Re: [SSI-devel] OPENSSI-FC-1-2-STABLE CVS-May21 shm.c oops

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 454-5900

httpd child exiting on node2 is not expected unless it is a child and
has served the max number of requests.  I would think the shm segments
are on node2 since the initnode failed over to node2 before this oops,
and that no httpd migrations were done at any time.

I also verified that httpd isn't marked to failover in the rc scripts,
so httpd wouldn't be disturbed (killed/restarted) on any nodedown
event either.  What it looks like is httpd was just running fine
locally until a few minutes after node1 down complete has been
announced on node2.

I think Laura would know the long story about my httpd and shm's, but
this oops looks different from the shm oops last time.  Last time I
thought it was something to do with phpa accelerator (shm cache) PHP
plug-in interaction.

-Roger

On 5/22/05, Walker, Bruce J <bru...@hp...> wrote:
> Any idea why httpd was exiting on node 2?  Is that expected and would
> it's shared
> Memory segment be from node 1 or node 2?
>=20
> Bruce
>=20
>=20
> -----Original Message-----
> From: ssi...@li...
> [mailto:ssi...@li...] On Behalf Of Roger
> Tsang
> Sent: Sunday, May 22, 2005 4:57 AM
> To: Ramirez, Laura L
> Cc: OpenSSI developers
> Subject: [SSI-devel] OPENSSI-FC-1-2-STABLE CVS-May21 shm.c oops
>=20
> Hi Laura,
>=20
> I'm getting the following shm.c oops a few minutes after init node
> failover.  I'm using ci/kernel from CVS May 21 OPENSSI-FC-1-2-STABLE.
> I tried the failover again and it is repeatable.  I'll try again with
> bonding properly setup, but I doubt it is going to change things.
>=20
> -Roger
>=20
> I was doing cat linuxrc at the time, so the unexpected output is just
> cosmetic.
>=20
> echo "Loading 8390.o module"
> insmod /lib/8390.o
> echo "Loading r8169.o module"
> ------------[ cut here ]------------
> kernel BUG at shm.c:232!module"
> invalid operand: 0000
> tun loop cls_u32 sch_sfq sch_htb softdog nfsd ip_vs_sed bonding
> ip_conntrack_tft p ip_conntrack_ftp ip_nat_ftp iptable_nat ipt_REJECT
> ipt_LOG ipt_limit ipt_mul
> CPU:    0""
> EIP:    0060:[<c01c4150>]    Not taintedrep HWaddr |
> EFLAGS: 00010246[0-9]*\).*HWaddr \(.*\)/\1-\2/'` do EIP is at shm_close
> [kernel] 0xb0 (2.4.22-1.2199.nptl_ssi_10up)
> eax: d8d4b000   ebx: c05e1e90   ecx: c05e1e90   edx: 00000000
> esi: 02000000   edi: bd311000   ebp: d08d3e64   esp: d08d3e5c
> ds: 0068   es: 0068   ss: 0068
> Process httpd (pid: 133055, stackpage=3Dd08d3000)
> Stack: 00041000 d08c7a00 d08d3e8c c013b316 d08c7a00 d08c7980 00041000
> 00000000
>        d08c7a80 d0a52580 00000000 d08d2000 d08d3ea0 c012196c d0a52580
> d0a52580
>        d0a52580 d08d3ec0 c0127859 d0a52580 d0bd9e00 00000007 00000007
> d08d2000
> Call Trace:
> [<c013b316>] exit_mmap [kernel] 0x166 (0xd08d3e68) [<c012196c>] mmput
> [kernel] 0x4c (0xd08d3e90)sk broadcast $bcast [<c0127859>] do_exit
> [kernel] 0xe9 (0xd08d3ea4) [<c0127c12>] do_group_exit [kernel] 0x32
> (0xd08d3ec4) [<c0131204>] get_signal_to_deliver [kernel] 0x2b4
> (0xd08d3ed8)
> [<c0114b4f>] restore_i387_fxsave [kernel] 0xaf (0xd08d3ee8)   23,1
> 14%
> [<c010b92f>] do_signal [kernel] 0x4f (0xd08d3f1c) [<c0109a78>]
> restore_sigcontext [kernel] 0x458 (0xd08d3f28) [<c0109ee6>]
> sys_sigreturn [kernel] 0x106 (0xd08d3f90) [<c010bb20>] signal_return
> [kernel] 0x14 (0xd08d3fc0)
>=20
> Code: 0f 0b e8 00 66 c4 37 c0 eb a5 8d b6 00 00 00 00 a1 a4 1e 5e
>=20
> Entering kdb (current=3D0xd08d2000, pid 133055) Oops: invalid operand due
> to oops @ 0xc01c4150 eax =3D 0xd8d4b000 ebx =3D 0xc05e1e90 ecx =3D 0xc05e=
1e90
> edx =3D 0x00000000 esi =3D 0x02000000 edi =3D 0xbd311000 esp =3D 0xd08d3e=
5c eip
> =3D 0xc01c4150 ebp =3D 0xd08d3e64 xss =3D 0xc0350068 xcs =3D 0x00000060 e=
flags =3D
> 0x00010246 xds =3D 0xbd310068 xes =3D 0x00000068 origeax =3D 0xffffffff &=
regs
> =3D 0xd08d3e28
> kdb> bt
> Stack traceback for pid 133055
> 0xd08d2000   133055   132992  1    0   R  0xd08d2350 *httpd
> EBP        EIP        Function (args)
> 0xd08d3e64 0xc01c4150 shm_close+0xb0 (0xd08c7a00, 0xd08c7980, 0x41000,
> 0x0, 0xd0
> 8c7a80)
>                                kernel .text 0xc0100000 0xc01c40a0
> 0xc01c4170 0xd08d3e8c 0xc013b316 exit_mmap+0x166 (0xd0a52580,
> 0xd0a52580, 0xd0a52580)
>                                kernel .text 0xc0100000 0xc013b1b0
> 0xc013b340 0xd08d3ea0 0xc012196c mmput+0x4c (0xd0a52580, 0xd0bd9e00,
> 0x7, 0x7, 0xd08d2000)
>                                kernel .text 0xc0100000 0xc0121920
> 0xc01219e0 0xd08d3ec0 0xc0127859 do_exit+0xe9 (0x7, 0xd08d2000, 0x7)
>                                kernel .text 0xc0100000 0xc0127770
> 0xc0127b60
> 0xd08d3ed4 0xc0127c12 do_group_exit+0x32 (0x7, 0x7, 0xd08d3fc4,
> 0xc0114b4f, 0xd0
> 8d23b0)
>                                kernel .text 0xc0100000 0xc0127be0
> 0xc0127c50
> 0xd08d3f18 0xc0131204 get_signal_to_deliver+0x2b4 (0xd08d3f30,
> 0xd08d3fc4, 0xc01 09a78, 0xbfec5980, 0x7)
>                                kernel .text 0xc0100000 0xc0130f50
> 0xc0131310 0xd08d3fbc 0xc010b92f do_signal+0x4f (0x1, 0x8899f50, 0x0,
> 0x889aa58, 0xbfec5d24
> )
>                                kernel .text 0xc0100000 0xc010b8e0
> 0xc010b9d0
>            0xc010bb20 signal_return+0x14
>                                kernel .text 0xc0100000 0xc010bb0c
> 0xc010bb24
> kdb>
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by Oracle Space Sweepstakes Want to be
> the first software developer in space?
> Enter now for the Oracle Space Sweepstakes!
> http://ads.osdn.com/?ad_idt12&alloc_id=16344&op=3Dick
> _______________________________________________
> ssic-linux-devel mailing list
> ssi...@li...
> https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel
>