Link to home
Start Free TrialLog in
Avatar of Khitrov
KhitrovFlag for Russian Federation

asked on

Riverbed virtual in-path issue

User generated image Hello, dear Experts! I've successfully tried to deploy physical in-path deployment of my riverbed 550 and 1050 and now I am trying to implement virtual in-path configuration. My scheme and pieces of config on the picture. swbr and swhq it is a layer3 cisco 3750. Router - it is a WANem, linux-based router. I am trying to deploy RVBDs via PBR. In network statistic I am seeing passtrough traffic on both RVBDs, but it can not optimize traffic. What is possible mistake or bug in my installation?
SOLUTION
Avatar of BDC-Net
BDC-Net

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
ASKER CERTIFIED SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of Khitrov

ASKER

BDC-Net, It is good, but our IT-security requires to use PBR. RVBD allows to do this, I've configured it relative to RVBD PBR deployment guide.
SOLUTION
Avatar of kevinhsieh
kevinhsieh
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of Khitrov

ASKER

kevinhsieh, PBR on cisco routers can verify next hop address before redirect traffic, it allows to avoid black holes.
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of Khitrov

ASKER

Thanks for all. I've configured WCCP solution. It works well.
@BDC-Net, Cisco can verify that the next hop is alive (responds to ping), but that doesn't mean that the Riverbed is actually able to forward the traffic. For example, the Cisco will send traffic to the Riverbed Steelhead appliance before the appliance has been able to start the optimization service, which wll cause those packets to get dropped. I have tested physical in-path, PBR, and WCCP, and I found WCCP to be the least disruptive to traffic when rebooting the appliance.