt guys. 

> *** This bug has been marked as a duplicate of bug 754961 ***

Reopening, because abrt did not detect this as a duplicate.

Comment 7 Christoph Wickert 2011-11-28 11:25:07 UTC
(In reply to comment #6)
> (In reply to comment #5)
> > Please always include a comment in the bug report, even if it's just that you
> > don't know how to reproduce the crash.
> 
> Then talk to RH's abrt guys.

Did that already in bug 702262.

ABRT is just a tool, it doesn't mean that you have to turn off your brain. ABRT even asks you to provide more info and include a comment.

> > *** This bug has been marked as a duplicate of bug 754961 ***
> 
> Reopening, because abrt did not detect this as a duplicate.

Why is that a bug in tumbler then? Did you forget to reassign the bug to the abrt component?

I agree that the automatic duplicate detection could be better, but it doubt that it will ever be perfect and capable to detect all dupes. This is where the maintainer or the reporter should help. I cannot see what is wrong with me closing this bug as a dupe. In fact you could easily have done that as you also filed the original bug just a few days ago.

If you reopen this bug, please state a reason and assign it to the proper component. Thank you.

*** This bug has been marked as a duplicate of bug 754961 ***

Comment 8 Ralf Corsepius 2011-11-28 12:24:07 UTC
(In reply to comment #7)
> (In reply to comment #6)
> > (In reply to comment #5)
> > > Please always include a comment in the bug report, even if it's just that you
> > > don't know how to reproduce the crash.
> > 
> > Then talk to RH's abrt guys.
> 
> Did that already in bug 702262.
> 
> ABRT is just a tool, it doesn't mean that you have to turn off your brain. ABRT
> even asks you to provide more info and include a comment.
Wrong ... today, abrt explictly asked me _if I'd like to add a comment_

As this abrt alert was risen midst of operation without any direct user interaction, I do not have any further information to provide and choose not to comment.
 
> > > *** This bug has been marked as a duplicate of bug 754961 ***
> > 
> > Reopening, because abrt did not detect this as a duplicate.
> 
> Why is that a bug in tumbler then?
Abrt. told me so.

> Did you forget to reassign the bug to the
> abrt component?
It was tumbler which for the n-th time segfaulted.
 
> If you reopen this bug, please state a reason and assign it to the proper
> component. Thank you.

Component: tumbler - It segfaulted
Not a duplicate, because there is no evidence this is a duplicate.
 
> *** This bug has been marked as a duplicate of bug 754961 ***
Please provide evidence this bug is a duplicate.

So far the fact of you closing this BZ to me qualifies as a bold attempt to play down the defectiveness of tumbler, I currently observing dozens of times per day. 

Provided how you reacted, you don't leave me any other alternative but to file each and every such segfault I am currently facing (ca. 10 per day).

Comment 9 Christoph Wickert 2011-11-28 19:51:51 UTC
(In reply to comment #8)
> (In reply to comment #7)
> > ABRT is just a tool, it doesn't mean that you have to turn off your brain. ABRT
> > even asks you to provide more info and include a comment.
> Wrong ... today, abrt explictly asked me _if I'd like to add a comment_

So? How is that a bug in tumbler?

> As this abrt alert was risen midst of operation without any direct user
> interaction, I do not have any further information to provide and choose not to
> comment.

That's at least something, why didn't you mention that earlier?

> > Why is that a bug in tumbler then?
> Abrt. told me so.

Again, you are human being and have a brain that is way more powerful than a tool like ABRT. Please use it.

> > If you reopen this bug, please state a reason and assign it to the proper
> > component. Thank you.
> 
> Component: tumbler - It segfaulted
> Not a duplicate, because there is no evidence this is a duplicate.

Read the backtraces of the two bugs you submitted.

> So far the fact of you closing this BZ to me qualifies as a bold attempt to
> play down the defectiveness of tumbler, 

I am not trying to play down defects in tumbler, why should I? I am trying to triage the report before I forward it upstream.

> I currently observing dozens of times per day.

I already told you to remove the offending gstreamer thumbnailer in bug 754961 but you did not follow my advice.

> Provided how you reacted, you don't leave me any other alternative but to file
> each and every such segfault I am currently facing (ca. 10 per day).

Sure, go ahead, but please use your brain and try to avoid duplicates.

Let me sum up:
- You are not providing helpful input unless explicitly asked to
- You refuse trying to avoid duplicates or to look at the backtraces
- You ignore the advice I give you
- You are blaming me or tumbler for problems in ABRT
- As a maintainer you completely ignore all ABRT bugs filed against any of your packages [1]

Do you really think you are in the position to question how I deal with bugs? Please try support other maintainers when they are trying to triage bugs in order to get them fixed.

[1] https://bugzilla.redhat.com/buglist.cgi?emailassigned_to1=1&short_desc=abrt&short_desc_type=allwordssubstr&email1=rc040203%40freenet.de&emailtype1=substring

Comment 10 Ralf Corsepius 2011-11-29 17:11:40 UTC
(In reply to comment #9)
> (In reply to comment #8)
> > (In reply to comment #7)
> > > ABRT is just a tool, it doesn't mean that you have to turn off your brain. ABRT
> > > even asks you to provide more info and include a comment.
> > Wrong ... today, abrt explictly asked me _if I'd like to add a comment_
> 
> So? How is that a bug in tumbler?
It's tumbler which segfaults. It's irrelevant which of its components triggers the segfault and it's not a user's duty to identify the component.

> > As this abrt alert was risen midst of operation without any direct user
> > interaction, I do not have any further information to provide and choose not to
> > comment.
> 
> That's at least something, why didn't you mention that earlier?
Quite simple: Abrt allows users to do so => You have NO RIGHT to insist on comments, but are OBLIGED to accept the reporter's decision.

If this doesn't taste you, use your FESCO position to have this stuff called abrt fixed or removed (A decision I would wildly applaude to - This stuff is unsuiteable to end-users).
 
> > I currently observing dozens of times per day.
> 
> I already told you to remove the offending gstreamer thumbnailer in bug 754961
> but you did not follow my advice.

Correct. Why should I corrupt my installation? Why don't you do it?

> Let me sum up:
> - You are not providing helpful input unless explicitly asked to.
Nonsense - You are not comprehending, there was nothing to say.

> - You refuse trying to avoid duplicates or to look at the backtraces
Correct, I refuse to do abrt's job.

> - You ignore the advice I give you
> - You are blaming me or tumbler for problems in ABRT
> - As a maintainer you completely ignore all ABRT bugs filed against any of your
> packages [1]
Correct. There is nothing I can do about these segfaults.

.... so do you - The only difference is you being more vocal before closing them.

I guess I'll do what I have been doing before f16: Deinstall abrt.

Comment 11 Christoph Wickert 2011-12-04 10:03:29 UTC
(In reply to comment #10)
> (In reply to comment #9)
> > That's at least something, why didn't you mention that earlier?
> Quite simple: Abrt allows users to do so => 

So you are trying to say everything that a tool allows to do also makes sense? yum allows you to remove the kernel and coreutils even allows you to 'rm -rf /'. Still you are not doing it.

> You have NO RIGHT to insist on
> comments, but are OBLIGED to accept the reporter's decision.

Ask yourself: Why did you file this bug? Do you want it to get fixed or do you just want to rant about it? In that case, please allow me to properly close it as duplicate (because it IS a duplicate). If you want it to get fixed, you should help me and this includes providing all necessary info.

> If this doesn't taste you, use your FESCO position to have this stuff called
> abrt fixed or removed (A decision I would wildly applaude to - This stuff is
> unsuiteable to end-users).

Everybody can submit proposals to FESCo, why don't you do it? I don't think it's unusable and I already managed to fix some bugs by 
a) properly triaging them and 
b) forwarding them upstream afterwards.

You haven't done any of that but simply ignored all reports. I don't think you are in the position to make a call on ABRT - but that should not stop you from submitting a propsal to FESCo.

> > I already told you to remove the offending gstreamer thumbnailer in bug 754961
> > but you did not follow my advice.
> 
> Correct. Why should I corrupt my installation? 

You are not corrupting anything, this has no impact on any other packages.

> Why don't you do it?

Why should I? A lot of people are using it and only some have problems. 

> > - As a maintainer you completely ignore all ABRT bugs filed against any of your
> > packages [1]
> Correct. There is nothing I can do about these segfaults.
> 
> .... so do you

Hold on. You say neither of us can fix any of these bugs. Then why did you file it if it useless anyway?

> - The only difference is you being more vocal before closing
> them.

Look into the history of this bug. I closed it as duplicate without a comment and then you reopened it with a rant. Who of us is vocal?

> I guess I'll do what I have been doing before f16: Deinstall abrt.

Yes, please do so if you are not willing to help maintainers triage bugs.

*** This bug has been marked as a duplicate of bug 754961 ***