31b7ddd122
fst: improve config
...
The config struct more or less "grew" to what it is today. This change moves things around to make more sense and fixes nonsensical comments describing obsolete behaviour.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-04-13 03:30:19 +09:00
e9a7cd526f
app: improve shim process management
...
This ensures a signal gets delivered to the process instead of relying on parent death behaviour.
SIGCONT was chosen as it is the only signal an unprivileged process is allowed to send to processes with different credentials.
A custom signal handler is installed because the Go runtime does not expose signal information other than which signal was received, and shim must check pid to ensure reasonable behaviour.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-04-07 03:55:17 +09:00
8aeb06f53c
app: share path setup on demand
...
This removes the unnecessary creation and destruction of share paths when none of the enablements making use of them are set.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-04-01 00:47:32 +09:00
297b444dfb
test: separate app and sandbox
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-30 22:09:46 +09:00
ff3cfbb437
test/sandbox: check seccomp outcome
...
This is as ugly as it is because it has to have CAP_SYS_ADMIN and not be in seccomp mode.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-28 02:24:27 +09:00
d613257841
sandbox/init: clear inheritable set
...
Inheritable should not be able to affect anything regardless of its value, due to no_new_privs.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-26 07:46:13 +09:00
52fcc48ac1
sandbox/init: drop capabilities
...
During development the syscall filter caused me to make an incorrect assumption about SysProcAttr.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-26 06:32:08 +09:00
4836d570ae
test: raise long timeout to 15 seconds
...
The race detector really slows down container tooling.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-26 01:59:05 +09:00
ee51320abf
test: check revert type selection
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-25 04:37:58 +09:00
5c4058d5ac
app: run in native sandbox
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-25 01:52:49 +09:00
61d86c5e10
test/sandbox: fix stdout tty check
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-24 16:23:50 +09:00
806ce18c0a
test/sandbox: check mapuid outcome
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-23 17:56:07 +09:00
b71d2bf534
test/sandbox: check tty outcome
...
This makes no difference currently but has different behaviour in the native sandbox.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-23 17:28:57 +09:00
d2c329bcea
test: format path aid offsets
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-23 17:21:14 +09:00
75e0c5d406
test/sandbox: parse full test case
...
This makes declaring multiple tests much cleaner.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-23 14:53:50 +09:00
5d3c8dcc92
test: raise timeout
...
Native container tooling is severely slowed down by race detector. Raise timeout so it reliably completes.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-16 23:51:17 +09:00
beb3918809
test: run go test under regular user
...
By default test vm commands run as root, this causes buildFHSEnv bwrap to cover some parts of /proc, making it impossible to mount proc in a mount namespace created under it. Running as a regular user gets around this issue.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-13 20:56:32 +09:00
2871426df2
test: print output of failed test
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-13 16:40:15 +09:00
61e58aa14d
helper/proc: expose setup file
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-03-09 17:22:31 +09:00
d8e9d71f87
test/sandbox: check mount outcome
...
Do this at the beginning of the test for early failure.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-02-28 15:56:15 +09:00
5de28800ad
test: verify fsu ppid check
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-02-26 16:51:57 +09:00
8e50293ab7
test: remove sway process check
...
This eliminates the race where systemd restarts sway too quick.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-02-26 13:52:44 +09:00
c62689e17f
nix: interrupt via tty
...
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-02-25 18:20:47 +09:00
b4549c72be
nix: verify silent signal exit
...
This catches errors in the cleanup process initiated by a signal.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-02-25 01:22:16 +09:00
b6af8caffe
nix: clean up directory structure
...
Tests for fpkg is going to be in ./cmd/fpkg, so this central tests directory is no longer necessary.
Signed-off-by: Ophestra <cat@gensokyo.uk >
2025-02-23 18:48:01 +09:00
