On 05/19/2010 11:33 AM, Riccardo wrote:
This is a multi-part message in MIME format.
Hi,
I have a server dual xeon quad core with gentoo and qemu:
app-emulation/qemu-kvm-0.12.3-r1 USE="aio gnutls ncurses sasl vde -alsa
-bluetooth -curl -esd -fdt -hardened -kvm-trace -pulseaudio -qemu-ifup -sdl
-static"
There is a lot of vm running with ubuntu and fedora that running without problems.
I installed one vm with latest gentoo amd64, stage3 and portage.
When I try to do an emerge -e world the process freeze after a while, any time
in differente package and there aren't any errors in the logs.
This is a screenshot of the freezed vm:
http://yfrog.com/0iscre1j
top - 10:00:50 up 10:53, 2 users, load average: 0.00, 0.00, 0.00
Tasks: 130 total, 1 running, 124 sleeping, 0 stopped, 5 zombie
Cpu(s): 0.1%us, 0.0%sy, 0.0%ni, 99.9%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 8194776k total, 886508k used, 7308268k free, 225080k buffers
Swap: 2048248k total, 0k used, 2048248k free, 476956k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1 root 20 0 3760 656 552 S 0 0.0 0:00.86 init
2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd
3 root RT 0 0 0 0 S 0 0.0 0:01.78 migration/0
4 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/0
5 root RT 0 0 0 0 S 0 0.0 0:02.01 migration/1
6 root 20 0 0 0 0 S 0 0.0 0:00.01 ksoftirqd/1
7 root RT 0 0 0 0 S 0 0.0 0:02.05 migration/2
8 root 20 0 0 0 0 S 0 0.0 0:00.01 ksoftirqd/2
9 root RT 0 0 0 0 S 0 0.0 0:02.15 migration/3
10 root 20 0 0 0 0 S 0 0.0 0:00.01 ksoftirqd/3
11 root RT 0 0 0 0 S 0 0.0 0:01.69 migration/4
12 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/4
13 root RT 0 0 0 0 S 0 0.0 0:01.49 migration/5
14 root 20 0 0 0 0 S 0 0.0 0:00.00 ksoftirqd/5
15 root 20 0 0 0 0 S 0 0.0 0:00.11 events/0
16 root 20 0 0 0 0 S 0 0.0 0:00.21 events/1
ps -elf:
1 S root 776 2 0 80 0 - 0 scsi_e May18 ? 00:00:00
[scsi_eh_1]
1 S root 810 2 0 80 0 - 0 worker May18 ? 00:00:00
[kpsmoused]
1 S root 818 2 0 80 0 - 0 worker May18 ? 00:00:00
[kstriped]
1 S root 821 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpathd/0]
1 S root 822 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpathd/1]
1 S root 823 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpathd/2]
1 S root 824 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpathd/3]
1 S root 825 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpathd/4]
1 S root 826 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpathd/5]
1 S root 827 2 0 80 0 - 0 worker May18 ? 00:00:00
[kmpath_handlerd]
1 S root 828 2 0 80 0 - 0 worker May18 ? 00:00:00
[ksnapd]
1 S root 859 2 0 80 0 - 0 worker May18 ? 00:00:00
[usbhid_resumer]
1 S root 900 2 0 80 0 - 0 kjourn May18 ? 00:00:00
[jbd2/vda3-8]
1 S root 901 2 0 80 0 - 0 worker May18 ? 00:00:00
[ext4-dio-unwrit]
1 S root 902 2 0 80 0 - 0 worker May18 ? 00:00:00
[ext4-dio-unwrit]
1 S root 903 2 0 80 0 - 0 worker May18 ? 00:00:00
[ext4-dio-unwrit]
1 S root 904 2 0 80 0 - 0 worker May18 ? 00:00:00
[ext4-dio-unwrit]
1 S root 905 2 0 80 0 - 0 worker May18 ? 00:00:00
[ext4-dio-unwrit]
1 S root 906 2 0 80 0 - 0 worker May18 ? 00:00:00
[ext4-dio-unwrit]
5 S root 1005 1 0 76 -4 - 3098 poll_s May18 ? 00:00:00
/sbin/udevd --daemon
1 S root 2661 1 0 80 0 - 7492 wait May18 ? 00:00:00
supervising syslog-ng
5 S root 2662 2661 0 80 0 - 7525 poll_s May18 ? 00:00:00
/usr/sbin/syslog-ng
1 S root 3250 1 0 80 0 - 9477 poll_s May18 ? 00:00:00
/usr/sbin/sshd
1 S root 3370 1 0 80 0 - 4086 hrtime May18 ? 00:00:00
/usr/sbin/cron
4 S root 3437 1 0 80 0 - 13988 wait May18 tty1 00:00:00
/bin/login --
0 S root 3438 1 0 80 0 - 1464 n_tty_ May18 tty2 00:00:00
/sbin/agetty 38400 tty2 linux
0 S root 3439 1 0 80 0 - 1465 n_tty_ May18 tty3 00:00:00
/sbin/agetty 38400 tty3 linux
0 S root 3440 1 0 80 0 - 1464 n_tty_ May18 tty4 00:00:00
/sbin/agetty 38400 tty4 linux
0 S root 3441 1 0 80 0 - 1465 n_tty_ May18 tty5 00:00:00
/sbin/agetty 38400 tty5 linux
0 S root 3442 1 0 80 0 - 1465 n_tty_ May18 tty6 00:00:00
/sbin/agetty 38400 tty6 linux
4 S root 3443 3437 0 80 0 - 4431 n_tty_ May18 tty1 00:00:00 -bash
5 S root 3449 1 0 80 0 - 6152 poll_s May18 ? 00:01:08
SCREEN -S cs1
4 S root 3450 3449 0 80 0 - 4403 wait May18 pts/0 00:00:00
-/bin/bash
4 S root 3454 3450 0 75 -5 - 35103 n_tty_ May18 pts/0 00:00:49
/usr/bin/python2.6 /usr/bin/emerge --resume
1 S root 5222 2 0 80 0 - 0 bdi_wr 09:50 ? 00:00:00
[flush-253:0]
4 S root 5241 3250 0 80 0 - 17913 poll_s 10:00 ? 00:00:00
sshd: root@pts/2
4 S root 5244 5241 0 80 0 - 4401 wait 10:00 pts/2 00:00:00 -bash
4 R root 5251 5244 0 80 0 - 3671 - 10:02 pts/2 00:00:00 ps -elf
0 S root 12724 3454 0 75 -5 - 955 wait May18 pts/0 00:00:00
[sys-libs/glibc-2.10.1-r1] sandbox "/usr/lib64/portage/bin/ebuild.sh" compile
4 S root 12725 12724 0 75 -5 - 6431 wait May18 pts/0 00:00:00
/bin/bash /usr/lib64/portage/bin/ebuild.sh compile
5 S root 12756 12725 0 75 -5 - 6936 wait May18 pts/0 00:00:00
/bin/bash /usr/lib64/portage/bin/ebuild.sh compile
4 S root 14901 12756 0 75 -5 - 3631 wait May18 pts/0 00:00:00
make PARALLELMFLAGS=-j8
4 S root 14902 14901 0 75 -5 - 5155 wait May18 pts/0 00:00:01
make -r PARALLELMFLAGS=-j8 CVSOPTS= -C
/var/tmp/portage/sys-libs/glibc-2.10..1-r1/work/glibc-2.10.1 objdir=/var/tmp/
4 S root 30286 14902 0 75 -5 - 5784 n_tty_ May18 pts/0 00:00:03
make -j8 subdir=nptl -C nptl ..=../ subdir_lib
4 Z root 31542 30286 0 75 -5 - 0 exit 01:11 pts/0 00:00:00
[x86_64-pc-linux]<defunct>
4 Z root 31543 30286 0 75 -5 - 0 exit 01:11 pts/0 00:00:00
[x86_64-pc-linux]<defunct>
4 Z root 31544 30286 0 75 -5 - 0 exit 01:11 pts/0 00:00:00
[x86_64-pc-linux]<defunct>
4 Z root 31545 30286 0 75 -5 - 0 exit 01:11 pts/0 00:00:00
[x86_64-pc-linux]<defunct>
4 Z root 31547 30286 0 75 -5 - 0 exit 01:11 pts/0 00:00:00
[x86_64-pc-linux]<defunct>
I found the same problem with
gentoo-sources-2.6.32-r7,2.6.33,2.6.33-r1,2.6.33-2 always with -smp 2, 4 or 6.
Kernel .config:
http://pastebin.org/252943
With 2.6.32-r7 I notice the emerge will freeze after a lot of compiled
packages, while with the 2.6.33 series after 30-40 compiled packages, emerge
freeze.
If I run vm without -smp there aren't any freeze of emerge.
In the same server there are other VMs with ubuntu and -smp 6 and works
without any problems.
There is a limit max cpus for -smp ? for the total of the VMs running in the
server?
Any suggestions?
It's possible to enable a log for what service?
There are almost impossible to debug.
Try copying vmlinux out of your guest and attach with gdb when it
hangs. Then issue the command
(gdb) thread apply all backtrace
to see what the guest is doing.
--
Do not meddle in the internals of kernels, for they are subtle and quick to panic.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
