Hey everyone! I’ve been diving into Versa Networks over the last few months, and I’m really excited about the potential for SD-WAN and SASE. I’m coming from a data analytics background, so networking is a bit new for me, but I love learning how data flows through these pipelines!
I’m hoping someone can walk me through a specific issue I hit during a lab test. I set up application routing policies to steer certain app traffic (like our BI tool) over a specific, more reliable WAN link. Everything works perfectly... until I simulate a link failure.
After the failover happens, the backup link comes up as expected, but my application-specific routing policies seem to stop applying. All traffic just follows the new default route instead of the policy I defined for that application. Once the primary link recovers and I fail back, the policies start working again.
Here’s what I’ve checked so far:
* The policy is definitely activated and committed in the director.
* The application is correctly identified in the policy (using the built-in application DB).
* Path monitoring is enabled for the primary link.
* The failover event itself is logged and successful.
My gut, from a data pipeline mindset, is that maybe the policy evaluation is somehow tied to the original path state? Or perhaps there’s a specific failover behavior I’m missing.
Could anyone provide a detailed walkthrough of how application routing policies are *supposed* to behave during and after a failover? Specifically:
* Do the policies need to reference the WAN link by name, or by a “role” like `primary-internet`?
* Is there a re-evaluation delay or a dependency on the application identification service coming back up?
* Are there any common configuration pitfalls in the policy order or match criteria that could cause this?
I’d be super grateful for any pointers, recommended docs, or even a step-by-step of your working config. Thanks in advance! This community has been incredibly helpful.