Hi,

I recently updated my fedora 13 system with the latest package versions. Following this update the gconf daemon started crashing continuously (i.e. it crashes, gets restarted by the system, crashes, ...).

I am no longer to log into the gnome desktop, i assume because of the dependency it has on gconf.

I am able to run the kde desktop but can realistically only run xterms because most other apps I run (e.g. firefox, thunderbird) cause the system to communicate with the gconf daemon.

Is there a way to disable gconf so kde won't use it or alternatively is there a way to restore gconf to a usable state?

I have opened a bug with redhat and looking through the bugzilla list it looks like others are having the same issue. One has to wonder why such a heavyweight system is used to store desktop and application settings (according to wikipedia gconf uses corba for client server communication).

Here is the crash info:

[New Thread 2342]
Core was generated by `/usr/libexec/gconfd-2'.
Program terminated with signal 11, Segmentation fault.
#0 IA__g_slist_foreach (list=0xe64b847c,
func=0x381583a520 <IOP_component_free>, user_data=0x0) at gslist.c:855
855 gslist.c: No such file or directory.
in gslist.c

Thread 1 (Thread 2342):
#0 IA__g_slist_foreach (list=0xe64b847c,
func=0x381583a520 <IOP_component_free>, user_data=0x0) at gslist.c:855
next = <value optimized out>
#1 0x000000381583a5d5 in IOP_components_free (components=0x70e030)
at iop-profiles.c:671
No locals.
#2 0x000000381583b256 in IOP_TAG_GENERIC_IOP_demarshal (orb=0x6fbd80,
buf=0x71a1b0, ret_type_id=0x7fff72a6f5a8,
ret_profiles=<value optimized out>) at iop-profiles.c:1532
retval = <value optimized out>
v2 = <value optimized out>
len = <value optimized out>
buf = <value optimized out>
v1 = <value optimized out>
version = <value optimized out>
#3 IOP_profile_demarshal (orb=0x6fbd80, buf=0x71a1b0,
ret_type_id=0x7fff72a6f5a8, ret_profiles=<value optimized out>)
at iop-profiles.c:1662
p = <value optimized out>
retval = <value optimized out>
#4 ORBit_demarshal_IOR (orb=0x6fbd80, buf=0x71a1b0,
ret_type_id=0x7fff72a6f5a8, ret_profiles=<value optimized out>)
at iop-profiles.c:1716
profile = 0x381583a520
num_profiles = 3
len = <value optimized out>
profiles = 0x0
type_id = 0x7fff72a6f5d8 "IDL:ConfigListener:1.0"
i = <value optimized out>
#5 0x0000003815832782 in ORBit_demarshal_object (obj=0x7fff72a6f730,
buf=<value optimized out>, orb=0x6fbd80) at corba-object.c:602
type_id = 0x0
profiles = 0x0
__PRETTY_FUNCTION__ = "ORBit_demarshal_object"
#6 0x000000381582e286 in CORBA_ORB_string_to_object (orb=0x6fbd80,
string=<value optimized out>, ev=0x7fff72a6f790) at corba-orb.c:630
retval = 0x0
len = 592
buf = 0x71a1b0
tmpbuf = <value optimized out>
i = <value optimized out>
#7 0x0000000000409e22 in restore_listener (key=0x708c80,
value=<value optimized out>, data=<value optimized out>) at gconfd.c:2085
cl = <value optimized out>
ev = {_id = 0x0, _major = 0, _any = {_type = 0x0, _value = 0x0,
_release = 0 '\000'}}
new_cnxn = <value optimized out>
err = <value optimized out>
#8 listener_logentry_restore_and_destroy_foreach (key=0x708c80,
value=<value optimized out>, data=<value optimized out>) at gconfd.c:2198
lle = 0x708c80
db = 0x6fe330
#9 0x000000380902d483 in IA__g_hash_table_foreach (hash_table=0x6f2e80,
func=0x409da0 <listener_logentry_restore_and_destroy_foreach>,
user_data=0x0) at ghash.c:1325
node = <value optimized out>
i = <value optimized out>
__PRETTY_FUNCTION__ = "IA__g_hash_table_foreach"
#10 0x0000000000409960 in logfile_read () at gconfd.c:2318
logfile = 0x702c10 "/home/wds/.gconfd/saved_state"
logdir = <value optimized out>
entries = <value optimized out>
clients = <value optimized out>
f = <value optimized out>
line = 0x0
lines = 0x712710
#11 0x000000000040a996 in main (argc=<value optimized out>,
argv=<value optimized out>) at gconfd.c:904
act = {__sigaction_handler = {sa_handler = 0x1, sa_sigaction = 0x1},
sa_mask = {__val = {0 <repeats 16 times>}}, sa_flags = 0,
sa_restorer = 0x7fff72a6fac0}
empty_mask = {__val = {0 <repeats 16 times>}}
full_mask = {__val = {18446744067267100671,
18446744073709551615 <repeats 15 times>}}
ev = {_id = 0x0, _major = 0, _any = {_type = 0x0, _value = 0x0,
_release = 0 '\000'}}
orb = <value optimized out>
ior = <value optimized out>
exit_code = 0
err = <value optimized out>
dev_null_fd = <value optimized out>
write_byte_fd = -1
connection = 0x6feae0
From To Syms Read Shared Object Library
0x0000003816811b30 0x000000381682da28 Yes /usr/lib64/libgconf-2.so.4
0x0000003815827990 0x000000381584b6a8 Yes /usr/lib64/libORBit-2.so.0
0x0000003930e09060 0x0000003930e19878 Yes /usr/lib64/libdbus-glib-1.so.2
0x000000380b807090 0x000000380b82e4c8 Yes /lib64/libdbus-1.so.3
0x0000003808005640 0x0000003808010e48 Yes /lib64/libpthread.so.0
0x00000038090155c0 0x0000003809099d58 Yes /lib64/libglib-2.0.so.0
0x000000380781e9a0 0x000000380792b820 Yes /lib64/libc.so.6
0x000000380f401080 0x000000380f401fc8 Yes /lib64/libgmodule-2.0.so.0
0x000000380b008d20 0x000000380b032a78 Yes /lib64/libgobject-2.0.so.0
0x000000380a801590 0x000000380a8029f8 Yes /lib64/libgthread-2.0.so.0
0x0000003808802140 0x00000038088055a8 Yes /lib64/librt.so.1
0x0000003807400af0 0x0000003807418904 Yes /lib64/ld-linux-x86-64.so.2
0x0000003807c00de0 0x0000003807c01998 Yes /lib64/libdl.so.2
0x00007fd010ff2110 0x00007fd010ffa258 Yes /lib64/libnss_files.so.2
0x00007fd010de51f0 0x00007fd010dec2e8 Yes /usr/lib64/GConf/2/libgconfbackend-xml.so
$1 = 0x0
$2 = 0x0
rax 0x7fff73a5f648 140735133644360
rbx 0x381583a520 240879117600
rcx 0x70e010 7397392
rdx 0x0 0
rsi 0x381583a520 240879117600
rdi 0xe64b847c 3863708796
rbp 0x7fff72a6f5a8 0x7fff72a6f5a8
rsp 0x7fff72a6f4e0 0x7fff72a6f4e0
r8 0x70e030 7397424
r9 0x3807883030 240644534320
r10 0xa 10
r11 0x3807882c20 240644533280
r12 0x0 0
r13 0x0 0
r14 0x7fff72a6f5d8 140735116932568
r15 0x3 3
rip 0x380905ad60 0x380905ad60 <IA__g_slist_foreach+16>
eflags 0x10202 [ IF RF ]
cs 0x33 51
ss 0x2b 43
ds 0x0 0
es 0x0 0
fs 0x0 0
gs 0x0 0
Dump of assembler code for function IA__g_slist_foreach:
0x000000380905ad50 <+0>: push %r12
0x000000380905ad52 <+2>: test %rdi,%rdi
0x000000380905ad55 <+5>: mov %rdx,%r12
0x000000380905ad58 <+8>: push %rbp
0x000000380905ad59 <+9>: push %rbx
0x000000380905ad5a <+10>: mov %rsi,%rbx
0x000000380905ad5d <+13>: je 0x380905ad74 <IA__g_slist_foreach+36>
0x000000380905ad5f <+15>: nop
=> 0x000000380905ad60 <+16>: mov 0x8(%rdi),%rbp
0x000000380905ad64 <+20>: mov %r12,%rsi
0x000000380905ad67 <+23>: mov (%rdi),%rdi
0x000000380905ad6a <+26>: callq *%rbx
0x000000380905ad6c <+28>: test %rbp,%rbp
0x000000380905ad6f <+31>: mov %rbp,%rdi
0x000000380905ad72 <+34>: jne 0x380905ad60 <IA__g_slist_foreach+16>
0x000000380905ad74 <+36>: pop %rbx
0x000000380905ad75 <+37>: pop %rbp
0x000000380905ad76 <+38>: pop %r12
0x000000380905ad78 <+40>: retq
End of assembler dump.

You can reverse the changes using yum

Assuming that you have not updated the system again since this error began, use

Then if you know for certain what package caused the problem you can update with

man yum

Thanks for the suggestion.

The gconf site (http://projects.gnome.org/gconf/) stated that it's database is stored in $HOME/.gconf. Do I deleted this directory (actually moved it to another name) as well as a $HOME/.gconfd directory which existed as well. After logging out and back in again no more crashes in gconfd. Looks like the database somehow got corrupted and gconf didn't deal with it gracefully.