SRX

what's happen if all fab interfaces down in Active/Active Cluster ?

Hi,

If all the Fab Links are down , in the latest code , Priority of Nodes in the RG groups will become 0.

But nodes states will be same ( primary node will be primary , secondary node will be secondary)

In older versions , when fab link goes down , secondary node will do to disabled state.

Regards
rparthi
 

Please Mark My Solution Accepted if it Helped, Kudos are Appreciated Too

Hi ,

HA cluster state diagram:

http://kb.juniper.net/InfoCenter/index?page=content&id=KB22717.

In active/active HA , when fab links are down , if traffic received on node0 that needs to be sent out through Node1 then these packets will get dropped. for Z mode traffic , packets will be dropped.

But secondary device will be up in the latest junos codes.

Regards
rparthi
 

Please Mark My Solution Accepted if it Helped, Kudos are Appreciated Too

I agree with rparthi that traffic was dropped.  From My scenario I have implement SRX chassis cluster in transparent active/active  with asymmetric routing  if fab link down return traffic cannot pass firewall because other node cannot find RTO from other node 

Hi Seanmine,

Thanks for the update.

Please Mark My Solution Accepted if it Helped, Kudos are Appreciated Too

Regards,

Parthi

I remember that when fab interfaces fail RG group  priority do not reduce to 0. 

Hi seanmine,

RG group  priority do not reduce to 0 on fab-failure is because Fabric Monitoring is disabled since 10.4.

Below documents can provide more details on fab-failures on SRX.

http://kb.juniper.net/KB23862

https://www.juniper.net/documentation/en_US/junos11.4/topics/concept/chassis-cluster-data-plane-understanding.html

Thanks,

Suraj

Please Mark My Solution Accepted if it Helped, Kudos are Appreciated too

Oh good solution .I will test with your advice

• When packets arrive on an interface on one node, but must be processed on the other node

  If the ingress and egress interfaces for a packet are on one node, but the packet must be processed on the other node because its session was established there, it must traverse the data link twice. This can be the case for some complex media sessions, such as voice-over-IP (VoIP) sessions.

In case of transparent active/active   

if session initial  through node0 session mark as active in node0  and mak as backup in node1 but  return traffic through node1  . Will traffic go to fab link to processing in node0 then return back to node1 ?

Fabric link mornitoring on high end is hidden cli