Login failure after restart. Every time.

[Mike Kupfer-2]

Kevin O'Gorman wrote:

> The Xorg logs are too big for this email list, so they are in pastebin:
>     http://pastebin.com/38syicdp  is Xorg.0.log
>     http://pastebin.com/NgxzkSia is Xorg.0.log.old
>
> The .xsession_errors is below.  There seem to be several things going
> wrong, so I'm not sure which ones are caused by others.

The messages in .xsession_errors all seem to be from applications
complaining that the X server has died.

And sure enough, your Xorg.0.log.old has

    Fatal server error:
    Failed to submit batchbuffer: No space left on device

That's probably the thing to key off of for further research.  Googling
for "Failed to submit batchbuffer" shows many hits.  
http://ubuntuforums.org/showthread.php?t=1413061 explains what the
message refers to, and it links to
https://wiki.ubuntu.com/X/Troubleshooting/Freeze, which looks useful.

mike

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Kevin O'Gorman]

On Sat, Mar 24, 2012 at 4:47 PM, Mike Kupfer <[hidden email]> wrote:

> Kevin O'Gorman wrote:
>
>> The Xorg logs are too big for this email list, so they are in pastebin:
>>     http://pastebin.com/38syicdp  is Xorg.0.log
>>     http://pastebin.com/NgxzkSia is Xorg.0.log.old
>>
>> The .xsession_errors is below.  There seem to be several things going
>> wrong, so I'm not sure which ones are caused by others.
>
> The messages in .xsession_errors all seem to be from applications
> complaining that the X server has died.
>
> And sure enough, your Xorg.0.log.old has
>
>    Fatal server error:
>    Failed to submit batchbuffer: No space left on device
>
> That's probably the thing to key off of for further research.  Googling
> for "Failed to submit batchbuffer" shows many hits.
> http://ubuntuforums.org/showthread.php?t=1413061 explains what the
> message refers to, and it links to
> https://wiki.ubuntu.com/X/Troubleshooting/Freeze, which looks useful.

It would look more useful if I were actually experiencing a freeze.
I'm not.  To repeat: when I reboot, the first X gnome login aborts
during the playing of the welcome sound, and takes me back to the
login screen.  Almost always the subsequent identical login works
fine.  Once I had to try three times, I think.  All operations after
that are normal.

Judging from the help in the indicated thread, it may be time for a bug report.

--
Kevin O'Gorman, PhD

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Mike Kupfer-2]

Kevin O'Gorman wrote:

> On Sat, Mar 24, 2012 at 4:47 PM, Mike Kupfer <[hidden email]> wrote:

> > https://wiki.ubuntu.com/X/Troubleshooting/Freeze, which looks useful.
>
> It would look more useful if I were actually experiencing a freeze.
> I'm not.  

Yes, I got that.  Sorry, I guess I should have explained my thinking in
a little more detail.

Your problem appears to be due to a GPU lockup.  In your case, the X
server apparently detects the problem and aborts, rather than itself
freezing.  But the underlying problem, the thing that will need to get
fixed, appears to be the same.  

> Judging from the help in the indicated thread, it may be time for a
> bug report.

Agreed.  The X/Troubleshooting/Freeze page has a section on "Reporting
GPU lockup Bugs".  So in additional to the information you've already
provided, you'll want to collect the dmesg log (check in /var/log),
information about your video card (though I suppose you could just
attach the Xorg.log.old, as that would have the hardware info), and what
desktop visual effects you have enabled, if any.

regards,
mike

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Basil Chupin-2]

On 24/03/12 03:00, Kevin O'Gorman wrote:

> On Fri, Mar 23, 2012 at 6:00 AM, Basil Chupin<[hidden email]>  wrote:
>> On 23/03/12 19:13, Kevin O'Gorman wrote:
>>> On Thu, Mar 22, 2012 at 9:18 PM, Basil Chupin<[hidden email]>
>>>   wrote:
>>>> I just checked 'man fsck' and there is no "-f" parameter/[switch] for
>>>> fsck.
>>>>
>>>> There is "-r" but no "-f".
>>>>
>>>>
>>>> BC
>>> You have to read quite carefully to get it and be familiar with the
>>> filesystem-specific options for your filesystem.  Since the -f option
>>> is not recognized by fsck proper, it is passed on the the actual
>>> filesystem-specific executable.  For most filesystems, it's a
>>> force-flag that makes fsck.??? do a full check.
>>
>> I see.
>>
>> What you are trying to say is that it is an undocumented "feature".
> I am not.  Not at all.
>
>  From output of "man fsck":
>         fs-specific-options
>                Options  which  are  not  understood  by  fsck are
> passed to the filesystem-specific checker.
>                These arguments must not take arguments, as there is no
> way for fsck to be able  to  properly
>                guess which arguments take options and which don't.
>
>                Options  and  arguments which follow the -- are treated
> as file system-specific options to be
>                passed to the file system-specific checker.

[pruned]

Sorry for the delayed response but I have been trying to solve a couple
of (nasty) hassles - one solved the other still in play :'( .

OK, what all of the above states is that if fsck does not understand
what it is you are trying to do then it sends the command to the
"filesystem-specific checker" - which in case, assuming that you are
using one of the extX filesystems, to e2fsck and with the parameter
"-f"; and the "-f" parameter in e2fsck means to "force a check". If I
remember correctly, the "f" parameter in Windows/OS2 chkdsk means "fix
errors".

Now, what I stated right from the start is that fsck is performed
automatically on each boot of system and it found NOTHING wrong with the
files system (ext4) on my wife's computer. Nothing - no errors, zilch,
and yet her system was playing up something aweful. So, as I stated, I
ran e2fsck manually and this found a host of errors; and because I did
not use the "-p" parameter in e2fsck I was able to see what those errors
were and answered YES everytime e2fsck asked if I wanted them to be
corrected.

If you are happy with the result of your exercise in running 'fsck -f'
then I shall debate the issue with you no further. I can only relate to
you my personal, real life, experience and how I solved my wife's
computer hassles :-) .

BC

--
The more sand has escaped from the hourglass of our life, the clearer we should see through it.
              Niccolo Machiavelli


--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Kevin O'Gorman]

On Mon, Mar 26, 2012 at 4:51 AM, Basil Chupin <[hidden email]> wrote:

> On 24/03/12 03:00, Kevin O'Gorman wrote:
>>
>> On Fri, Mar 23, 2012 at 6:00 AM, Basil Chupin<[hidden email]>
>>  wrote:
>>>
>>> On 23/03/12 19:13, Kevin O'Gorman wrote:
>>>>
>>>> On Thu, Mar 22, 2012 at 9:18 PM, Basil Chupin<[hidden email]>
>>>>  wrote:
>>>>>
>>>>> I just checked 'man fsck' and there is no "-f" parameter/[switch] for
>>>>> fsck.
>>>>>
>>>>> There is "-r" but no "-f".
>>>>>
>>>>>
>>>>> BC
>>>>
>>>> You have to read quite carefully to get it and be familiar with the
>>>> filesystem-specific options for your filesystem.  Since the -f option
>>>> is not recognized by fsck proper, it is passed on the the actual
>>>> filesystem-specific executable.  For most filesystems, it's a
>>>> force-flag that makes fsck.??? do a full check.
>>>
>>>
>>> I see.
>>>
>>> What you are trying to say is that it is an undocumented "feature".
>>
>> I am not.  Not at all.
>>
>>  From output of "man fsck":
>>        fs-specific-options
>>               Options  which  are  not  understood  by  fsck are
>> passed to the filesystem-specific checker.
>>               These arguments must not take arguments, as there is no
>> way for fsck to be able  to  properly
>>               guess which arguments take options and which don't.
>>
>>               Options  and  arguments which follow the -- are treated
>> as file system-specific options to be
>>               passed to the file system-specific checker.
>
> [pruned]
>
> Sorry for the delayed response but I have been trying to solve a couple of
> (nasty) hassles - one solved the other still in play :'( .
>
> OK, what all of the above states is that if fsck does not understand what it
> is you are trying to do then it sends the command to the
> "filesystem-specific checker" - which in case, assuming that you are using
> one of the extX filesystems, to e2fsck and with the parameter "-f"; and the
> "-f" parameter in e2fsck means to "force a check". If I remember correctly,
> the "f" parameter in Windows/OS2 chkdsk means "fix errors".
>
> Now, what I stated right from the start is that fsck is performed
> automatically on each boot of system and it found NOTHING wrong with the
> files system (ext4) on my wife's computer. Nothing - no errors, zilch, and
> yet her system was playing up something aweful. So, as I stated, I ran
> e2fsck manually and this found a host of errors; and because I did not use
> the "-p" parameter in e2fsck I was able to see what those errors were and
> answered YES everytime e2fsck asked if I wanted them to be corrected.
>
> If you are happy with the result of your exercise in running 'fsck -f' then
> I shall debate the issue with you no further. I can only relate to you my
> personal, real life, experience and how I solved my wife's computer hassles
> :-) .

Happy I'm not, but I'm satisfied the filesystem is clean.  I still
have the original problem.  I've filed a bug, but from the looks of it
it's just one of a great many bugs involving X crashes, freezes or
misbehavior, so I don't expect fast action.  I would be *happy* to be
wrong about that last bit.

BTW, for e2fsck, the fix-everything switch is "-y" (mnemonic: "assume yes".)


--
Kevin O'Gorman, PhD

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Colin Law-4]

On 27 March 2012 06:25, Kevin O'Gorman <[hidden email]> wrote:
>
> Happy I'm not, but I'm satisfied the filesystem is clean.  I still
> have the original problem.  I've filed a bug, but from the looks of it
> it's just one of a great many bugs involving X crashes, freezes or
> misbehavior, so I don't expect fast action.  I would be *happy* to be
> wrong about that last bit.

Have you tried logging in as a different user to see whether it is a
system problem or something to do with your user setup?

Colin

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Kevin O'Gorman]

On Tue, Mar 27, 2012 at 1:56 AM, Colin Law <[hidden email]> wrote:

> On 27 March 2012 06:25, Kevin O'Gorman <[hidden email]> wrote:
>>
>> Happy I'm not, but I'm satisfied the filesystem is clean.  I still
>> have the original problem.  I've filed a bug, but from the looks of it
>> it's just one of a great many bugs involving X crashes, freezes or
>> misbehavior, so I don't expect fast action.  I would be *happy* to be
>> wrong about that last bit.
>
> Have you tried logging in as a different user to see whether it is a
> system problem or something to do with your user setup?

I have not, because there is no other user.  Well, there *is* root, I
suppose.  The machine is busy right now; I may try later.

It has also occurred to me to mark the beginning and end of the first
two login attempts in the various logs, to see what's different.

--
Kevin O'Gorman, PhD

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Colin Law-4]

On 27 March 2012 14:35, Kevin O'Gorman <[hidden email]> wrote:

> On Tue, Mar 27, 2012 at 1:56 AM, Colin Law <[hidden email]> wrote:
>> On 27 March 2012 06:25, Kevin O'Gorman <[hidden email]> wrote:
>>>
>>> Happy I'm not, but I'm satisfied the filesystem is clean.  I still
>>> have the original problem.  I've filed a bug, but from the looks of it
>>> it's just one of a great many bugs involving X crashes, freezes or
>>> misbehavior, so I don't expect fast action.  I would be *happy* to be
>>> wrong about that last bit.
>>
>> Have you tried logging in as a different user to see whether it is a
>> system problem or something to do with your user setup?
>
> I have not, because there is no other user.  Well, there *is* root, I
> suppose.  The machine is busy right now; I may try later.

It is worth making a test user for just this sort of situation.

>
> It has also occurred to me to mark the beginning and end of the first
> two login attempts in the various logs, to see what's different.

Good idea.

Colin Law BSc (Hons)

> Kevin O'Gorman, PhD
>

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users

[Kevin O'Gorman]

On Tue, Mar 27, 2012 at 9:36 AM, Colin Law <[hidden email]> wrote:

> On 27 March 2012 14:35, Kevin O'Gorman <[hidden email]> wrote:
>> On Tue, Mar 27, 2012 at 1:56 AM, Colin Law <[hidden email]> wrote:
>>> On 27 March 2012 06:25, Kevin O'Gorman <[hidden email]> wrote:
>>>>
>>>> Happy I'm not, but I'm satisfied the filesystem is clean.  I still
>>>> have the original problem.  I've filed a bug, but from the looks of it
>>>> it's just one of a great many bugs involving X crashes, freezes or
>>>> misbehavior, so I don't expect fast action.  I would be *happy* to be
>>>> wrong about that last bit.
>>>
>>> Have you tried logging in as a different user to see whether it is a
>>> system problem or something to do with your user setup?
>>
>> I have not, because there is no other user.  Well, there *is* root, I
>> suppose.  The machine is busy right now; I may try later.
>
> It is worth making a test user for just this sort of situation.
>
>>
>> It has also occurred to me to mark the beginning and end of the first
>> two login attempts in the various logs, to see what's different.
>
> Good idea.
>
> Colin Law BSc (Hons)

It happens that I can no longer do this, as the problem went away more
or less by itself.  I don't keep a close watch on the updates I
install, but there have been several and one of them may have fixed
it.

Anyway, it's fixed.  For now.

--
Kevin O'Gorman, PhD

--
ubuntu-users mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-users