OpenVZ Forum: Support » No pty and signal 9 VE autokill

Home » General » Support » No pty and signal 9 VE autokill

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

No pty and signal 9 VE autokill [message #40440]

Wed, 18 August 2010 17:29

PASSCE
Messages: 4
Registered: August 2010
Location: France

Junior Member

Hi everybody,
I'm running OpenVZ on 2.6.32.15 on my new server.
When I would like to start a fresh new VE, it says that is start correctly
ex:

vzctl start 126
Starting container ...
Container is unmounted
Container is mounted
Setting CPU units: 1000
Container start in progress...
root@

but it died a few seconds later.

I already tried setting CONFIG_LEGACY_PTYS in the kernel, the result is the same.

When I tried to entered just after starting the VE, I go the message that it can't find any pty dev.
If I strace vzctl enter, I see that I got a sigkill.
If I remove udev, same problem, I can enter a few seconds in the VE and then got a Signal 9.

I'm running Gentoo inside VE and in Hypervisor.

Do you have any ideas of what can I do ?

PS : Nothing in dmesg (just CTXXX start and CTXXX stop)

EDIT : open a bug on bugzilla sorry
http://bugzilla.openvz.org/show_bug.cgi?id=1613

[Updated on: Wed, 18 August 2010 17:41]

Report message to a moderator

Re: No pty and signal 9 VE autokill [message #40465 is a reply to message #40440]

Fri, 20 August 2010 10:58

maratrus
Messages: 1495
Registered: August 2007
Location: Moscow

Senior Member

Hi,

most probably the reason is some sort of incompatibility between kernel and init-scripts inside the VE.
A couple of years ago the was a thread with similar problem (please, read the thread )
To find out the reason in your particular case I can recommend you to exclude all init-scripts by turns until you catch the bad one. It might be some call that an OpenVZ kernel processes incorrectly.

Report message to a moderator

Re: No pty and signal 9 VE autokill [message #40467 is a reply to message #40440]

Fri, 20 August 2010 12:04

PASSCE
Messages: 4
Registered: August 2010
Location: France

Junior Member

Hi,
Thanks for you answer, I found that

/sbin/reboot -f

was called during the boot process, but I don't know exactly which script is calling it .. (Maybe checkroot)

Is there any possibility to see the boot process of a VE ? (ie see the initscripts loading to know which one is launching reboot and maybe the error)?
Or anything to have a /dev/vcs1 in a VE ?

Thanks

[Updated on: Fri, 20 August 2010 12:04]

Report message to a moderator