Re: [Users] New behavior noted

2016-06-30 Thread jjs - mainphrame
Odd - I wonder what my corner case is here. I have 2 OVZ 7 boxes, and they both exhibit the same behavior. It would be handy to travel back in time to before the change that introduced this behavior. Sanity check: the lxc and lxd CTs are still all starting automatically My environment: OS:

Re: [Users] New behavior noted

2016-06-30 Thread jjs - mainphrame
Hi Kyrill - It sounds like this is unexpected, so I'll file a bug. Regards, Jake On Thu, Jun 30, 2016 at 4:14 PM, Kirill Kolyshkin wrote: > I suggest to file a bug saying that ONBOOT flag in ctid.conf is not > honored. > > On 30 June 2016 at 15:04, jjs - mainphrame

Re: [Users] New behavior noted

2016-06-30 Thread jjs - mainphrame
Hi Scott, Thanks for the background. On OVZ 7 at least, this seemed to be a change in default behavior. Perhaps just a consequence of starting my OVZ 7 involvement with the early pre-releases. Jake On Thu, Jun 30, 2016 at 9:56 AM, Scott Dowdle wrote: > Greetings, > >

Re: [Users] New behavior noted

2016-06-30 Thread Scott Dowdle
Greetings, - Original Message - > Yes, the kernel version told me that your first results were with > openvz legacy. But my question was more about your "second round of > tests" - > > Do you have any OVZ 7 hosts to look at? Yes, I do. But I don't think I currently have any containers

Re: [Users] New behavior noted

2016-06-30 Thread jjs - mainphrame
Hi Scott, Yes, the kernel version told me that your first results were with openvz legacy. But my question was more about your "second round of tests" - Do you have any OVZ 7 hosts to look at? Regards, Jake On Thu, Jun 30, 2016 at 2:16 PM, Scott Dowdle wrote: >

Re: [Users] New behavior noted

2016-06-30 Thread Scott Dowdle
Greetings, - Original Message - > The containers we're expecting to start on boot are precisely those > which have "ONBOOT="yes" in their config files. And up until a few > updates ago, they always started on boot. Well, that is news. I haven't tried to verify yet. > That said, this

Re: [Users] New behavior noted

2016-06-30 Thread Scott Dowdle
Greeting, - Original Message - > Just to be clear, you're testing with legacy openvz, right? Right. The kernel version should tell you. 2.6.32-x = OpenVZ Legacy 3.10.x = Virtuozzo / OpenVZ 7 TYL, -- Scott Dowdle 704 Church Street Belgrade, MT 59714 (406)388-0827 [home] (406)994-3931

Re: [Users] New behavior noted

2016-06-30 Thread jjs - mainphrame
Ah, I'm on OVZ 7 - sorry if I did not make that clear. vzkernel-3.10.0-327.18.2.vz7.14.22.x86_64 vzctl-7.0.105-1.vz7.x86_64 Jake On Thu, Jun 30, 2016 at 10:43 AM, Scott Dowdle wrote: > Greetings, > > - Original Message - > > The containers we're expecting

Re: [Users] New behavior noted

2016-06-30 Thread Scott Dowdle
Greetings, - Original Message - > The containers we're expecting to start on boot are precisely those > which have "ONBOOT="yes" in their config files. And up until a few > updates ago, they always started on boot. I tried a second round of tests... this time with two containers (one

Re: [Users] New behavior noted

2016-06-30 Thread Scott Dowdle
Greetings, - Original Message - > The containers we're expecting to start on boot are precisely those > which have "ONBOOT="yes" in their config files. And up until a few > updates ago, they always started on boot. I just tried on one of my hosts and I can not duplicate your claim. Here

Re: [Users] New behavior noted

2016-06-30 Thread jjs - mainphrame
Hi Scott, The containers we're expecting to start on boot are precisely those which have "ONBOOT="yes" in their config files. And up until a few updates ago, they always started on boot. That said, this new behavior of suspending running containers on shutdown seems useful though. Joe On Thu,

Re: [Users] New behavior noted

2016-06-30 Thread Scott Dowdle
Greetings, - Original Message - > I've noticed a new behavior from the openvz hosts regarding shutdown > and startup. We've long preferred to migrate the really important > CTs to another host before we rebooting for e.g. a kernel upgrade. > Other CTs are redundant or are not continuously