9:21 p.m.
In the past few weeks, I am not sure exactly when, clamav
(www.clamav.net) has ceased to work. I have tried their cvs and new
versions... nothing works. It dies with a SIGILL.
# freshclam
ClamAV update process started at Mon Mar 15 22:17:21 2004
Reading CVD header (main.cvd): OK
main.cvd is up to date (version: 21, sigs: 20094, f-level: 1, builder:
tkojm)
Reading CVD header (daily.cvd): OK
Downloading daily.cvd [*]
Illegal instruction
This seems to me to be either a change in glibc, gnuc et al, or the
kernel. Has a non-executable stack patch been added lately? Does it
throw a SIGILL instead of a SIGSEGV?
I am running a recent Athlon (2200+). According to cpuinfo it supports
cmov, so I don't think that is it.
processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 8
model name : AMD Athlon(tm) XP 2200+
stepping : 0
cpu MHz : 1797.047
cache size : 256 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow
bogomips : 3555.32
I am running the latest development as of this morning (including
kernel-2.6.3-2.1.253). I know devel/rawhide can be treacherous, but if
no one tries it, bugs don't get found and fixed.
So, how do I go about filing this bug?
Trever
--
"ADVISORY: There Is an Extremely Small but Nonzero Chance That, Through
a Process Known as 'Tunneling,' This Message May Spontaneously Disappear
from Its Present Location and Reappear at Any Random Place in the
Universe, Including Your Neighbour's Kitchen. The Author Will Not Be
Responsible for Any Damages or Inconveniences That May Result." --
Unknown
12:20 a.m.
On Mon, 15 Mar 2004, Trever L. Adams wrote:
In the past few weeks, I am not sure exactly when, clamav
(www.clamav.net) has ceased to work. I have tried their cvs and new
versions... nothing works. It dies with a SIGILL.
As you are not sure what might cause the issue the only given thing is
clamav. So I suggest you file it against clamav and take it from there.
Hugo.
--
All email sent to me is bound to the rules described on my homepage.
hvdkooij(a)vanderkooij.org http://hvdkooij.xs4all.nl/
Don't meddle in the affairs of sysadmins,
for they are subtle and quick to anger.
11:46 a.m.
On Tue, 2004-03-16 at 07:20 +0100, Hugo van der Kooij wrote:
On Mon, 15 Mar 2004, Trever L. Adams wrote:
> In the past few weeks, I am not sure exactly when, clamav
> (www.clamav.net) has ceased to work. I have tried their cvs and new
> versions... nothing works. It dies with a SIGILL.
As you are not sure what might cause the issue the only given thing is
clamav. So I suggest you file it against clamav and take it from there.
Hugo.
I would gladly agree with you, except it was working until the last week
or so of rawhide. I will ask them about it, but it does seem to be
related to the work in rawhide.
Trever
--
"It was as true as taxes is. And nothing's truer than them."
-- Charles Dickens (1812-70)
1:42 p.m.
On Tue, 16 Mar 2004, Trever L. Adams wrote:
Date: Tue, 16 Mar 2004 12:46:14 -0500
From: Trever L. Adams <tadams-lists(a)myrealbox.com>
To: Development discussions related to Fedora Core
<fedora-devel-list(a)redhat.com>
Content-Type: text/plain
Reply-To: Development discussions related to Fedora Core
<fedora-devel-list(a)redhat.com>
X-BeenThere: fedora-devel-list(a)redhat.com
Subject: Re: Not sure how to file this bug
On Tue, 2004-03-16 at 07:20 +0100, Hugo van der Kooij wrote:
> On Mon, 15 Mar 2004, Trever L. Adams wrote:
>
> > In the past few weeks, I am not sure exactly when, clamav
> > (www.clamav.net) has ceased to work. I have tried their cvs and new
> > versions... nothing works. It dies with a SIGILL.
>
> As you are not sure what might cause the issue the only given thing is
> clamav. So I suggest you file it against clamav and take it from there.
>
> Hugo.
I would gladly agree with you, except it was working until the last week
or so of rawhide. I will ask them about it, but it does seem to be
related to the work in rawhide.
The best thing to do is to get a stack backtrace and see where
the SIGILL is actually occuring. If it occurs inside clamav,
then it is probably a clamav bug that should be filed to the
clamav people. If it occurs inside a system supplied library,
then a bug report should probably be filed against that component
in Red Hat bugzilla.
It could also possibly be a compiler bug or something, or it
could be a coding error.
Try recompiling clamav with -O0 to see if the problem goes away.
--
Mike A. Harris ftp://people.redhat.com/mharris
OS Systems Engineer - XFree86 maintainer - Red Hat
1:59 p.m.
Before the below trace back, it is definitely in clamav, but I am not
seem to see the files in clamav source code, are these known to you?
[Switching to Thread -150235456 (LWP 10936)]
0x00e93dc8 in __gmpz_set_str (x=0xfeece860, str=0x58eb9e "", base=10)
at ../../mpz/set_str.c:133
133 ../../mpz/set_str.c: No such file or directory.
in ../../mpz/set_str.c
(gdb) back
#0 0x00e93dc8 in __gmpz_set_str (x=0xfeece860, str=0x58eb9e "",
base=10)
at ../../mpz/set_str.c:133
#1 0x00e8f4a3 in __gmpz_init_set_str (x=0xfeece860, str=0xea52e0 "",
base=15356640) at ../../mpz/iset_str.c:45
On Tue, 2004-03-16 at 14:42 -0500, Mike A. Harris wrote:
On Tue, 16 Mar 2004, Trever L. Adams wrote:
>Date: Tue, 16 Mar 2004 12:46:14 -0500
>From: Trever L. Adams <tadams-lists(a)myrealbox.com>
>To: Development discussions related to Fedora Core
> <fedora-devel-list(a)redhat.com>
>Content-Type: text/plain
>Reply-To: Development discussions related to Fedora Core
> <fedora-devel-list(a)redhat.com>
>X-BeenThere: fedora-devel-list(a)redhat.com
>Subject: Re: Not sure how to file this bug
>
>On Tue, 2004-03-16 at 07:20 +0100, Hugo van der Kooij wrote:
>
>> On Mon, 15 Mar 2004, Trever L. Adams wrote:
>>
>> > In the past few weeks, I am not sure exactly when, clamav
>> > (www.clamav.net) has ceased to work. I have tried their cvs and new
>> > versions... nothing works. It dies with a SIGILL.
>>
>> As you are not sure what might cause the issue the only given thing is
>> clamav. So I suggest you file it against clamav and take it from there.
>>
>> Hugo.
>
>I would gladly agree with you, except it was working until the last week
>or so of rawhide. I will ask them about it, but it does seem to be
>related to the work in rawhide.
The best thing to do is to get a stack backtrace and see where
the SIGILL is actually occuring. If it occurs inside clamav,
then it is probably a clamav bug that should be filed to the
clamav people. If it occurs inside a system supplied library,
then a bug report should probably be filed against that component
in Red Hat bugzilla.
It could also possibly be a compiler bug or something, or it
could be a coding error.
Try recompiling clamav with -O0 to see if the problem goes away.
--
Mike A. Harris ftp://people.redhat.com/mharris
OS Systems Engineer - XFree86 maintainer - Red Hat
--
fedora-devel-list mailing list
fedora-devel-list(a)redhat.com
http://www.redhat.com/mailman/listinfo/fedora-devel-list
--
"ADVISORY: There Is an Extremely Small but Nonzero Chance That, Through
a Process Known as 'Tunneling,' This Message May Spontaneously Disappear
from Its Present Location and Reappear at Any Random Place in the
Universe, Including Your Neighbour's Kitchen. The Author Will Not Be
Responsible for Any Damages or Inconveniences That May Result."
-- Unknown
7346
days inactive
7346
days old
5 comments
4 participants
participants (4)
-
Hugo van der Kooij -
Mike A. Harris -
Miloslav Trmač -
Trever L. Adams