[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#408002: konqueror: Konqueror blocks forever reading file descriptor 8



On Mon, Jan 22, 2007 at 11:30:53PM +0100, Sune Vuorela wrote:
> severity 408002 normal
> tags 408002 +unreproducible moreinfo
> thanks
> 
> > On amd64, Konqueror frequently fails to launch or occasionally freezes
> > up completely while it's already running. In either case, it's becase
> > konqueror is blocking forever trying to read file descriptor 8.
> >
> > Killing kded and gam_server and restarting any KDE apps lets konqueror
> > load again. I have never seen this bug occur on x86.
> 
> No one else have seen this on amd64 - so it must be specific to your setup. 
> Downgrading and tagging unreproducible.
> 
> Please try to provide more info about your setup.

When this occurs, konqueror, gam_server and kded all block
reading/writing various unix sockets:

[bloom@little-cat-a ~]$ strace -p 9321 #konqueror
Process 9321 attached - interrupt to quit
read(8,  <unfinished ...>
Process 9321 detached
[bloom@little-cat-a ~]$ strace -p 5863 #kded
Process 5863 attached - interrupt to quit
write(9, "\n\0\1\0\v\0\3\0\0\0", 10 <unfinished ...>
Process 5863 detached
[bloom@little-cat-a ~]$ strace -p 5865 #gam_server
Process 5865 attached - interrupt to quit
write(7, "E\0\1\0\31\1\7\0;\0/home/bloom/.kde/share"..., 69 <unfinished ...>
Process 5865 detached

<partial LSOF listing>
konqueror  9321 bloom    8u  unix 0xffff81000bc43400      19947 socket
kded       5863 bloom    9u  unix 0xffff81000b71fc80      13356 socket
gam_serve  5865 bloom    7u  unix 0xffff810005176e40      13357 socket

I sent you my backtrace for konqueror. Here's a (useless) backtrace for gam_server:
#0  0x00002b49fb9233d2 in write () from /lib/libc.so.6
#1  0x0000000000408c80 in ?? ()
#2  0x000000000040a202 in ?? ()
#3  0x0000000000409788 in ?? ()
#4  0x0000000000409b06 in ?? ()
#5  0x00000000004084c9 in ?? ()
#6  0x00002b49fb5d4c73 in g_main_context_dispatch ()
   from /usr/lib/libglib-2.0.so.0
#7  0x00002b49fb5d7abd in g_main_context_check ()
   from /usr/lib/libglib-2.0.so.0
#8  0x00002b49fb5d7da6 in g_main_loop_run () from /usr/lib/libglib-2.0.so.0
#9  0x00000000004050e7 in ?? ()
#10 0x00002b49fb8834ca in __libc_start_main () from /lib/libc.so.6
#11 0x00000000004038da in ?? ()
#12 0x00007fffaf618f38 in ?? ()
#13 0x0000000000000000 in ?? ()

Here's an even more useless backtrace for kded
#0  0x00002ab7772433d2 in ?? ()
#1  0x00002ab779381128 in ?? ()
#2  0x00007fff3403bf01 in ?? ()
#3  0x00000000008adfe0 in ?? ()
#4  0x0000000000894d60 in ?? ()
#5  0x000000000000000a in ?? ()
#6  0x00007fff3403bd60 in ?? ()
#7  0x0000000900894d70 in ?? ()
#8  0x00000000008adfa0 in ?? ()
#9  0x0000000a0000000a in ?? ()
#10 0x00007fff3403cd90 in ?? ()
#11 0x00002ab779381453 in ?? ()
#12 0x00002ab7773bf0e0 in ?? ()
#13 0x00002ab7771b6268 in ?? ()
#14 0x000000003403bd50 in ?? ()
#15 0x00000000005c7280 in ?? ()
#16 0x0000000000000000 in ?? ()

I have my .kde directory shared between an my amd64 laptop and several
x86 boxes through Subversion. I have tried deleting ~/.kde/ksycoca
(the only binary file I could find in my my .kde directory) as well
/var/tmp/kdecache-bloom to no avail. I could try moving the whole .kde
directory out of the way to see what happens.

--Ken Bloom

-- 
Ken Bloom. PhD candidate. Linguistic Cognition Laboratory.
Department of Computer Science. Illinois Institute of Technology.
http://www.iit.edu/~kbloom1/

Attachment: signature.asc
Description: Digital signature


Reply to: