[delroth]

> The systemd notification protocol could have been as simple as just writing a newline to a pipe

It basically is. libsystemd links to liblzma for other features not related to notifications.

(The protocol is that systemd passes the path to a unix socket in the `NOTIFY_SOCKET` env variable, and the daemon writes "READY=1" into it.)

[agwa]

Is that protocol documented/stable? For whatever reason, daemons are choosing to link to libsystemd instead of implementing it themselves.

It doesn't matter that libsystemd links to liblzma for other reasons. It's still in the address space of any daemon that is using libsystemd for the notification protocol.

[wickberg]

I know Golang has their own implementation of sd_notify().

For Slurm, I looked at what a PITA pulling libsystemd into our autoconf tooling would be, stumbled on the Golang implementation, and realized it's trivial to implement directly.

[tripflag]

indeed; it should be trivial in any language. Here's python: https://github.com/9001/copyparty/blob/a080759a03ef5c0a6b06c...

[cesarb]

It should be trivial in any language which has AF_UNIX. Last time I looked, Java didn't have it, so the only way was to call into non-Java code.

[fullstop]

At first I thought that this surely could not be true as of today, but it looks like it is.

There is AF_UNIX support, but only for streams and not datagrams: https://bugs.openjdk.org/browse/JDK-8297837

What an odd decision. I suppose that you could execute systemd-notify but that's a solution that I would not like.

[ptx]

It looks like the FFI (Project Panama) finally landed in Java 22, released a few days ago: https://openjdk.org/jeps/454

Unless that feature also has some weird limitation, you could probably use that to call the socket API in libc.

[cesarb]

> I suppose that you could execute systemd-notify but that's a solution that I would not like.

What I did was to use JNA to call sd_notify() in libsystemd.so.0 (when that library exists), which works but obviously does not avoid using libsystemd. I suppose I could have done all the socket calls into glibc by hand, but doing that single call into libsystemd directly was simpler (and it can be expected to exist whenever systemd is being used).

[reftel]

Then I suggest you have another look =) https://inside.java/2021/02/03/jep380-unix-domain-sockets-ch...

[fullstop]

Under Limitations: Datagram support

[KerrAvon]

Caveat is that golang is not a good enough actor to be a reliable indicator of whether this interface is supported, though. They’ll go to the metal because they can, not because it’s stable.

[pkaye]

Can me point me to the Golang implementation? Is it a standard package?

[yencabulator]

Most likely https://github.com/coreos/go-systemd

[wiml]

> libsystemd links to liblzma for other features not related to notifications

Which is pretty emblematic of systemd's primary architectural fault!

[IAmNotACellist]

systemd getting its tentacles everywhere they can squeeze is a feature, not a bug

[yrro]

The funny thing is that libsystemd _used_ to be split into several different libraries. I certainly remember libsystemd-journal (which is presumably the part of libsystemd that pulls in liblzma) being separate to libsystemd-daemon (which is the part that implements sd_notify, as used by OpenSSH [after patching by distros]).

If that split had never happened, then liblzma wouldn't have ended up being linked into sshd...

[tonyg]

Strange protocol. Why not pass a path to a file that should be `touch`d and/or written to, I wonder? Would avoid the complexity of sockets.

[Bu9818]

Services may be in a different mount namespace from systemd for sandboxing or other reasons (also means you have to worry about filesystem permissions I suppose). Passing an fd from the parent (systemd) is a nice direct channel between the processes

[NewJazz]

But systemd precisely doesn't pass an FD. If it did, you would just need to write() and close().

[Bu9818]

Yeah I was wrong about that, I confused it with socket-activation passing. The systemd-side socket is available from the process.