CAcert System Admins discussion list

[Bas van den Dikkenberg <bas AT Dikkenberg.net>]

Just my two cent's ,

 

 

A sysadmin must alway have abilety to reboot server/host/container.

 

I know restarting a service is beter most of the times but, not always the fastestes way.

So i think it should be fixed that a host/container can be rebooted

 

With kind regards,

 

Bas van den Dikkenberg

 

Van: Michael Tänzer
Verzonden: ‎23‎-‎sep‎-‎13 ‎21‎:‎00
Aan: cacert-sysadm AT lists.cacert.org
Onderwerp: Re: Bugs.c.o not rebooting properly

 

Hi,

On 23.09.2013 20:43, Mario Lipinski wrote:
> Am 23.09.2013 17:32, schrieb Wytze van der Raay:
>> but the vm is still not reachable ... and I don't know why.
>> I guess that Mario or Jan need to look at this ...
>
> First, you should not reboot or shutdown the lxc containers. It is
> sufficient to restart the services.

Ok, I didn't know about that, I thought that they behave just like a
normal virtual machine. I wanted to make sure that every service was
running with the updated binaries and libraries. How do I restart all
services? Is there an easy way or do I have to restart each running
service manually? I guess changing runlevels is not a good option as it
might leave me without any running services => no SSH access.

> Second, there seems to be a problem with Wheezy and setting the default
> gateway. I added this to /etc/rc.local which should bring up the network
> properly after boot.

Can we build a knowledge base of some sorts for these little things you
have to know if you're administering a host on the CAcert infrastructure
so I can have a look once in a while if I do any special operations on
the services like upgrades? This should contain known issues with some
component (like the wheezy thing), curiousities of the environment (e.g.
that you should not shut it down, what you should do instead etc).

--
Thanks,
Michael Tänzer