[arch-general] Strange (sd-pam) process
Hi `systemd --user` has a child `(sd-pam)`. It's certainly related to pam, but why has it such a strange name ? Thanks
On 13 June 2014 10:21, Yamakaky <yamakaky@yamaworld.fr> wrote:
Hi
`systemd --user` has a child `(sd-pam)`. It's certainly related to pam, but why has it such a strange name ?
Thanks
Hi, It's so that it looks pretty in ps output [1]. Cheers, WorMzy [1] http://cgit.freedesktop.org/systemd/systemd/tree/src/core/execute.c#n809
Le 13/06/2014 13:50, WorMzy Tykashi a écrit :
On 13 June 2014 10:21, Yamakaky <yamakaky@yamaworld.fr> wrote:
Hi
`systemd --user` has a child `(sd-pam)`. It's certainly related to pam, but why has it such a strange name ?
Thanks
Hi,
It's so that it looks pretty in ps output [1].
Cheers,
WorMzy
[1] http://cgit.freedesktop.org/systemd/systemd/tree/src/core/execute.c#n809
Oh, why not ^^ And what is it's purpose ?
On 13 June 2014 12:57, Yamakaky <yamakaky@yamaworld.fr> wrote:
And what is it's purpose ?
Hi, 's:#n809:#n806:' on my earlier link. ;) I can't tell you any more than is written there. If it doesn't answer your question, you may want to ask on the systemd mailing list [2], they're more likely to have the answers you seek. Cheers, WorMzy [2] http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Le 13/06/2014 14:08, WorMzy Tykashi a écrit :
On 13 June 2014 12:57, Yamakaky <yamakaky@yamaworld.fr> wrote:
And what is it's purpose ?
Hi,
's:#n809:#n806:' on my earlier link. ;)
I can't tell you any more than is written there. If it doesn't answer your question, you may want to ask on the systemd mailing list [2], they're more likely to have the answers you seek.
Cheers,
WorMzy
[2] http://lists.freedesktop.org/mailman/listinfo/systemd-devel
... I read the code upper, but not this comment ^^ It's strange it's not `systemd --user` who do this job. Thanks for you quick and precise answer !
`systemd --user` has a child `(sd-pam)`. It's certainly related to pam, but why has it such a strange name ?
And what is it's purpose ?
I think it's intended to guarantee that pam_close_session() is closed when its parent "systemd --user" exits. However, on my system it seems that on user logout the main systemd (pid 1) usually SIGKILLs it before it has the chance. I think that's a bug, because systemd is supposed to wait TimeoutStopSec between sending SIGTERM and SIGKILL. Carl
participants (3)
-
Carl Schaefer
-
WorMzy Tykashi
-
Yamakaky