Not detecting systemd-nspawn container

[rascalDan]

I'm new to needrestart, so accept I might be doing something wrong, for the most part it works great :-) But I have a problem where its not picking up a systemd-nspawn based container and thus I believe it thinks the service (not container) always needs restarting.

Here I'm running OpenWRT in a container, started by PaceMaker as a systemd service.

firebrand ~ # needrestart --version
needrestart 3.8 - Restart daemons after library updates.

firebrand ~ # systemctl --version 
systemd 256 (256.8)
+PAM -AUDIT -SELINUX -APPARMOR +IMA +SMACK +SECCOMP +GCRYPT -GNUTLS +OPENSSL +ACL +BLKID +CURL -ELFUTILS -FIDO2 +IDN2 -IDN -IPTC +KMOD -LIBCRYPTSETUP -LIBCRYPTSETUP_PLUGINS +LIBFDISK +PCRE2 -PWQUALITY -P11KIT -QRENCODE -TPM2 -BZIP2 +LZ4 +XZ -ZLIB +ZSTD -BPF_FRAMEWORK -XKBCOMMON +UTMP -SYSVINIT +LIBARCHIVE

firebrand ~ # machinectl 
MACHINE CLASS     SERVICE        OS      VERSION ADDRESSES 
openwrt container systemd-nspawn openwrt 19.07.7 10.10.0.1…

firebrand ~ # systemctl status systemd-nspawn@openwrt.service 
● systemd-nspawn@openwrt.service - Cluster Controlled systemd-nspawn@openwrt
     Loaded: loaded (/usr/lib/systemd/system/systemd-nspawn@.service; disabled; preset: disabled)
    Drop-In: /run/systemd/system/systemd-nspawn@openwrt.service.d
             └─50-pacemaker.conf
             /etc/systemd/system/systemd-nspawn@openwrt.service.d
             └─main.conf
     Active: active (running) since Tue 2024-12-03 12:25:25 GMT; 1 day 23h ago
 Invocation: b1b099469c274101ae28ac6d73811f5a
       Docs: man:systemd-nspawn(1)
   Main PID: 107789 (systemd-nspawn)
     Status: "Container running."
      Tasks: 12 (limit: 16384)
        CPU: 52.449s
     CGroup: /machine.slice/systemd-nspawn@openwrt.service
             ├─payload
             │ ├─107791 /sbin/procd
             │ ├─108072 /sbin/ubusd
             │ ├─108088 /sbin/urngd
             │ ├─108209 /sbin/logd -S 640
             │ ├─108248 /sbin/rpcd -s /var/run/ubus.sock -t 30
             │ ├─108357 /sbin/netifd
             │ ├─108382 /usr/sbin/crond -f -c /etc/crontabs -l 5
             │ ├─108482 /usr/sbin/pppd nodetach ipparam plusnet ifname pppoe-plusnet lcp-echo-interva>
             │ ├─108887 /usr/sbin/miniupnpd -f /var/etc/miniupnpd.conf
             │ ├─108924 /usr/sbin/dropbear -F -P /var/run/dropbear.1.pid -s -g -p 10.10.0.1:22 -p 192>
             │ └─109119 /usr/sbin/uhttpd -f -h /www -r OpenWrt -x /cgi-bin -t 60 -T 30 -k 20 -A 1 -n >
             └─supervisor
               └─107789 systemd-nspawn --quiet --keep-unit --boot --link-journal=try-guest --network->

Running in list mode:

firebrand ~ # needrestart -ra -lk -v
[main] eval /etc/needrestart/needrestart.conf
[main] eval /etc/needrestart/conf.d/pacemaker.conf
[main] needrestart v3.8
[main] running in root mode
[Core] Using UI 'NeedRestart::UI::stdio'...
[main] systemd detected
[main] #107791 uses obsolete binary /sbin/procd
# many of these
...
[main] #107789 unexpected cgroup '/machine.slice/systemd-nspawn@openwrt.service/supervisor'
...
[main] #107791 unexpected cgroup '/machine.slice/systemd-nspawn@openwrt.service/payload'
[main] trying systemctl status
[main] #108357 is systemd-nspawn@openwrt.service
...
Restarting services...
[main] using restart.d file systemd-nspawn@openwrt.service
 /etc/needrestart/restart.d/systemd-nspawn@openwrt.service

No containers need to be restarted.

No user sessions are running outdated binaries.

No VM guests are running outdated hypervisor (qemu) binaries on this host.

I presume that if it detects a container, it understands that the binary paths a relative to the container root and thus all is well, but I've not seen it do that, so I'm guessing there.

I created a script to handle restarting OpenWRT in a custom way (its currently empty) but the problem remains that it always wants to do it, not that its doing in some problematic way.

Thanks.

[rascalDan]

I just noticed there's a PR for this... even if I didn't see an existing issue.

I think just fffe4a0 is all that's needed.