Bug 871118 - On some crashes abrt doing something but did not file a bug
Summary: On some crashes abrt doing something but did not file a bug
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Fedora
Classification: Fedora
Component: abrt
Version: 18
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: abrt
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2012-10-29 16:50 UTC by Mikhail
Modified: 2012-11-06 20:07 UTC (History)
9 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2012-11-06 20:07:13 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
dmesg output (83.36 KB, text/plain)
2012-11-01 18:05 UTC, Mikhail
no flags Details

Comment 1 Jakub Filak 2012-10-30 07:43:33 UTC
I see that the crash you wanted to submit was for tainted kernel. Please, see bug 861933

Comment 2 Mikhail 2012-10-30 08:29:55 UTC
My kernel is not tainted. Why ABRT think so?

Comment 3 Jakub Filak 2012-10-30 09:24:47 UTC
Due to flags G (All modules loaded have a GPL or compatible license), D (Kernel has oopsed before) and F (Module has been forcibly loaded).

Without any negative feelings, do you find the "This problem should not be reported (it is likely a known problem). A kernel problem occurred, but your kernel has been tainted (flags:GF D). Kernel maintainers are unable to diagnose tainted reports." message not being good enough? Should we find a better message? Or just stop bothering users with tainted kernel crashes?

Comment 4 Mikhail 2012-10-30 18:37:50 UTC
I think message must contain information why kernel is tainted, and have instruction how make the kernel untainted.

I my case, I am very surprised because I am use the original kernel from Fedora repository, without proprietary drivers and without any type of virtualization such as Virtual Box. Why is my kernel is tainted?

Comment 5 Jiri Moskovcak 2012-10-31 07:12:12 UTC
If you didn't load any custom modules then it's probably because: "Kernel has oopsed before" like the comment#3 says...

Comment 6 Mikhail 2012-10-31 07:36:24 UTC
I didn't load any custom modules by myself. May I know which tainted modules are loaded?

Comment 7 Jiri Moskovcak 2012-10-31 07:59:42 UTC
(In reply to comment #6)
> I didn't load any custom modules by myself. May I know which tainted modules
> are loaded?

Please read my comment carefully I said: "If you *didn't*..." - so I believe you didn't load anything manually, the problem is probably the fact that there has been another oops before the one you're trying to report. The kernel is tainted after the first oops, so if you have oops and don't reboot then every other oops will not be reportable, because the kernel is tainted since the first oops.

Comment 8 Mikhail 2012-10-31 08:09:11 UTC
(In reply to comment #7)
> Please read my comment carefully I said: "If you *didn't*..." - so I believe
> you didn't load anything manually, the problem is probably the fact that
> there has been another oops before the one you're trying to report. The
> kernel is tainted after the first oops, so if you have oops and don't reboot
> then every other oops will not be reportable, because the kernel is tainted
> since the first oops.

I'am sorry, thanks, It' would be good if it ABRT message contains this explain :)
I have many another oops before :(
So I want send them on bug tracker.

Comment 9 Mikhail 2012-11-01 18:05:48 UTC
Created attachment 636790 [details]
dmesg output

Comment 10 Mikhail 2012-11-01 18:14:12 UTC
At current time occurs crash, before this no any kernel crash. You can see it from dmesg log here https://bugzilla.redhat.com/attachment.cgi?id=636790 Why ABRT can't able to send a bug report? (Instead it write again that my kernel is tainted )

Comment 11 Jiri Moskovcak 2012-11-02 08:20:03 UTC
This may happen if you have the old oops from tainted kernel and abrt recognizes the new oops on not-tainted as a duplicate of the previous one. In that case abrt deletes the new oops and keep only the old one which is still marked as not_reportable. Please remove the non-reportable oops from ABRT and try catch that oops again.

Comment 12 Jiri Moskovcak 2012-11-02 08:21:50 UTC
The oops deduplicator should compare even the tainted info and if it differs it should not consider it as a dupe.

Comment 13 Jakub Filak 2012-11-02 09:30:46 UTC
(In reply to comment #10)
> At current time occurs crash, before this no any kernel crash. You can see
> it from dmesg log here https://bugzilla.redhat.com/attachment.cgi?id=636790
> Why ABRT can't able to send a bug report? (Instead it write again that my
> kernel is tainted )

You've probably found a bug! Can you please notify kernel guys that fresh and clean Fedora 18 kernel is tainted with F flag.

Comment 14 Nikola Pajkovsky 2012-11-02 10:20:02 UTC
it looks like you don't have correctly singed modules.

[    5.767903] Request for unknown module key 'Fedora kernel signing key: cfabd3c4f3cc79200c45f4f2eea9db8c339c0d66' err -11
[    5.768196] iscsi: registered transport (iser)

and I'm guessing here, kernel load unsigned modules forcibly

Comment 15 Mikhail 2012-11-02 13:00:43 UTC
(In reply to comment #14)
> it looks like you don't have correctly singed modules.
> 
> [    5.767903] Request for unknown module key 'Fedora kernel signing key:
> cfabd3c4f3cc79200c45f4f2eea9db8c339c0d66' err -11
> [    5.768196] iscsi: registered transport (iser)
> 
> and I'm guessing here, kernel load unsigned modules forcibly

Why modules are not correctly signed?

Comment 16 Jakub Filak 2012-11-06 20:07:13 UTC
(In reply to comment #15)
> (In reply to comment #14)
> > it looks like you don't have correctly singed modules.
> > 
> > [    5.767903] Request for unknown module key 'Fedora kernel signing key:
> > cfabd3c4f3cc79200c45f4f2eea9db8c339c0d66' err -11
> > [    5.768196] iscsi: registered transport (iser)
> > 
> > and I'm guessing here, kernel load unsigned modules forcibly
> 
> Why modules are not correctly signed?

It was due to bug 873796. Thanks for your cooperation.


Note You need to log in before you can comment on or make changes to this bug.