AAP/FIP: flow gets stuck in H state upon s/o
Affects | Status | Importance | Assigned to | Milestone | ||
---|---|---|---|---|---|---|
Juniper Openstack | Status tracked in Trunk | |||||
R3.1 |
New
|
Medium
|
Naveen N | |||
Trunk |
Fix Committed
|
Medium
|
Naveen N |
Bug Description
Consider a case where a FIP from a public VN has been associated with an AAP (which in-turn front ends a vrrp group that has 2 VMs). This is done by setting the 'floating_
If a ping is initiated from a VM that is on the same compute as the current active (these pings pass), and then a s/o is done (pings sent to new active and they pass too), a flow is Hold state is seen on the local compute. If another s/o happens while this H flow is present, the pings will fail.
To begin with: 44.1.1.3 is source VM, 192.169.0.17 is FIP, 10.0.0.50 is AAP. Active VM is on the same compute.
Index Source:
-------
392780<=>519484 44.1.1.3:16778 1 (3->1)
(Gen: 152, K(nh):55, Action:N(D), Flags:, QOS:-1, S(nh):0, Stats:27/2646, SPort 54838 TTL 0)
519484<=>392780 10.0.0.50:16778 1 (1->2)
(Gen: 46, K(nh):19, Action:N(S), Flags:, QOS:-1, S(nh):0, Stats:27/2646, SPort 62967 TTL 0)
After s/o H flow is seen:
Listing flows matching ([44.1.1.3]:*)
Index Source:
-------
164112 44.1.1.3:16778 1 (1)
(Gen: 5, K(nh):19, Action:H, Flags:, QOS:-1, S(nh):0, Stats:1/98, SPort 0 TTL 0)
166664<=>392780 192.169.0.17:16778 1 (3)
(Gen: 74, K(nh):55, Action:F, Flags:, QOS:-1, S(nh):23, Stats:34/2856, SPort 57541 TTL 0)
392780<=>166664 44.1.1.3:16778 1 (3)
(Gen: 153, K(nh):55, Action:F, Flags:, QOS:-1, S(nh):55, Stats:34/3332, SPort 54838 TTL 0)
Review in progress for https:/ /review. opencontrail. org/24618
Submitter: Naveen N (<email address hidden>)