Let's not check the fd type beforehand, let's instead gracefully handle
if we get EPERM back from epoll_ctl() because the fd doesn't do epoll.
THis should be safer and more generic.
The epoll_ctl(2) man page clearly documents EPERM is being returned in
this case, hence it's safe to check for exactly that case.
Follow-up for: #28644
This is a follow-up for #28596.
I think the suggestion to use Type=exec uses too strong wording:
Type=exec has non-trivial drawbacks over Type=simple, and they deserve
to be mentioned.
Hence drop the <emphasis> and turn this around so that Type=exec is
*recommended*, but Type=simple is not expressly discouraged, because
there are plenty reasons to use it.
Add a brief discussion where Type=simple might be preferable.
Also, fix the outright unruth that Type=exec was the "simplest and
fastest", because it certainly is a lot, but not that.
For a userns root user to be able to access the credentials, both
the uid and gid of the credentials directory have to be mapped into
the userns. Currently, the credentials directory group is root, which
we obviously do not want to map in to a userns, so let's make sure
that the credentials directory and files are owned by the service
group instead, which can generally be safely mapped into the userns.
Since we use permissions mode 0600, this shouldn't cause any change
in who is able to access the credentials.
Fixes#28747
But the directories are changed from /dev/loop/by-ref/ -> /dev/disk/by-loop-ref/
and /dev/loop/by-inode/ -> /dev/disk/by-loop-inode/.
As /dev/loop/ is used by losetup command for other purpose.
See issue #28475.
This effectively reverts commits 9915cc6086,
5022fab15f, and
c0d998248e.
This update introduces the explicit Dependencies= setting, instead
of relying on implicit dependencies via alphanumerical ordering.
We also take the opportunity to rename the "final" preset to the
"system" preset, which seems like a better name.
This is supposed to be a help for compilers to apply optimizations on
functions where they can't determine whether they are const/pure on
their own. For static, local functions the compiler can do this on its
own easily however, hence the decoration with pure/const is just noise.
Let's drop it, and let the compiler to its thing better.
(Use it for exported functions, since compilers can't 'reach-over' into
other modules to determine if they are pure, except if LTO is used)
Our coding style says static variables suck except for very special
cases, i.e. things like the log level or very per-process stuff, such as
parsed version of cmdline args and such. sysusers departed from that as
one of the very few exceptions in our codebases: it keeps its
operational state in global variables.
Address that. Introduce a Context object that carries the fields that so
far have been global, and pass it around as needed.
This has the nice effect that state and configuration is clearly
separated in code, and we can very clearly see which functions mangle
state and which ones do not.
No actual codeflow changes, just refactoring.
The command is deprecated, as per NEWS of 254. Let's go one step further
and remove it from the help text and man page, so that people are not
inspired to use it at this time anymore.
This is how we usually have done this before: remove it from visibility,
but support it for a while still.
Frantisek Sumsal
Lennart Poettering
Daan De Meyer
Yu Watanabe
Luca Boccassi
Mike Yuan
Xiaotian Wu
Zbigniew Jędrzejewski-Szmek