On Thursday 2012-02-09 03:44, Lennart Poettering wrote:
>Heya,
>
>http://www.freedesktop.org/software/systemd/systemd-41.tar.xz
Can I propose that the version number in AC_INIT and supplementary
documentation like NEWS is not bumped until the release is actually
made, and that the commit that doe
Heya,
http://www.freedesktop.org/software/systemd/systemd-41.tar.xz
Changes:
http://cgit.freedesktop.org/systemd/systemd/tree/NEWS
Make sure to read the NEWS, otherwise you might break your system, since
your /sbin/init symlink might point to nowhere.
Lennart
--
Lennart Poettering - Red Hat,
On Wed, 08.02.12 21:52, Tom Gundersen ([email protected]) wrote:
Thanks a lot! Both commited.
I must say I like it if people just start me patches for stuff we list
on the TODO list! ;-)
> ---
> Makefile.am|6 +++-
> TODO |2 -
> src/modules-load.c | 70 ++
---
Makefile.am|6 +++-
TODO |2 -
src/modules-load.c | 70 ---
3 files changed, 38 insertions(+), 40 deletions(-)
diff --git a/Makefile.am b/Makefile.am
index 2856fab..fbf2358 100644
--- a/Makefile.am
+++ b/Makefile.
---
Makefile.am |6 +++-
TODO |2 +-
configure.ac |2 +
src/kmod-setup.c | 68 +++--
4 files changed, 47 insertions(+), 31 deletions(-)
diff --git a/Makefile.am b/Makefile.am
index e4e3510..2856fab 100644
--- a/Mak
On Wed, 08.02.12 10:10, Michael Olbrich ([email protected]) wrote:
> Signed-off-by: Michael Olbrich
> ---
> changes in v2:
> - Directly set ratelimit.interval/ratelimit.burst
>There are some issues with that:
>The names are now inconsistent: ratelimit vs. start_limit, maybe we
>
On Wed, 08.02.12 10:10, Michael Olbrich ([email protected]) wrote:
Applied, thanks!
> ---
> src/service.c |6 ++
> 1 files changed, 6 insertions(+), 0 deletions(-)
>
> diff --git a/src/service.c b/src/service.c
> index 1631595..a190a73 100644
> --- a/src/service.c
> +++ b/src/ser
On Wed, 08.02.12 10:10, Michael Olbrich ([email protected]) wrote:
Thanks,
applied.
> +
> sd_notify3
> +regularly with "WATCHDOG=1". If the time
> +between two such calls is larger than
> +
On Tue, 07.02.12 20:43, Tom Gundersen ([email protected]) wrote:
Heya,
> On Tue, Feb 7, 2012 at 8:15 PM, Phil Miller wrote:
> > Since it get's more and more complex to keep up our patch to add our distro
> > to
> > systemd it would be nice to have it added upstream. Attached your current
> > patch f
Am 08.02.2012 17:59, schrieb Bill Nottingham:
> Reindl Harald ([email protected]) said:
>> a kernel boot-param would be nice but until know i did not
>> find any working one :-(
>
> ipv6.disable_ipv6=1
hmm should i file a bugreport for the kernel?
i tried "disable_ipv6=1" today with no su
Reindl Harald ([email protected]) said:
> a kernel boot-param would be nice but until know i did not
> find any working one :-(
ipv6.disable_ipv6=1
Bill
___
systemd-devel mailing list
[email protected]
http://lists.freedesktop.or
On Wed, 08.02.12 17:47, Reindl Harald ([email protected]) wrote:
>
> Am 08.02.2012 17:34, schrieb Bill Nottingham:
> > Reindl Harald ([email protected]) said:
> >> not network manager
> >> classical configuration
> >>
> >> net.ipv6.conf.all.disable_ipv6 = 1
> >
> > Your configuration
Am 08.02.2012 17:34, schrieb Bill Nottingham:
> Reindl Harald ([email protected]) said:
>> not network manager
>> classical configuration
>>
>> net.ipv6.conf.all.disable_ipv6 = 1
>
> Your configuration is wrong, I believe. The rule at udev time is:
>
> SUBSYSTEM=="net", KERNEL!="lo", RUN+=
Reindl Harald ([email protected]) said:
> not network manager
> classical configuration
>
> net.ipv6.conf.all.disable_ipv6 = 1
Your configuration is wrong, I believe. The rule at udev time is:
SUBSYSTEM=="net", KERNEL!="lo", RUN+="/lib/systemd/systemd-sysctl \
--prefix=/proc/sys/net/ipv4/
Am 08.02.2012 17:06, schrieb Lennart Poettering:
> On Wed, 08.02.12 17:00, Reindl Harald ([email protected]) wrote:
>
>>> Hmm, so we actually apply sysctls very early in the boot, and we do not
>>> spawn normal services before this finished. There's one exception
>>> however: settings on ne
On Wed, 08.02.12 17:00, Reindl Harald ([email protected]) wrote:
> > Hmm, so we actually apply sysctls very early in the boot, and we do not
> > spawn normal services before this finished. There's one exception
> > however: settings on network interfaces are applied as the network
> > interfa
Am 08.02.2012 16:40, schrieb Lennart Poettering:
> On Wed, 08.02.12 16:07, Reindl Harald ([email protected]) wrote:
>> can someone please take a look at this
>>
>> in myopinion it is not optimal starting services which
>> possibly read settings controlled by "systcl.conf"
>> and after that t
On Wed, 08.02.12 16:07, Reindl Harald ([email protected]) wrote:
> hi
>
> can someone please take a look at this
>
> in myopinion it is not optimal starting services which
> possibly read settings controlled by "systcl.conf"
> and after that the local settings get applied
>
> sample below:
On Tue, 07.02.12 22:39, Benjamin Franzke ([email protected]) wrote:
> Fixes segfault in systemd-logind, triggered by:
> systemd-loginctl show-session $XDG_SESSION_ID.
>
> Bug introduced by d200735e13c52dcfe36c0e066f9f6c2fbfb85a9c,
> so only systemd v39 is affected.
Applied!
Thanks!
On Tue, 07.02.12 15:34, William Douglas ([email protected]) wrote:
>
> >From b165f17de89c1d61d640234e6516d9f517aa25b2 Mon Sep 17 00:00:00 2001
> From: William Douglas
> Date: Tue, 7 Feb 2012 12:31:20 -0800
> Subject: [PATCH] systemd: Add systemd.setenv for /proc/cmdline parsing.
>
> Che
hi
can someone please take a look at this
in myopinion it is not optimal starting services which
possibly read settings controlled by "systcl.conf"
and after that the local settings get applied
sample below:
after boot smbd is listening on ipv6, sysctl disables this,
but seems to do it after sm
Signed-off-by: Michael Olbrich
---
changes in v2:
- Directly set ratelimit.interval/ratelimit.burst
There are some issues with that:
The names are now inconsistent: ratelimit vs. start_limit, maybe we
should call the variable ratelimit_action?
setting StartLimitBurst=0 will trigger an
---
src/service.c |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/src/service.c b/src/service.c
index 1631595..a190a73 100644
--- a/src/service.c
+++ b/src/service.c
@@ -1735,6 +1735,12 @@ static int service_spawn(
goto fail;
---
changes in v2:
- the watchdog is trigger for the first time when the service starts.
Is service_enter_start_post() the right place for this?
- Call it SERVICE_FAILURE_WATCHDOG and define corresponding string
- Watchdog timer handling reworked to catch timeouts during
daemon-reexec/daem
Hi,
Here is the next version of my watchdog patches. I've documented the
changes in each patch. Patch 2/3 is unchanged.
Regards,
Michael
Michael Olbrich (3):
introduce WatchdogSec and hook up the watchdog with the existing
failure logic
set WATCHDOG_USEC environmen variable
service: ad
25 matches
Mail list logo
