FS#58892 - [systemd] nspawn --keep-unit failed to attach to systemd cgroup
Attached to Project:
Arch Linux
Opened by bartus (bartoszek) - Wednesday, 06 June 2018, 16:43 GMT
Last edited by freswa (frederik) - Thursday, 05 March 2020, 21:53 GMT
Opened by bartus (bartoszek) - Wednesday, 06 June 2018, 16:43 GMT
Last edited by freswa (frederik) - Thursday, 05 March 2020, 21:53 GMT
|
Details
Description:
When building packages within chroot using arch-nspawn occasionally build renders systemd nspawn container unusable. Every consecutive invocation of systemd-nspawn crashes with this type of error: Failed to attach 8185 to compat systemd cgroup /user.slice/user-1000.slice/session-c1.scope/payload: No such file or directory Failed to attach 8157 to compat systemd cgroup /user.slice/user-1000.slice/session-c1.scope/supervisor: No such file or directory Failed to chown() cgroup /sys/fs/cgroup/systemd/user.slice/user-1000.slice/session-c1.scope/payload: No such file or directory Steps to reproduce: $ aursync -c -C /tmp/ -T ananicy-git // first time works fine $ aursync -c -C /tmp/ -T ananicy-git // this time it crashes Add1. The problem was mentioned on mailing list: https://lists.archlinux.org/pipermail/arch-dev-public/2018-January/029105.html Add2. `payload` and `supervisor` directorys are located in /sys/fs/cgroups/undefinded instead of /sys/fs/cgroups/systemd ... $ tree /sys/fs/cgroup . ├── blkio ... ├── cpu -> cpu,cpuacct ├── cpuacct -> cpu,cpuacct ├── cpu,cpuacct ... ├── cpuset ... ├── devices ... ├── freezer ... ├── hugetlb ... ├── memory ... ├── net_cls -> net_cls,net_prio ├── net_cls,net_prio ... ├── net_prio -> net_cls,net_prio ├── perf_event ... ├── pids ... ├── rdma ... ├── systemd │ ├── cgroup.clone_children │ ├── cgroup.procs │ ├── cgroup.sane_behavior │ ├── init.scope ... │ ├── notify_on_release │ ├── release_agent │ ├── system.slice ... │ ├── tasks │ └── user.slice │ ├── cgroup.clone_children │ ├── cgroup.procs │ ├── notify_on_release │ ├── tasks │ └── user-1000.slice │ ├── cgroup.clone_children │ ├── cgroup.procs │ ├── notify_on_release │ ├── session-c1.scope │ │ ├── cgroup.clone_children │ │ ├── cgroup.procs │ │ ├── notify_on_release │ │ └── tasks │ ├── tasks │ └── user@1000.service ... └── unified ├── cgroup.controllers ├── cgroup.max.depth ├── cgroup.max.descendants ├── cgroup.procs ├── cgroup.stat ├── cgroup.subtree_control ├── cgroup.threads ├── init.scope │ ├── cgroup.controllers │ ├── cgroup.events │ ├── cgroup.max.depth │ ├── cgroup.max.descendants │ ├── cgroup.procs │ ├── cgroup.stat │ ├── cgroup.subtree_control │ ├── cgroup.threads │ ├── cgroup.type │ └── cpu.stat ├── system.slice ... └── user.slice ├── cgroup.controllers ├── cgroup.events ├── cgroup.max.depth ├── cgroup.max.descendants ├── cgroup.procs ├── cgroup.stat ├── cgroup.subtree_control ├── cgroup.threads ├── cgroup.type ├── cpu.stat └── user-1000.slice ├── cgroup.controllers ├── cgroup.events ├── cgroup.max.depth ├── cgroup.max.descendants ├── cgroup.procs ├── cgroup.stat ├── cgroup.subtree_control ├── cgroup.threads ├── cgroup.type ├── cpu.stat ├── session-c1.scope │ ├── cgroup.controllers │ ├── cgroup.events │ ├── cgroup.max.depth │ ├── cgroup.max.descendants │ ├── cgroup.procs │ ├── cgroup.stat │ ├── cgroup.subtree_control │ ├── cgroup.threads │ ├── cgroup.type │ ├── cpu.stat │ ├── payload │ │ ├── cgroup.controllers │ │ ├── cgroup.events │ │ ├── cgroup.max.depth │ │ ├── cgroup.max.descendants │ │ ├── cgroup.procs │ │ ├── cgroup.stat │ │ ├── cgroup.subtree_control │ │ ├── cgroup.threads │ │ ├── cgroup.type │ │ └── cpu.stat │ └── supervisor │ ├── cgroup.controllers │ ├── cgroup.events │ ├── cgroup.max.depth │ ├── cgroup.max.descendants │ ├── cgroup.procs │ ├── cgroup.stat │ ├── cgroup.subtree_control │ ├── cgroup.threads │ ├── cgroup.type │ └── cpu.stat └── user@1000.service ... 588 directories, 8918 files |
This task depends upon
Closed by freswa (frederik)
Thursday, 05 March 2020, 21:53 GMT
Reason for closing: Fixed
Additional comments about closing: Reporter: I haven't encounter this problem since ~1yer - think it was resolved with `systemd` update.
Thursday, 05 March 2020, 21:53 GMT
Reason for closing: Fixed
Additional comments about closing: Reporter: I haven't encounter this problem since ~1yer - think it was resolved with `systemd` update.
Failed to attach 5172 to compat systemd cgroup /user.slice/user-1000.slice/session-1.scope/payload: No such file or directory
Failed to attach 5151 to compat systemd cgroup /user.slice/user-1000.slice/session-1.scope/supervisor: No such file or directory
Failed to chown() cgroup /sys/fs/cgroup/systemd/user.slice/user-1000.slice/session-1.scope/payload: No such file or directory
==> ERROR: Aborting...
This worked a couple of days ago, but has stopped working on two separate systems. I have tried re-logging in, and also rebooted both.
Thanks to sangy for the tip about removing "--keep-unit".
FS#55082, right? Is --keep-unit even supposed to be used like this?