On Thu, 02 Dec 2021, Coly Li wrote: > On 12/2/21 12:28 AM, Benjamin Brunner wrote: > > Hi all, > > > > On 01.12.21 17:23, Coly Li wrote: > >> On 12/2/21 12:08 AM, mtkaczyk wrote: > >>> Hi Coly, > >>> > >>>> This patch changes KillMode in above listed service files from "none" > >>>> to "mixed", to follow systemd recommendation and avoid potential > >>>> unnecessary issue. > >>> What about mdmonitor.service? Should we add it there too? Now it is not > >>> defined. > >> > >> It was overlooked when I did grep KillMode. Yes, I agree > >> mdmonitor.service should have a KillMode key word as well. > >> > >> Let me post a v2 version. > >> > > JFYI, when KillMode is not set, it defaults to KillMode=control-group, > > see https://www.freedesktop.org/software/systemd/man/systemd.kill.html. > > > > Therefore, it shouldn't be necessary to explicitly add it (as long as > > control-group is working in case of mdmonitor.service, of course). > > Hi Benjamin, > > Please correct me if I am wrong, I see the difference of the KillMode is, > -- KillMode=mixed stops the processes more gentally, it kill the main > process with SIGTERM and the remaining processes with SIGKILL. > -- KillMode=control-group kills all in-cgroup processes with SIGKILL, > which I feel a bit cruel for the main process. There is no point sending SIGTERM to a process which doesn't respond to it. mdmon is the only mdadm service which handles SIGTERM. So it might make sense to uise KillMode=mixed for that. For anything else, SIGKILL via KillMode=control-group is perfectly acceptable. NeilBrown > > IMHO both method (explicit mixed mode or implicit control-group mode) > should works for the fixing issue, but I feel removing the KillMode > lines might be better as you indicated since the files can be a bit > simpler. Do I understand you correct ? > > Thanks. > > Coly Li > >
