sdn: fix initialization order to prevent crash on node startup #13767
Conversation
dcbw
force-pushed
the
15-sdn-multicast-init-order
branch
from
84502cb to
b60f14a
Compare
|
[test] |
|
[test]
…On Wed, Apr 19, 2017 at 6:53 PM, OpenShift Bot ***@***.***> wrote:
continuous-integration/openshift-jenkins/test FAILURE (
https://ci.openshift.redhat.com/jenkins/job/test_pull_request_origin/804/)
(Base Commit: 8ef58c5
<8ef58c5>
)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#13767 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABG_p9DigiExDmTjNEZYN9NTewllyPRbks5rxpBsgaJpZM4M9Zua>
.
|
dcbw
force-pushed
the
15-sdn-multicast-init-order
branch
from
b60f14a to
521c79f
Compare
|
(updated to match final code in #13766) |
There was a problem hiding this comment.
(updated to match final code in #13766)
I'd said there that I wasn't comfortable putting the kubelet init change into 1.5 without it getting testing in master first, so if we're making that change here, then it's not ready to land yet
OsdnNode.Start() (node.pm == nil at this point) -> node.policy.Start() (which is multitenant policy) -> mp.vnids.Start() -> go vmap.watchNetNamespaces() -> (net namespace event happens) -> watchNetNamespaces() -> vmap.policy.AddNetNamespace() (policy is multitenant) -> mp.updatePodNetwork() -> mp.node.podManager.UpdateLocalMulticastRules() (and podManager is still nil) Create the PodManager earlier so it's not nil if we get early events. Fixes: openshift#13742
dcbw
force-pushed
the
15-sdn-multicast-init-order
branch
from
521c79f to
030e3ec
Compare
|
@danwinship updated to move the kubelet init bits back. |
|
Evaluated for origin test up to 030e3ec |
|
[merge] |
|
[merge] now that i think things are alivish |
|
[merge] fun times |
|
Evaluated for origin merge up to 030e3ec |
|
continuous-integration/openshift-jenkins/merge SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pull_request_origin/880/) (Base Commit: 2aadab3) (Image: devenv-rhel7_6171) |
|
continuous-integration/openshift-jenkins/test SUCCESS (https://ci.openshift.redhat.com/jenkins/job/test_pull_request_origin/880/) (Base Commit: 2aadab3) |
|
[merge]
…On Sun, Apr 23, 2017 at 3:25 PM, OpenShift Bot ***@***.***> wrote:
continuous-integration/openshift-jenkins/test SUCCESS (
https://ci.openshift.redhat.com/jenkins/job/test_pull_request_origin/880/)
(Base Commit: 2aadab3
<2aadab3>
)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#13767 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/ABG_px3cxiYEYe4vQ19AozNlK3MPfQ8Lks5ry6WYgaJpZM4M9Zua>
.
|
OsdnNode.Start()
(node.pm == nil at this point)
-> node.policy.Start() (which is multitenant policy)
-> mp.vnids.Start()
-> go vmap.watchNetNamespaces()
-> (net namespace event happens)
-> watchNetNamespaces()
-> vmap.policy.AddNetNamespace() (policy is multitenant)
-> mp.updatePodNetwork()
-> mp.node.podManager.UpdateLocalMulticastRules() (and podManager is still nil)
Create the PodManager earlier so it's not nil if we get early events.
Fixes: #13742
(cherry picked from commit fc95b86)
@openshift/networking @smarterclayton