poststop: reboot: wait for lxc to exit before rebooting

otherwise it'll leak cgroup directories...

Note that we need to escape the lxc@.service context (by
entering a new scope) as well as close our ties to the lxc
monitor (the stdout pipe), otherwise this never finishes
properly.

diff --git a/src/lxc-pve-poststop-hook b/src/lxc-pve-poststop-hook
index a4302d6bbf13f3efc4cbd01c79d2032c9063d0d5..e3a8956ed1b644b8e3222924dfd343650af15644 100755 (executable)
--- a/src/lxc-pve-poststop-hook
+++ b/src/lxc-pve-poststop-hook
@@ -9,6 +9,7 @@ exit 0 if $ENV{LXC_NAME} && $ENV{LXC_NAME} !~ /^\d+$/;
 
 use POSIX;
 use File::Path;
+use IO::Pipe;
 
 use PVE::SafeSyslog;
 use PVE::Tools;
@@ -81,9 +82,39 @@ __PACKAGE__->register_method ({
            # that we must not block because we're part of the service cgroup
            # systemd waits for to die before issuing the new lxc-start command.
            PVE::LXC::update_lxc_config($vmid, $conf);
-           PVE::Tools::run_command(['systemctl', '--no-block', 'restart', "lxc\@$vmid"]);
+           $SIG{HUP} = 'IGNORE';
+           # Synchronization pipe
+           my $scope = IO::Pipe->new() or die "pipe failed: $!\n";
+           my $pid = fork();
+           die "fork failed: $!\n" if !defined($pid);
+           if (!$pid) {
+               $scope->writer();
+               # We inherit a pipe from LXC, replace it with stderr otherwise
+               # lxc will keep waiting for us...
+               POSIX::dup2(2, 1);
+               POSIX::setsid();
+               eval {
+                   # Change scope otherwise we're part of lxc@.service and then
+                   # if lxc finishes cleaning up before we restart it systemd
+                   # might clean US up as well (read: kill us) => race
+                   PVE::Tools::enter_systemd_scope("restart-$vmid", "Restarter for Proxmox VE CT $vmid",
+                       Slice => 'lxc.slice',
+                       KillMode => 'none');
+                   # Tell the main stop hook we're "in the clear":
+                   close($scope);
+                   # Wait for the container to clean up everything...
+                   PVE::Tools::run_command(['lxc-wait', "--name=$vmid", '--state=STOPPED']);
+                   # ... before finally triggering a restart:
+                   PVE::Tools::run_command(['systemctl', 'restart', "lxc\@$vmid"]);
+               };
+               warn "$@" if $@;
+               POSIX::_exit(0);
+           }
+           # Wait for the restarter scope to make sure systemd doesn't kill it before it started...
+           $scope->reader();
+           <$scope>;
            # cause lxc to stop instead of rebooting
-           POSIX::_exit(1);
+           exit(1);
        }
 
        return undef;