[gentoo-user] glibc upgrade: emerge frozen...

[gentoo-user] glibc upgrade: emerge frozen...

[Jarry]

Hi,

I'm trying to update to glibc 3.4.4, this is what I did:
emerge -uav gcc
gcc-config i686-pc-linux-gnu-3.4.4
source /etc/profile
emerge --oneshot -av libtool
emerge --oneshot sys-libs/libstdc++-v3
emerge -e system

Up to this point, everything was OK. As the last part I started:

emerge -e world

Emerge started updating all packages (~150), but then suddenly stopped
(after ~20 packages). No error message, just sitting here, like frozen:
...
gcc -W -Wall -Wstrict-prototypes -Wmissing-prototypes -march=athlon-xp -pipe -O2
-o .libs/ld-new ldgram.o ldlex.o lexsup.o ldlang.o mri.o ldctor.o ldmain.o
ldwrite.o ldexp.o ldemul.o ldver.o ldmisc.o ldfile.o ldcref.o eelf_i386.o
ei386linux.o eelf_x86_64.o ../bfd/.libs/libbfd.so
-L/var/tmp/portage/binulsl-2.16.1/work/build/libberty/pic -liberty
../blibiberty/libiberety.a -Wl, --rpath
-Wl,/usr/lib/binutils/i686-pc-linux-gnu/2.16.1
creating ld-new
make[3]: leaving directory '/var/tmp/portage/binutils-2.16.1/work/build/ld'
make[2]: leaving directory '/var/tmp/portage/binutils-2.16.1/work/build/ld'
make[1]: leaving directory '/var/tmp/portage/binutils-2.16.1/work/build/ld'

And nothing more. Emerge is waiting there, doing nothing. Computer is
responsive, cpu load 0%, no disc activity, nothing. Following processes
are still running (but taking no cpu-time):

emerge, sandbox, ebuild.sh, emake, make (2x), grep

What to do now? Do I have to interrupt it? What then, start again?

Jarry
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] glibc upgrade: emerge frozen...

[Jarry]

Jarry wrote:

> emerge -e world
> Emerge started updating all packages (~150), but then suddenly stopped
> (after ~20 packages). No error message, just sitting here, like frozen:
> ...

Searching for reasons why emerge so suddenly stopped I found following
entries in my /var/log/messages (time corresponds pretty well with that
frozen "emerge -e world"):

Feb  4 14:18:21 obelix kernel: ------------[ cut here ]------------
Feb  4 14:18:21 obelix kernel: CPU:    0
Feb  4 14:18:21 obelix kernel: EIP:    0060:[<c0148ee0>]    Not tainted VLI
Feb  4 14:18:21 obelix kernel: EFLAGS: 00010286   (2.6.15-gentoo-r1)
Feb  4 14:18:21 obelix kernel: EIP is at page_remove_rmap+0x30/0x40
Feb  4 14:18:21 obelix kernel: eax: ffffffff   ebx: c2bafcb0   ecx: c4eacc70
edx: c11a9000
Feb  4 14:18:21 obelix kernel: esi: 08b2c000   edi: c11a9000   ebp: c2075f34
esp: c2075e94
Feb  4 14:18:21 obelix kernel: ds: 007b   es: 007b   ss: 0068
Feb  4 14:18:21 obelix kernel: Process sh (pid: 25717, threadinfo=c2074000
task=d0b5da70)
Feb  4 14:18:21 obelix kernel: Stack: c013b508 c11a9000 c0142c66 c11a9000
08b2c000 0d480045 0d480045 fffffdae
Feb  4 14:18:21 obelix kernel:        ffffffff df51c220 08c00000 dcbfd088
08c00000 c2075f34 c0142df3 c03fbcf8
Feb  4 14:18:21 obelix kernel:        dabc9544 dcbfd088 088da000 08c00000
c2075f34 00000000 08d23fff dcbfd088
Feb  4 14:18:21 obelix kernel: Call Trace:
Feb  4 14:18:21 obelix kernel:  [<c013b508>] set_page_dirty+0x38/0x50
Feb  4 14:18:21 obelix kernel:  [<c0142c66>] zap_pte_range+0x136/0x200
Feb  4 14:18:21 obelix kernel:  [<c0142df3>] unmap_page_range+0xc3/0x130
Feb  4 14:18:21 obelix kernel:  [<c0142f4b>] unmap_vmas+0xeb/0x1a0
Feb  4 14:18:21 obelix kernel:  [<c014745b>] exit_mmap+0x6b/0xf0
Feb  4 14:18:21 obelix kernel:  [<c011613c>] mmput+0x2c/0x80
Feb  4 14:18:21 obelix kernel:  [<c011a2d3>] do_exit+0xf3/0x330
Feb  4 14:18:21 obelix kernel:  [<c011a584>] do_group_exit+0x34/0x70
Feb  4 14:18:21 obelix kernel:  [<c0102cf5>] syscall_call+0x7/0xb
Feb  4 14:18:21 obelix kernel: Code: 24 0c 83 42 08 ff 0f 98 c0 84 c0 74 1a 8b
42 08 40 78 18 c7 44 24 04 ff ff ff ff c7 04 24 10 00 00 00 e8 74 0d ff ff 83 c4
08 c3 <0f> 0b e6 01 29 42 34 c0 eb de 8d b6 00 00 00 00 55 57 56 53 83
Feb  4 14:18:21 obelix kernel:  <1>Fixing recursive fault but reboot is needed!
------------------------------------------------------------------------

I have never seen messages like these. What does it mean?
Could this be the reason why emerge suddenly stopped?

Jarry

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] glibc upgrade: emerge frozen...

[Benno Schulenberg]

Jarry wrote:
> Feb  4 14:18:21 obelix kernel: ------------[ cut here ]------------
> Feb  4 14:18:21 obelix kernel: CPU:    0
> Feb  4 14:18:21 obelix kernel: EIP:    0060:[<c0148ee0>]    Not tainted VLI
> Feb  4 14:18:21 obelix kernel: EFLAGS: 00010286   (2.6.15-gentoo-r1)
> Feb  4 14:18:21 obelix kernel: EIP is at page_remove_rmap+0x30/0x40
> [...]
>
> I have never seen messages like these. What does it mean?

Your kernel oopsed.  You stumbled over a kernel bug.

If it happens again, or is reproducible, you may want to send all that 
output between the lines to the kernel mailing list.

Benno
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] glibc upgrade: emerge frozen...

[Richard Fish]

On 2/4/06, Jarry <jarry@gmx.net> wrote:
> emerge -e world

Your question has already been answered, but I just want to point out
that an emerge -e world is not necessary.  revdep-rebuld
--library=libstdc++.so.5 will suffice.

-Richard

-- 
gentoo-user@gentoo.org mailing list