Bug#863601: systemd-notify doesn't work because of race condition

[Андрей Доценко]

Sorry for late answer. I'll try to test with old solution but it wouldn't
be sooner than in a month or two since I'm very limited of time.

вс, 10 мар. 2019 г. в 02:22, Michael Biebl :

> Control: forwarded -1 https://github.com/systemd/systemd/issues/2739
>
> On Thu, 1 Jun 2017 17:50:27 +0200 Michael Biebl  wrote:
> > On Mon, 29 May 2017 09:20:52 +0200
> > =?UTF-8?B?0JDQvdC00YDQtdC5INCU0L7RhtC10L3QutC+?= 
> > wrote:
> > > Package: systemd
> > > Version: 232
> > >
> > > I used unit-file with options below:
> > >
> > > Type=notify
> > > NotifyAccess=all
> > >
> > > Command `systemd-notify --ready` worked only once. All other times it
> did
> > > nothing. Service was killed by systemd after timeout. systemd-notify
> > > returned zero code (success). No error was reported.
> > >
> > > Workaround script I used to solve the issue:
> > >
> > > inport systemd.daemon;
> > > systemd.daemon.notify('READY=1')
> > >
> > > Another bug report might describe the same issue:
> > > https://bugzilla.redhat.com/show_bug.cgi?id=982376
> > > But in my case everything works without any sleeps.
> >
> > Can you please share the complete unit file and the script you used so
> > we have a minimal test case to reproduce the issue.
>
> I tried the test service that was linked in the bugzilla bug report but
> was not able to reproduce the issue.
> Do you still see the problem with a recent version of systemd?
>
> That said, I suppose the referenced upstream bug report is about the
> same issue and still marked as unfixed. So tentatively marking the bug
> as forwarded.
>
> Regards,
> Michael
>
> --
> Why is it that all of the instruments seeking intelligent life in the
> universe are pointed away from Earth?
>
>

-- 
Андрей Николаевич, инженер-программист.

Bug#863601: systemd-notify doesn't work because of race condition

[Michael Biebl]

Control: forwarded -1 https://github.com/systemd/systemd/issues/2739

On Thu, 1 Jun 2017 17:50:27 +0200 Michael Biebl  wrote:
> On Mon, 29 May 2017 09:20:52 +0200
> =?UTF-8?B?0JDQvdC00YDQtdC5INCU0L7RhtC10L3QutC+?= 
> wrote:
> > Package: systemd
> > Version: 232
> > 
> > I used unit-file with options below:
> > 
> > Type=notify
> > NotifyAccess=all
> > 
> > Command `systemd-notify --ready` worked only once. All other times it did
> > nothing. Service was killed by systemd after timeout. systemd-notify
> > returned zero code (success). No error was reported.
> > 
> > Workaround script I used to solve the issue:
> > 
> > inport systemd.daemon;
> > systemd.daemon.notify('READY=1')
> > 
> > Another bug report might describe the same issue:
> > https://bugzilla.redhat.com/show_bug.cgi?id=982376
> > But in my case everything works without any sleeps.
> 
> Can you please share the complete unit file and the script you used so
> we have a minimal test case to reproduce the issue.

I tried the test service that was linked in the bugzilla bug report but
was not able to reproduce the issue.
Do you still see the problem with a recent version of systemd?

That said, I suppose the referenced upstream bug report is about the
same issue and still marked as unfixed. So tentatively marking the bug
as forwarded.

Regards,
Michael

-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?



signature.asc
Description: OpenPGP digital signature

Bug#863601: systemd-notify doesn't work because of race condition

[Michael Biebl]

On Mon, 29 May 2017 09:20:52 +0200
=?UTF-8?B?0JDQvdC00YDQtdC5INCU0L7RhtC10L3QutC+?= 
wrote:
> Package: systemd
> Version: 232
> 
> I used unit-file with options below:
> 
> Type=notify
> NotifyAccess=all
> 
> Command `systemd-notify --ready` worked only once. All other times it did
> nothing. Service was killed by systemd after timeout. systemd-notify
> returned zero code (success). No error was reported.
> 
> Workaround script I used to solve the issue:
> 
> inport systemd.daemon;
> systemd.daemon.notify('READY=1')
> 
> Another bug report might describe the same issue:
> https://bugzilla.redhat.com/show_bug.cgi?id=982376
> But in my case everything works without any sleeps.

Can you please share the complete unit file and the script you used so
we have a minimal test case to reproduce the issue.

Thanks.

-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?



signature.asc
Description: OpenPGP digital signature

Bug#863601: systemd-notify doesn't work because of race condition

[Андрей Доценко]

Package: systemd
Version: 232

I used unit-file with options below:

Type=notify
NotifyAccess=all

Command `systemd-notify --ready` worked only once. All other times it did
nothing. Service was killed by systemd after timeout. systemd-notify
returned zero code (success). No error was reported.

Workaround script I used to solve the issue:

inport systemd.daemon;
systemd.daemon.notify('READY=1')

Another bug report might describe the same issue:
https://bugzilla.redhat.com/show_bug.cgi?id=982376
But in my case everything works without any sleeps.

I use Debian 9 Stretch.