vduong at ics.com
Wed Apr 5 12:06:51 EDT 2017
Thank you for sending me the links.
For GDP-11, we are using systemd 229, which already
included the mentioned patch codes to systemd. But it was
noted that the fix can only reduce the chances of encountering
the problem and that the problem still persists.
>From the thread, it is mentioned that the unified cgroups are being
actively worked on, but the 'perfect solution' is in the works:
Some of the commenters mentioned reverting the systemd back to a previous
version which didn't have this problem, but did have a racing condition
And a possible workaround for this problem
And it seems to still be around by the end of last year :
What is everyone's opinion on what we should do in this case?
Should we revert the systemd? Do a workaround? And if we do a workaround,
should we apply it to all Genivi apps?
Thanks in advance for your inputs!
On Wed, Apr 5, 2017 at 10:14 AM, James Thomas <james.thomas at codethink.co.uk>
> On 05/04/17 14:27, Vy Duong wrote:
> > Hi Everyone,
> > I have encountered something that may be a bug. It seems that the
> systemd is not
> > consistently sending SIGTERM when the system reboots. I had tested this
> > out by having two programs, the one that Justin had sent in the other
> email to
> > test whether SIGTERM was being received and another one to record a
> > occuring.
> Possibly the same issue as , with a fix here ?
>  https://bugzilla.redhat.com/show_bug.cgi?id=1170765
>  https://github.com/systemd/systemd/issues/317
Vy Duong, Software Engineer, Integrated Computer Solutions.
ICS - Delivering World-Class Applications for Embedded & Mobile Devices
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the genivi-projects