[systemd-devel] socket unit refusing connection when JOB_STOP is pending

Michal Sekletar msekleta at redhat.com
Wed May 31 08:14:44 UTC 2017


On Mon, May 29, 2017 at 5:44 PM, Lennart Poettering
<lennart at poettering.net> wrote:

> This is indeed a shortcoming in systemd's model right now: we don't
> permit a start and a stop job to be enqueued for the same unit at the
> same time. But to do what you want to do we'd need to permit that: the
> service is supposed to stop, but also temporarily start.

AFAIU, this is not exactly the case Stanislav is talking about. He
wants systemd to activate instance of a service during shutdown while
stop job is already enqueued for respective socket unit (which is
different unit). At that time there can't be any stop job enqueued for
service instance since that isn't running yet. Hence there is no
conflict between start and stop jobs. *But* this is only true when we
talk about the service instance itself. That instance can have
dependencies that are already running and are scheduled to be stopped,
and here we have the problem that Lennart is talking about.


More information about the systemd-devel mailing list