It might be something that lxc-start needs to check based upon some
config parameter. I dont know if there is a way of getting info from the
ABI or if its just a matter of checking the kernel version and/or
/proc/cmdline.
Such a config parameter could then be placed in the centos6 template and
others as needed. At start up, lxc-start could then fail gracefully with
a helpful message?
On 06/12/16 19:09, Evgeni Golov wrote:
Hey,
On Tue, Dec 06, 2016 at 05:46:58PM +1100, Dean Hamstead wrote:
Yes, Solution was to use "vsyscall=emulate" on the host kernel command line
(ie on my laptop).
Exactly the same issue as:
https://github.com/CentOS/sig-cloud-instance-images/issues/62
So I am guessing the issue was actually the kernel upgrading.
See also: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1018415
Yeah, I knew what it was as soon as you posted the dmesg output, but was
on mobile, so was not in the mood to write a proper answer, sorry.
Glad you have a solution here. Let's see if we can talk the kernel ppl
into re-thinking this. Or we need a big fat warning somewhere.
Thanks for debugging
