Using MAINPID breaks systemd's exit detection, as it stops watching the
original pid, but is unable to watch the new pid as it is not a child
of systemd itself. The best we can do is just notify when execing the child
process.
We also need to consolidate forking into a sigle place so that we don't
end up with multiple levels of child processes if both redirecting log
output and reaping child processes.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
Before this change, Ubuntu provisioning script used the package list
from Alpine. But a lot of packages from that list have name mismatches.
Biggest example - `docker`, which has nothing to do with Docker as a
container engine in Ubuntu (the proper package is `docker.io`). Let's
just keep the full list for Ubuntu in its own provisioning script.
Signed-off-by: Michal Rostecki <mrostecki@opensuse.org>
This change adds support for libvirt provider. Please note that libvirt
doesn't support the "virtualbox" mount type, so an another MOUNT_TYPE
has to be provided while using libvirt. 9p works well.
Signed-off-by: Michal Rostecki <mrostecki@opensuse.org>
On many Linux distributions (including openSUSE) the OS env variable is
already used with value "Linux". This results in weird errors like:
==> .1: Box 'generic/Linux' could not be found. Attempting to find and install...
To avoid the conflict, use the more neutral and less likely to be taken
DISTRO variable.
Signed-off-by: Michal Rostecki <mrostecki@opensuse.org>
* Make sure there are no duplicates in etcd member list
Signed-off-by: galal-hussein <hussein.galal.ahmed.11@gmail.com>
* fix node names with hyphens
Signed-off-by: galal-hussein <hussein.galal.ahmed.11@gmail.com>
* use full server name for etcd node name
Signed-off-by: galal-hussein <hussein.galal.ahmed.11@gmail.com>
Works around issue with Job controller not tracking job pods that
are in CrashloopBackoff during upgrade from 1.21 to 1.22.
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
Problem:
A false-negative in check-config.sh for cgroups v2 systems was fixed but the
commit introduced a regression based on a small assumption that content of
/sys/fs/cgroup/cgroup.controllers would have the same format as the content
of /proc/self/cgroup. It doesn't.
Solution:
This just tweaks the regex to count occurrences of either cgroup
subsystem-names on each line (as occurs in the sysfs pseudo-file), or those
names with colons either side (as occurs in the procfs pseudo-file).
Signed-off-by: Rowan Thorpe <rowan@rowanthorpe.com>
* Update the default containerd config template with support for adding extra container runtimes. Add logic to discover nvidia container runtimes installed via the the gpu operator or package manager.
Signed-off-by: Joe Kralicky <joe.kralicky@suse.com>
