Kudzu at Startup(BOOT HANG)

sean bruno sean.bruno at dsl-only.net
Mon Apr 12 04:40:04 UTC 2004


And continuing the insanity...I removed the entry from modprobe.conf and
the ifcfg-eth0 config file and rebooted.  The first time I booted, kudzu
hung the system as it has been.

The second boot, I skipped kudzu and ran kudzu from the command line. 
I'm not familiar with kernel errors, but what does this mean:

Apr 11 21:17:11 localhost kernel: ------------[ cut here ]------------
Apr 11 21:17:11 localhost kernel: kernel BUG at fs/proc/generic.c:660!
Apr 11 21:17:11 localhost kernel: invalid operand: 0000 [#1]
Apr 11 21:17:11 localhost kernel: CPU:    0
Apr 11 21:17:11 localhost kernel: EIP:    0060:[<0219c223>]    Tainted:
P
Apr 11 21:17:11 localhost kernel: EFLAGS: 00010202   (2.6.5-1.315)
Apr 11 21:17:11 localhost kernel: EIP is at remove_proc_entry+0xa5/0xf6
Apr 11 21:17:11 localhost kernel: eax: 00000001   ebx: 17dba9cc   ecx:
00000000   edx: 21fde880
Apr 11 21:17:11 localhost kernel: esi: 21fde8b0   edi: 17d67800   ebp:
00000005   esp: 17f6fee0
Apr 11 21:17:11 localhost kernel: ds: 007b   es: 007b   ss: 0068
Apr 11 21:17:11 localhost kernel: Process modprobe (pid: 1653,
threadinfo=17f6f000 task=17f5cc50)
Apr 11 21:17:11 localhost kernel: Stack: 21fde880 17dba9cc 17dba000
17dba000 17dba000 17f6f000 228a6092 17dba000
Apr 11 21:17:11 localhost kernel:        228a07d6 17dba000 229cb80e
17dba000 03467000 229d00fe 03467000 229d2e04
Apr 11 21:17:11 localhost kernel:        229d2e04 021d2582 03467044
0221fb80 229d2e04 229d2e68 0221fba2 0233b1ec
Apr 11 21:17:11 localhost kernel: Call Trace:
Apr 11 21:17:11 localhost kernel:  [<228a6092>]
ieee80211_proc_remove+0x28/0x34 [wlan]
Apr 11 21:17:11 localhost kernel:  [<228a07d6>]
ieee80211_ifdetach+0x5a/0x7f [wlan]
Apr 11 21:17:11 localhost kernel:  [<229cb80e>] ath_detach+0x48/0x4d
[ath_pci]
Apr 11 21:17:11 localhost kernel:  [<229d00fe>] ath_pci_remove+0x15/0x68
[ath_pci]
Apr 11 21:17:11 localhost kernel:  [<021d2582>]
pci_device_remove+0x16/0x28
Apr 11 21:17:11 localhost kernel:  [<0221fb80>]
device_release_driver+0x3c/0x46
Apr 11 21:17:11 localhost kernel:  [<0221fba2>] driver_detach+0x18/0x26
Apr 11 21:17:11 localhost kernel:  [<0221fd71>]
bus_remove_driver+0x37/0x64
Apr 11 21:17:11 localhost kernel:  [<02220043>]
driver_unregister+0xc/0x2a
Apr 11 21:17:11 localhost kernel:  [<021d26de>]
pci_unregister_driver+0xb/0x13
Apr 11 21:17:11 localhost kernel:  [<229d01c7>] exit_ath_pci+0xf/0x22
[ath_pci]
Apr 11 21:17:11 localhost kernel:  [<02134cb9>]
sys_delete_module+0x122/0x162
Apr 11 21:17:11 localhost kernel:  [<0212581c>]
ptrace_notify+0x109/0x1ab
Apr 11 21:17:11 localhost kernel:
Apr 11 21:17:11 localhost kernel: Code: 0f 0b 94 02 d2 d6 2f 02 8b 47 40
85 c0 75 09 89 f8 e8 0d ff


This went away after I placed the module entry back into modprobe.conf. 
It still halts the boot process, but it runs fine from the command
line...Is this a REDHAT bug or a bug with the driver?

On Sun, 2004-04-11 at 20:52, sean bruno wrote:
> Here I go again...replying to my own post...hmmm...
> 
> Anyway, I ran another strace of Kudzu at boot time.  It is still
> hanging.  I took Charles Anderson's suggestion and modified my strace
> with a "-f" so that it would follow forks.
> 
> The last thing Kudzu tries to do is delete my network adaptor(madwifi). 
> It doesn't even complete the delete for some reason.  Here is a snippet:
> 925   close(0 <unfinished ...>
> 925   <... close resumed> )             = 0
> 925   close(1)                          = 0
> 925   close(2)                          = 0
> 925   dup2(15, 0)                       = 0
> 925   dup2(15, 1)                       = 1
> 925   dup2(15, 2)                       = 2
> 925   execve("/sbin/modprobe", ["/sbin/modprobe", "-r", "ath_pci"], [/*
> 13 vars */]) = 0
> 925   uname({sys="Linux", node="localhost.localdomain", ...}) = 0
> 925   brk(0)                            = 0x85e6000
> 925   open("/etc/ld.so.preload", O_RDONLY) = -1 ENOENT (No such file or
> directory)
> 925   open("/etc/ld.so.cache", O_RDONLY) = 16
> 925   fstat64(16, {st_mode=S_IFREG|0644, st_size=57319, ...}) = 0
> 925   old_mmap(NULL, 57319, PROT_READ, MAP_PRIVATE, 16, 0) = 0xf705b000
> 925   close(16)                         = 0
> 925   open("/lib/tls/libc.so.6", O_RDONLY) = 16
> 925   read(16,
> "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0000\220"..., 512) = 512
> 925   fstat64(16, {st_mode=S_IFREG|0755, st_size=1455072, ...}) = 0
> 925   old_mmap(0x4e4000, 1162188, PROT_READ|PROT_EXEC, MAP_PRIVATE, 16,
> 0) = 0x4e4000
> 925   old_mmap(0x5fa000, 16384, PROT_READ|PROT_WRITE,
> MAP_PRIVATE|MAP_FIXED, 16, 0x115000) = 0x5fa000
> 925   old_mmap(0x5fe000, 7116, PROT_READ|PROT_WRITE,
> MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x5fe000
> 925   close(16)                         = 0
> 925   old_mmap(NULL, 4096, PROT_READ|PROT_WRITE,
> MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xf705a000
> 925   mprotect(0x5fa000, 4096, PROT_READ) = 0
> 925   mprotect(0x4e0000, 4096, PROT_READ) = 0
> 925   set_thread_area({entry_number:-1 -> 6, base_addr:0xf705a2a0,
> limit:1048575, seg_32bit:1, contents:0, read_exec_only:0,
> limit_in_pages:1, seg_not_present:0, useable:1}) = 0
> 925   munmap(0xf705b000, 57319)         = 0
> 925   uname({sys="Linux", node="localhost.localdomain", ...}) = 0
> 925   uname({sys="Linux", node="localhost.localdomain", ...}) = 0
> 925   fstat64(2, {st_mode=S_IFCHR|0666, st_rdev=makedev(1, 3), ...}) = 0
> 925   brk(0)                            = 0x85e6000
> 925   brk(0x8607000)                    = 0x8607000
> 925   brk(0)                            = 0x8607000
> 925   open("/etc/modprobe.conf.dist", O_RDONLY) = 16
> 925   fstat64(16, {st_mode=S_IFREG|0644, st_size=5851, ...}) = 0
> 925   mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
> -1, 0) = 0xf7059000
> 925   read(16, "# default modutils aliases\nalias"..., 4096) = 4096
> 925   read(16, "probe -o dummy0 --ignore-install"..., 4096) = 1755
> 925   read(16, "", 4096)                = 0
> 925   close(16)                         = 0
> 925   munmap(0xf7059000, 4096)          = 0
> 925   open("/etc/modprobe.conf", O_RDONLY) = 16
> 925   fstat64(16, {st_mode=S_IFREG|0644, st_size=356, ...}) = 0
> 925   mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
> -1, 0) = 0xf7058000
> 925   read(16, "alias ath0 ath_pci\nalias sound-s"..., 4096) = 356
> 925   read(16, "", 4096)                = 0
> 925   close(16)                         = 0
> 925   munmap(0xf7058000, 4096)          = 0
> 925   open("/lib/modules/2.6.5-1.315/modules.dep", O_RDONLY) = 16
> 925   fstat64(16, {st_mode=S_IFREG|0644, st_size=169653, ...}) = 0
> 925   mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
> -1, 0) = 0xf7057000
> 925   read(16, "/lib/modules/2.6.5-1.315/net/ath"..., 4096) = 4096
> 925   close(16)                         = 0
> 925   munmap(0xf7057000, 4096)          = 0
> 925   open("/proc/modules", O_RDONLY)   = 16
> 925   fstat64(16, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
> 925   mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS,
> -1, 0) = 0xf7056000
> 925   read(16, "e100 33668 0 - Live 0x229c2000\nm"..., 1024) = 943
> 925   close(16)                         = 0
> 925   munmap(0xf7056000, 4096)          = 0
> 925   delete_module("ath_pci"
> 
> Here endeth the log...Anyone?  Anyone?
> 
> :)
> 
> 
> On Sun, 2004-04-11 at 12:05, Sean Bruno wrote:
> > I believe so...I have two output from strace...one from the command line
> > after the system starts, and one from the init script.
> > 
> > 
> > On Sun, 2004-04-11 at 11:51, Charles R. Anderson wrote:
> > > On Sun, Apr 11, 2004 at 11:19:35AM -0700, Sean Bruno wrote:
> > > > Well...I have modified the init script for kudzu and had it run an
> > > > "strace" on kudzu to see what it is hanging on.
> > > > 
> > > > It appears to be sitting in a wait call for a pid to exit.  I didn't
> > > > want to post the entire strace out put to the list, since nobody else is
> > > > having this issue...Does someone want to look at it for me and see if
> > > > they can figure out why it is hanging?
> > > 
> > > Did you run "strace -f" so the strace follows fork?
> > -- 
> > Sean Bruno
> > Telecommunications Engineer
> > Metro One Telecommunications
> > Desk (503)524-1632
> > Cell (503)358-6832
> > 
> 





More information about the fedora-test-list mailing list