Description of problem:

Check the openflow after deleted the pod, still remained one rule for each pod

[root@atomic /]# ovs-ofctl dump-flows br0 -O OpenFlow13 | grep 10.1.2.8
 cookie=0x0, duration=761.751s, table=4, n_packets=0, n_bytes=0, priority=100,ip,in_port=10,nw_src=10.1.2.8 actions=load:0xc->NXM_NX_REG0[],goto_table:5

[root@atomic /]# ovs-ofctl dump-flows br0 -O OpenFlow13 | grep 10.1.2.7
 cookie=0x0, duration=766.576s, table=4, n_packets=0, n_bytes=0, priority=100,ip,in_port=9,nw_src=10.1.2.7 actions=load:0xc->NXM_NX_REG0[],goto_table:5


Version-Release number of selected component (if applicable):
origin
oc v1.1-603-gb1cffde-dirty
kubernetes v1.1.0-origin-1107-g4c8e6f4
plugin:  redhat/openshift-ovs-multitenant

ose
oc v3.1.1.0
kubernetes v1.1.0-origin-1107-g4c8e6f4
plugin:  redhat/openshift-ovs-multitenant


How reproducible:
Always

Steps to Reproduce:
1. Set up multi-node env with multitenant plugin
2.  Create some pod
#oc create -f https://raw.githubusercontent.com/openshift-qe/v3-testfiles/master/networking/list_for_pods.json
# oc describe pod  | grep IP
IP:                10.1.2.8
IP:                10.1.2.7
3. Check the openflow list on the node
[root@atomic /]# ovs-ofctl dump-flows br0 -O OpenFlow13 | grep 10.1.2.8
 cookie=0x0, duration=92.340s, table=4, n_packets=0, n_bytes=0, priority=100,ip,in_port=10,nw_src=10.1.2.8 actions=load:0xc->NXM_NX_REG0[],goto_table:5
 cookie=0x0, duration=92.337s, table=7, n_packets=0, n_bytes=0, priority=100,ip,reg0=0xc,nw_dst=10.1.2.8 actions=output:10
 
[root@atomic /]# ovs-ofctl dump-flows br0 -O OpenFlow13 | grep 10.1.2.7
 cookie=0x0, duration=95.979s, table=4, n_packets=0, n_bytes=0, priority=100,ip,in_port=9,nw_src=10.1.2.7 actions=load:0xc->NXM_NX_REG0[],goto_table:5
 cookie=0x0, duration=95.975s, table=7, n_packets=0, n_bytes=0, priority=100,ip,reg0=0xc,nw_dst=10.1.2.7 actions=output:9
 
4. Delete the pods
5. Check the openflow list again


Actual results:
step5: Still remained one rule for each pod after deleted the pod

[root@atomic /]# ovs-ofctl dump-flows br0 -O OpenFlow13 | grep 10.1.2.8
 cookie=0x0, duration=761.751s, table=4, n_packets=0, n_bytes=0, priority=100,ip,in_port=10,nw_src=10.1.2.8 actions=load:0xc->NXM_NX_REG0[],goto_table:5
 
[root@atomic /]# ovs-ofctl dump-flows br0 -O OpenFlow13 | grep 10.1.2.7
 cookie=0x0, duration=766.576s, table=4, n_packets=0, n_bytes=0, priority=100,ip,in_port=9,nw_src=10.1.2.7 actions=load:0xc->NXM_NX_REG0[],goto_table:5

Wait a few minutes and check again,  the rules are still there.


Expected results:
All pod openflow rules could be cleaned after delete the pod


Additional info:
Service openflow rules could be cleaned successfully after deleted service.