Init is not very talkative after process start even when verbose. Suspending output here is pointless and does more harm than good.
Signed-off-by: Ophestra <cat@gensokyo.uk>
The SIGTERM signal is delivered in many other cases and can lead to strange behaviour. The unconditional resume of the logger also causes strange behaviour in the cancellation forwarding path. This change also passes through additional signals.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This does not yet have full coverage. Test cases covering failsafe paths and error injection will be added eventually.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This helps with other areas of the test suite as they're all based on hst.Template. This also helps contributors understand the behaviour of internal/app as hst.Template covers almost every aspect of it.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This is quite a simple piece of code, this many test cases is excessive and wastes time in the integration vm.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This is useful as package "syscall" does not provide such a wrapper. This change also improves error handling to fully conform to the manpage.
Signed-off-by: Ophestra <cat@gensokyo.uk>
Goroutine scheduling is not helpful in the setuid wrapper, it is not particularly harmful but lock here anyway.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This is useful for a user-facing representation other than JSON. This also gets rid of the ugly, outdated flags string builder in cmd/hakurei.
Signed-off-by: Ophestra <cat@gensokyo.uk>
The hst package only deals with data serialisation, however since many parts of hakurei make use of C libraries in some way it can be easy to inadvertently depend on cgo.
Signed-off-by: Ophestra <cat@gensokyo.uk>
There's quite a bit of code duplication here, but since this is already quite simple it is best to leave it as is for now.
Signed-off-by: Ophestra <cat@gensokyo.uk>
The loadFile behaviour does not guarantee the buffer to be zeroed or not clobbered if an error is returned, but for the current implementation it is good to check.
Signed-off-by: Ophestra <cat@gensokyo.uk>
These are now considered stable, defined behaviour and can be used by external programs to determine shim outcome.
Signed-off-by: Ophestra <cat@gensokyo.uk>
When using two channels it is possible for the other case to be reached before all pending winfo are consumed, causing incorrect reporting.
Signed-off-by: Ophestra <cat@gensokyo.uk>
The booleans are getting packed into a single field. This requires non-insignificant amount of code for JSON serialisation to stay compatible.
Signed-off-by: Ophestra <cat@gensokyo.uk>
This is not done very cleanly, however this op is pending removal for the in-process dbus proxy so not worth spending too much effort here. As long as it checks all paths it is good enough.
Signed-off-by: Ophestra <cat@gensokyo.uk>
