- This introduces ERRNO_IS_FS_WRITE_REFUSED(), and apply it where
usable.
- This makes unexpected errors in access_fd() called by binfmt_mounted()
propagated to the caller.
- Renames binfmt_mounted() to binfmt_mounted_and_writable(), as it also
checks the fs is writable.
- Voidifies one disable_binfmt() call in shutdown.c.
To implement --bind-user in systemd-vmspawn, we need a transient
version of these credentials. These are useful when the home directory
of the user is mounted into the container/vm and every trace of the user
will be (mostly) gone again when the container/vm is shut down.
Both kernel-core and kernel-uki-virt call kernel-install upon removal. Need an additional argument to avoid complete removal for both traditional kernel and UKI.
Signed-off-by: Li Tian <litian@redhat.com>
Even if we're not using --accept=, it's very useful to be able to
synchronize on systemd-socket-activate having binded to its listen
socket, so let's always send READY=1. This means the payload can't
send READY=1 anymore but it's doubtful whether that's useful in this
case in the first place.
We don't need a full blown SCSI controller just to present the main
root drive device to the VM. Let's simplify the storage stack by using
virtio-blk-pci instead.
Additionally, virtio-blk-pci is a builtin module in Arch and Fedora
which means we can do qemu direct kernel boot without needing an initrd.
hpet is an emulated clocksource that is generally discouraged in favor
of kvm-clock or tsc for virtual machines. While vmspawn's virtual machines
already use kvm-clock, leaving hpet enabled causes qemu on the host to
consume a non-trivial amount of cpu, so let's disable the hpet feature since
we're not making use of it anyway.
Otherwise, integration-test-wrapper.py will fail.
```
Traceback (most recent call last):
File "/home/runner/work/systemd/systemd/test/integration-tests/integration-test-wrapper.py", line 693, in <module>
main()
~~~~^^
File "/home/runner/work/systemd/systemd/test/integration-tests/integration-test-wrapper.py", line 677, in main
runner = os.environ['TEST_RUNNER']
~~~~~~~~~~^^^^^^^^^^^^^^^
File "<frozen os>", line 717, in __getitem__
KeyError: 'TEST_RUNNER'
```
Follow-up for c0a5801f7b.
Fixes https://github.com/systemd/systemd/issues/34780. The number in the header
is the size of the *compressed* data, so for gzip we'd read the initial part of
the decompressed data (equal to the size of the compressed data) and not find
the version string. Later on, Fedora switched to zstd compression, and there we
correctly use the number as the size of the compressed data, so we stopped
hitting the issue, but we should still fix it for older kernels.
I verified that the fix works for gzip-compressed kernels. I also made the same
change for the code for lzma compression. I'm pretty sure it is the right thing,
even though I don't have such a kernel at hand to test.
>>> ukify.Uname.scrape('/lib/modules/6.12.0-0.rc2.24.fc42.aarch64/vmlinuz')
Real-Mode Kernel Header magic not found
+ readelf --notes /lib/modules/6.12.0-0.rc2.24.fc42.aarch64/vmlinuz
readelf: Error: Not an ELF file - it has the wrong magic bytes at the start
Found uname version: 6.12.0-0.rc2.24.fc42.aarch64
closes#37602, see there for extra motivation and considered
alternatives.
On typical systems, only few services need to create SUID/SGID files.
This often is limited to the user explicitly setting suid/sgid, the
`systemd-tmpfiles*` services, and the package manager. Allowing a
default to globally restrict creation of suid/sgid files makes it easier
to apply this restriction precisely.
## testing done
- built on aarch64-linux and x86_64-linux
- ran a VM test on x86_64-linux, checking for:
- VM system boots successfully
- defaults apply (both `yes`, `no`, and undefined)
- systemd tmpfiles can set suid/sgid on journal log path
- Other services explicitly defining `RestrictSUIDSGID=no` can create
suid files
The recently added test case TEST-07-PID1.subgroup-kill.sh surfaced a
race: if we enumerate PIDs in a cgroup, and the cgroup is unlinked at
the very same time reading will result in ENODEV. We need to handle that
gracefully. Hence let's do so.
Noticed while looking at:
https://github.com/systemd/systemd/actions/runs/16143084441/job/45554929264?pr=38120
closes#37602
On typical systems, only few services need to create SUID/SGID files.
This often is limited to the user explicitly setting suid/sgid, the
`systemd-tmpfiles*` services, and the package manager. Allowing a default
to globally restrict creation of suid/sgid files makes it easier to apply
this restriction precisely.
This is very similar to 327cd2d3db:
If emergency.target is started while initrd-cleanup.service/start is queued,
the initrd-cleanup job did not get canceled. In parallel to the emergency
units, it eventually runs the service, which in turn isolates and starts
initrd-switch-root.target. This stops the emergency units and effectively
starts the initrd boot process again, which likely fails again like the
initial attempt. The system is thus stuck in a loop, never really reaching
emergency.target.
This can be triggered if a service in between initrd-parse-etc.service
and initrd.target fails.
With this conflict added, starting emergency.target automatically cancels
initrd-cleanup.service/start, avoiding the loop.
Add a new option `PrivateBPF=` to mount a private instance of bpffs.
Add also four configuration options
`BPFDelegate{Commands,Maps,Programs,Attachments}=` which set the
corresponding bpffs mount options in order to create BPF tokens:
https://lwn.net/Articles/947173/
Closes#35108.
Add a new option PrivateBPF= to mount a new instance of bpffs within a
namespace.
PrivateBPF= can be set to "no" to use the host bpffs in readonly mode
and "yes" to do a new mount.
The mount is done with the new fsopen()/fsmount() API because in future
we'll hook some commands between the two calls.
Ubuntu
Lennart Poettering
DaanDeMeyer
Li Tian
Yu Watanabe
Eisuke Kawashima
Jörg Behrmann
Christian Hesse
Zbigniew Jędrzejewski-Szmek
Antonio Alvarez Feijoo
Matteo Croce
Grimmauld
Fabian Vogt
Mike Yuan