Red Hat Bugzilla – Bug 426575
system-config-network hangs after loading network interfaces, but before starting GUI or TUI
Last modified: 2008-04-22 18:39:13 EDT
Description of problem: When I start system-config-network (I've tried both -gui
and -tui), my CPU goes to 100% and stays there until i press Ctrl+C. When
started with -v -d, I get the following:
 Sat Dec 22 13:35:52 2007: Checking Desc: nVidia Corporation MCP61
Version-Release number of selected component (if applicable):
How reproducible: just start it. I'm not sure what causes this, so I cannot be
more precise, but I'm willing to provide more data on request.
Steps to Reproduce:
1. start system-config-network either from konsole as root or from menu
2. enter password (if started from menu)
3. die waiting ;-)
Obviously, I'd like app to start.
Linux sime.styx.org 184.108.40.206-85.fc8 #1 SMP Fri Dec 7 15:49:36 EST 2007 x86_64
x86_64 x86_64 GNU/Linux
Tried with the original F8 kernel (220.127.116.11-42.fc8) and one in between
(18.104.22.168-63.fc8). Version 22.214.171.124-63.fc8 didn't show modem in lspci, but
everything seemed to remained the same.
I have installed kmod-nvidia from Livna.
00:00.0 RAM memory: nVidia Corporation MCP61 Memory Controller (rev a1)
00:01.0 ISA bridge: nVidia Corporation MCP61 LPC Bridge (rev a2)
00:01.1 SMBus: nVidia Corporation MCP61 SMBus (rev a2)
00:01.2 RAM memory: nVidia Corporation MCP61 Memory Controller (rev a2)
00:02.0 USB Controller: nVidia Corporation MCP61 USB Controller (rev a2)
00:02.1 USB Controller: nVidia Corporation MCP61 USB Controller (rev a2)
00:04.0 PCI bridge: nVidia Corporation MCP61 PCI bridge (rev a1)
00:05.0 Audio device: nVidia Corporation MCP61 High Definition Audio (rev a2)
00:06.0 IDE interface: nVidia Corporation MCP61 IDE (rev a2)
00:07.0 Bridge: nVidia Corporation MCP61 Ethernet (rev a2)
00:08.0 IDE interface: nVidia Corporation MCP61 SATA Controller (rev a2)
00:08.1 IDE interface: nVidia Corporation MCP61 SATA Controller (rev a2)
00:09.0 PCI bridge: nVidia Corporation MCP61 PCI Express bridge (rev a2)
00:0b.0 PCI bridge: nVidia Corporation MCP61 PCI Express bridge (rev a2)
00:0c.0 PCI bridge: nVidia Corporation MCP61 PCI Express bridge (rev a2)
00:0d.0 VGA compatible controller: nVidia Corporation GeForce 6100 nForce 430
00:18.0 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron]
HyperTransport Technology Configuration
00:18.1 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] Address Map
00:18.2 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron] DRAM
00:18.3 Host bridge: Advanced Micro Devices [AMD] K8 [Athlon64/Opteron]
01:08.0 Communication controller: Motorola SM56 PCI Modem
eth0 Link encap:Ethernet HWaddr 00:19:66:30:10:BE
inet addr:192.168.100.6 Bcast:192.168.100.255 Mask:255.255.255.0
inet6 addr: fe80::219:66ff:fe30:10be/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:156 errors:0 dropped:0 overruns:0 frame:0
TX packets:73 errors:0 dropped:0 overruns:0 carrier:0
RX bytes:15446 (15.0 KiB) TX bytes:9949 (9.7 KiB)
Interrupt:252 Base address:0x6000
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:1057 errors:0 dropped:0 overruns:0 frame:0
TX packets:1057 errors:0 dropped:0 overruns:0 carrier:0
RX bytes:2865728 (2.7 MiB) TX bytes:2865728 (2.7 MiB)
I want to install modem since this machine is being set up for a friend who
doesn't have access to DSL. I cannot remove modem from the machine, since it is
sealed (due to the terms of guarantee).
Ok, I've decided to go through all the Python code (though I don't know Pythin,
but its close enough to Perl, C, PHP,...) and I've traced the problem to
/etc/hosts loading. It seems that handling /etc/hosts entries is too time and
memory consuming. The problem occures in NCProfileList.py lines 143 through 151:
for key in hoconf.keys()...
I have a large /etc/hosts, most of it being taken from here:
Since this list is huge (around 600kB), hoconf list gets clogged with items and
this loop just goes on forever.
I don't need this large hosts file on this particular machine and have cropped
it to only few entries, but the problem persists on my other machine where I do
need large hosts file.
Workaround: replace /etc/hosts with something short (i.e. the one having only
localhost in it), do whatever you need to do with system-config-network and,
after you're done (and quit s-c-network), replace /etc/hosts with the original file.
Suggestion: it might be wise to move the handling of /etc/hosts to another app
or at least another window (being started from menus or with the click of the
button), since antispam hosts files are not that uncommon.
Hi, this issue has been fixed in the 1.4.7 version of s-c-n. If you want to try
it you can install it by $ yum update system-config-network --
enablerepo=updates-testing. But thanks to your hosts file i noticed a minor bug
in this version - there is a problem on line 6423 it contains only hostname but
no ip, so you have to edit the file manually and add 127.0.0.1 at the begging
of that line.
system-config-network-1.5.5-1.fc8 has been submitted as an update for Fedora 8
system-config-network-1.5.5-1.fc8 has been pushed to the Fedora 8 testing repository. If problems still persist, please make note of it in this bug report.
If you want to test the update, you can install it with
su -c 'yum --enablerepo=updates-testing update system-config-network'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/F8/FEDORA-2008-2995
Ok, I've installed system-config-network-1.5.5-1.fc8 from F8 testing repository
and have tried it after updating my host file with enteries from
Upon starting, s-c-n has reported errors in the following lines of the hosts file:
127.0.0.1 norbert_sirot.club.fr #[Trojan-Spy.Win32.Banker.anv]
127.0.0.1 ___id___.c.mystat-in.net #[Wildcard DNS]
127.0.0.1 web--search.com #[McAfee.Adware-SBSoft]
127.0.0.1 search_fd.php.coolwebsearch.com #[Wildcard DNS]
Now, underscore IS illegal part of domain/server name (at least SeaMonkey
doesn't recognize it), but minus (lines 3 and 4) is not. I see that only a
double-minus is not accepted (i.e. www2.a-counter.kiev.ua is not labeled as an
error), but since browsers accept double-minuses, I think s-c-n should as well
(to allow blacklisting). Actually, I don't see why s-c-n doesn't just ignore
this kind of errors (except when entries are edited through s-c-n itself).
What confused me more is that 127.0.0.1 entries from the hosts file were not
available for editing in s-c-n (didn't notice this before). I'm not sure if it's
considered a bug or a feature, but upon saving, errors persisted and were still
being reported, whith no way to repair them (inside s-c-n) and I do consider
this a problem: why reporting if I'm not given a way to repair it?
Minor note: upon saving, some lines had spaces (or tabs?) added at the end.
Yes, double minuses should be accepted - I'm going to fix this asap. But you
not right about the 127.0.0.1 entries - there is a checkbox in the left lower
corner called "show loopback" which toggles showing of this entries.
Sorry, I've missed that checkbox (I usually edit hosts file in gvim).
Related to that (although it's not a bug report): since error reports provide
line numbers, you might consider adding "line number" column to the GUI (new
enteries would have this field empty or set to "new" or something). With
something over 18000 lines in hosts file, reports like "error in line xxxxx" are
quite useless if you don't see the line numbers.
system-config-network-1.5.5-1.fc8 has been pushed to the Fedora 8 stable repository. If problems still persist, please make note of it in this bug report.