| Message ID | f7tint4jc2a.fsf@redhat.com |
|---|---|
| State | Accepted |
| Delegated to: | Russell Bryant |
| Headers | show |
On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole <aconole@redhat.com> wrote: > During testing it was found that systemd would consider the openvswitch > service as a part of networking component, but the dependent services of > ovs-vswitchd and ovsdb-server were not likewise considered. This leads > to some strange race conditions, observed when using NFS over TCP, while > shutting down systems. > > Fixes: 84ad12083491 ("rhel: Improved Systemd Integration") > Co-authored-by: Flavio Leitner <fbl@redhat.com> > Signed-off-by: Aaron Conole <aconole@redhat.com> > This looks fine to me. Flavio, can you provide a Signed-off-by for this patch, please?
On Wed, Oct 12, 2016 at 11:56:51AM -0400, Russell Bryant wrote: > On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole <aconole@redhat.com> wrote: > > > During testing it was found that systemd would consider the openvswitch > > service as a part of networking component, but the dependent services of > > ovs-vswitchd and ovsdb-server were not likewise considered. This leads > > to some strange race conditions, observed when using NFS over TCP, while > > shutting down systems. > > > > Fixes: 84ad12083491 ("rhel: Improved Systemd Integration") > > Co-authored-by: Flavio Leitner <fbl@redhat.com> > > Signed-off-by: Aaron Conole <aconole@redhat.com> > > > > This looks fine to me. Flavio, can you provide a Signed-off-by for this > patch, please? Signed-off-by: Flavio Leitner <fbl@redhat.com>
On Thu, Oct 13, 2016 at 9:44 AM, Flavio Leitner <fbl@redhat.com> wrote: > On Wed, Oct 12, 2016 at 11:56:51AM -0400, Russell Bryant wrote: > > On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole <aconole@redhat.com> wrote: > > > > > During testing it was found that systemd would consider the openvswitch > > > service as a part of networking component, but the dependent services > of > > > ovs-vswitchd and ovsdb-server were not likewise considered. This leads > > > to some strange race conditions, observed when using NFS over TCP, > while > > > shutting down systems. > > > > > > Fixes: 84ad12083491 ("rhel: Improved Systemd Integration") > > > Co-authored-by: Flavio Leitner <fbl@redhat.com> > > > Signed-off-by: Aaron Conole <aconole@redhat.com> > > > > > > > This looks fine to me. Flavio, can you provide a Signed-off-by for this > > patch, please? > > Signed-off-by: Flavio Leitner <fbl@redhat.com> Thanks, I applied this to master.
On Fri, Oct 14, 2016 at 9:36 PM, Russell Bryant <russell@ovn.org> wrote: > > On Thu, Oct 13, 2016 at 9:44 AM, Flavio Leitner <fbl@redhat.com> wrote: > >> On Wed, Oct 12, 2016 at 11:56:51AM -0400, Russell Bryant wrote: >> > On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole <aconole@redhat.com> >> wrote: >> > >> > > During testing it was found that systemd would consider the >> openvswitch >> > > service as a part of networking component, but the dependent services >> of >> > > ovs-vswitchd and ovsdb-server were not likewise considered. This >> leads >> > > to some strange race conditions, observed when using NFS over TCP, >> while >> > > shutting down systems. >> > > >> > > Fixes: 84ad12083491 ("rhel: Improved Systemd Integration") >> > > Co-authored-by: Flavio Leitner <fbl@redhat.com> >> > > Signed-off-by: Aaron Conole <aconole@redhat.com> >> > > >> > >> > This looks fine to me. Flavio, can you provide a Signed-off-by for this >> > patch, please? >> >> Signed-off-by: Flavio Leitner <fbl@redhat.com> > > > Thanks, I applied this to master. > and I backported this to branch-2.6. I'll backport this along with the other systemd related fixes to branch-2.5 in a few minutes.
diff --git a/rhel/usr_lib_systemd_system_openvswitch.service b/rhel/usr_lib_systemd_system_openvswitch.service index a44574b..e823c54 100644 --- a/rhel/usr_lib_systemd_system_openvswitch.service +++ b/rhel/usr_lib_systemd_system_openvswitch.service @@ -1,5 +1,7 @@ [Unit] Description=Open vSwitch +Before=network.target network.service +After=network-pre.target PartOf=network.target BindsTo=ovsdb-server.service BindsTo=ovs-vswitchd.service diff --git a/rhel/usr_lib_systemd_system_ovs-vswitchd.service b/rhel/usr_lib_systemd_system_ovs-vswitchd.service index d3d020a..01d9cee 100644 --- a/rhel/usr_lib_systemd_system_ovs-vswitchd.service +++ b/rhel/usr_lib_systemd_system_ovs-vswitchd.service @@ -1,6 +1,7 @@ [Unit] Description=Open vSwitch Forwarding Unit -After=ovsdb-server.service +After=ovsdb-server.service network-pre.target +Before=network.target network.service Requires=ovsdb-server.service ReloadPropagatedFrom=ovsdb-server.service AssertPathIsReadWrite=/var/run/openvswitch/db.sock diff --git a/rhel/usr_lib_systemd_system_ovsdb-server.service b/rhel/usr_lib_systemd_system_ovsdb-server.service index 46e3e08..d3d7408 100644 --- a/rhel/usr_lib_systemd_system_ovsdb-server.service +++ b/rhel/usr_lib_systemd_system_ovsdb-server.service @@ -1,6 +1,7 @@ [Unit] Description=Open vSwitch Database Unit -After=syslog.target +After=syslog.target network-pre.target +Before=network.target network.service ReloadPropagatedFrom=openvswitch.service PartOf=openvswitch.service
During testing it was found that systemd would consider the openvswitch service as a part of networking component, but the dependent services of ovs-vswitchd and ovsdb-server were not likewise considered. This leads to some strange race conditions, observed when using NFS over TCP, while shutting down systems. Fixes: 84ad12083491 ("rhel: Improved Systemd Integration") Co-authored-by: Flavio Leitner <fbl@redhat.com> Signed-off-by: Aaron Conole <aconole@redhat.com> --- rhel/usr_lib_systemd_system_openvswitch.service | 2 ++ rhel/usr_lib_systemd_system_ovs-vswitchd.service | 3 ++- rhel/usr_lib_systemd_system_ovsdb-server.service | 3 ++- 3 files changed, 6 insertions(+), 2 deletions(-)