Bug 216173 - Kudzu-1.2.60-1 segfaults
Kudzu-1.2.60-1 segfaults
Product: Fedora
Classification: Fedora
Component: kudzu (Show other bugs)
powerpc Linux
medium Severity medium
: ---
: ---
Assigned To: Bill Nottingham
David Lawrence
Depends On:
  Show dependency treegraph
Reported: 2006-11-17 11:44 EST by Joseph Sacco
Modified: 2014-03-16 23:04 EDT (History)
1 user (show)

See Also:
Fixed In Version: 1.2.61-1
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2006-11-17 15:13:40 EST
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)
patch for this issue (468 bytes, patch)
2006-11-17 14:57 EST, Bill Nottingham
no flags Details | Diff

  None (edit)
Description Joseph Sacco 2006-11-17 11:44:41 EST
Kudzu now segfaults when run, which is something it did not do in the past. This
system is updated frequently, usually once a day. The segfault behavior started
some time in the past 7-10 days. When run from within GDB, the backtrace
provides no useful information [much like this bug report].

Comment 1 Bill Nottingham 2006-11-17 13:27:08 EST
Do you get a more useful backtrace if you install kudzu-debuginfo?
Comment 2 Bill Nottingham 2006-11-17 13:28:26 EST
Failing the backtrace, a strace of it segfaulting could be useful.
Comment 3 Joseph Sacco 2006-11-17 13:45:31 EST

I did rebuild from the SRC RPM and did install kudzu-debuginfo. Doing so allows
me to run gdb and stop in "main".  When I kick off gdb from there, kudzu
segfaults. The messages displayed are useless.

OK... I will boot back into rawhide and try strace.

One other thing which may be totally unrelated, but one never knows...

udev hangs during the boot sequence if there is media present in a ZIP250 drive.
See bug # 216175.

Comment 4 Joseph Sacco 2006-11-17 14:55:05 EST

Running 'strace /sbin/kudzu' was informative:

lstat64("/proc/device-tree/aliases/.", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0
lstat64("/proc/device-tree/AAPL,add-fcode-file", {st_mode=S_IFREG|0444,
st_size=4, ...}) = 0
lstat64("/proc/device-tree/..", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0
lstat64("/proc/device-tree/.", {st_mode=S_IFDIR|0555, st_size=0, ...}) = 0
lstat64("/proc/device-tree/#size-cells", {st_mode=S_IFREG|0444, st_size=4, ...}) = 0
lstat64("/proc/device-tree/#address-cells", {st_mode=S_IFREG|0444, st_size=4,
...}) = 0
open("/proc/device-tree/compatible", O_RDONLY) = 9
read(9, "PowerMac3,4\0MacRISC2\0MacRISC\0Pow"..., 256) = 45
close(9)                                = 0
access("/proc/iSeries", F_OK)           = -1 ENOENT (No such file or directory)
access("/proc/device-tree/mambo", R_OK) = -1 ENOENT (No such file or directory)
access("/proc/device-tree/mambo/bogus-net@0", R_OK) = -1 ENOENT (No such file or
open("/sys/bus/vio/devices", O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_DIRECTORY) = -1
ENOENT (No such file or directory)
--- SIGSEGV (Segmentation fault) @ 0 (0) ---
+++ killed by SIGSEGV +++
Process 3292 detached

Hmmm... kudzu appears to be looking for things that do not exist on a G4 PowerMac.

Comment 5 Bill Nottingham 2006-11-17 14:57:31 EST
Created attachment 141524 [details]
patch for this issue

Try this. Where's that brown paper bag....
Comment 6 Joseph Sacco 2006-11-17 15:11:10 EST

That fixed it.  Nothing like having the right information.

Comment 7 Bill Nottingham 2006-11-17 15:13:40 EST
1.2.61-1 should show up in rawhide tomorrow.

Note You need to log in before you can comment on or make changes to this bug.