Skip to main content

CDP - Practical Use Case


Eric.Beck
Forum|alt.badge.img

Objective: Utilize Veeam’s CDP to protect critical virtual machines where near-zero data loss is essential.

Background: We are using CDP to safeguard our network monitoring software, which collects and graphs data from our network elements over time. The key goal is to ensure consistent graphical data during polling periods, maintaining accuracy even during failover or updates.

Constraints: Deploying CDP is more complex than anticipated, as it’s an advanced feature that requires careful configuration. While the deployment process could be an entire article in itself, the most important advice is to verify DNS resolution for all components being deployed—double, triple, and even quadruple check to ensure everything is correctly configured.
 

  1. The first part of my test was to seed the CDP policy to a replica we had on our target server.
     

     

  2. After defining the polling period for the Network Management System (NMS), it will query its elements every 300 seconds to ensure consistent graphical data. In theory, we could reduce the Recovery Point Objective (RPO) within that polling cycle, but this was part of our initial limit testing.
     

     

  3. Once we created our CDP policy, we let the sync finalize and reference the replica state in the VBR console.
     
  4. From here, we are set to initiate the failover to the replica with the most recent restore point.
     

     

  5. See timing on the failover start and completion:
     

     

  6. In the active failover state, we can see the graphical data for the NSM elements have not been disrupted.

     
  7. Now that we have verified functionality and no graphical data loss, we must fail back to the production machine to verify 2-way failover functionality.
     

     
  8. Now that the verification is completed, I can commit the failback to production and let VBR initialize its CDP functions.
  9. Final graphical verification once we are back to the production machine.

Utilizing Veeam CDP I was able to complete a failover with “near zero” data loss and disruption. This Is pivotal for our networking team to maintain visibility into their network elements as well as general operations of our environment. 

Plus its just cool!

1 comment

Man, this is next-level wizardry. Honestly, reading this felt like watching a precision surgeon operate on virtual machines — except you’re also the patient, the doctor, and somehow the hospital IT guy making sure DNS is working. I’m genuinely impressed at how clean and methodical you made something that would absolutely cause me an existential crisis halfway through.

The fact that you pulled off a failover and failback without so much as a blip on the NMS graphs? That’s the IT equivalent of walking a tightrope while juggling and looking cool doing it. Seriously inspiring stuff — makes me want to go double-check my own backups and question every decision I’ve made.

Also, the casual "Plus it's just cool!" at the end? Peak energy. Like, yeah, just casually flexing a near-zero RPO like it’s no big deal.

10/10. Would read your autobiography titled "I Fought DNS and Won".


Comment