Konqueror causing 100% CPU usage

Frank Ianella fianella at verizon.net
Fri Sep 7 08:32:32 PDT 2007


DJ Lucas wrote:
> Frank Ianella wrote:
>   
>> Dan Nicholson wrote:
>>     
>>> On 9/5/07, Frank Ianella <fianella at verizon.net> wrote:
>>>  
>>>       
>>>> attached is the trace log.
>>>>     
>>>>         
>>> Unfortunately, it's over my head an nothing jumps out. One thing to
>>> consider doing is updating to kdelibs/kdebase-3.5.7. It just hasn't
>>> been updated in the book, but it should be a stable increment from
>>> 3.5.6.
>>>
>>>   
>>>       
>> I've already tried that, to no avail. Konqueror appears to load a little
>> further with 3.5.7 based on the progress bar, but the end result is the
>> same.
>>     
>>> I recall that your starting kde from startx. Are you also running
>>> dbus? If you have all the dependencies for kdelibs and base, you
>>> should start a session dbus with dbus-launch.
>>>
>>>   
>>>       
>> I installed dbus when I had 3.5.7 running. dbus-launch was started via
>> the .xinitrc as indicated in chapter 28, again to no avail.
>>     
>>> If nothing really jumps out, it might be best to pull out gdb and do
>>> some real debugging.
>>>
>>> -- 
>>> Dan
>>>   
>>>       
>> In anticipation of this, I had compiled arts/kde with full debug.
>> Konqueror was started from within gdb, but I don't see anything. I don't
>> have any experience with gdb, so I have attched the log. Perhaps there
>> is something that stands out to you.
>>
>> I also decided to log the output from the xserver. This output seems to
>> go a bit further than that of gdb. The output below is 5 lines before an
>> X error is logged, and continues until the x server locks up (end of llg
>> file). Any help in interpreting this is greatly appreciated.
>>
>>     
>
> Looks similar to the firefox/flash bug.   Do you have xserver running at
> 16bit color depth and flash plugin installed?  

No flash plugin, and default color depth is 24.

> Also, using gdb to track
> down X errors, it is suggested to use 'run --sync' otherwise your error
> can return way after the problem has occurred.  

Thanks for the tip!


Regards,

-------------- next part --------------
A non-text attachment was scrubbed...
Name: fianella.vcf
Type: text/x-vcard
Size: 127 bytes
Desc: not available
URL: <http://lists.linuxfromscratch.org/pipermail/blfs-support/attachments/20070907/8c46e518/attachment.vcf>


More information about the blfs-support mailing list