- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
3 weeks ago
Team,
We have a multi cloud environment. (AWS / Azure & GCP) & we want to perform discovery across all clouds.
As per the environment, cloud team is saying that they can provide connectivity from AWS to Azure & GCP.
It is a best practice? Can we Perform this type of Discovery?
Solved! Go to Solution.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
2 weeks ago
@ChaitanyaG26248 Wanted to check whether did you get chance to review this?
✅ Issue resolved? → Mark as Correct
Found value? → Mark as Helpful
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
3 weeks ago
Yes, technically this type of setup can work, but from an enterprise architecture and ServiceNow best practice perspective, it is usually NOT the recommended approach for large-scale multi-cloud discovery.
Recommended Best Practice
For multi-cloud environments (AWS, Azure, GCP), the preferred approach is:
- Deploy MID Servers close to each cloud environment
- Keep discovery traffic local to that cloud/network
- Use cloud-native APIs wherever possible
- Avoid routing all discovery traffic through one cloud unnecessarily
Example:
- AWS MID → AWS resources
- Azure MID → Azure resources
- GCP MID → GCP resources
This provides:
- Better scalability
- Lower latency
- Better security segmentation
- Reduced firewall complexity
- Improved troubleshooting
- Better resiliency
Can Discovery Work Across Clouds?
Yes, if:
- Proper routing exists
- Firewall ports are open
- DNS resolution works
- SSH/WMI/SNMP/API access is allowed
- Latency is acceptable
Then a MID Server hosted in AWS can technically discover Azure/GCP resources.
Important Considerations
1. Network Latency
Cross-cloud discovery may increase:
- SSH/WMI timeout issues
- Slower pattern execution
- ECC queue delays
2. Security Concerns
Security teams may not prefer:
- East-west cloud connectivity
- Broad firewall openings between clouds
3. Scalability
One centralized MID handling all clouds may become:
- overloaded
- harder to troubleshoot
- a bottleneck
4. Cloud Discovery vs Traditional Discovery
For cloud resources, prefer:
- AWS Cloud Discovery
- Azure Cloud Discovery
- GCP Cloud Discovery
using cloud APIs instead of only traditional IP-based discovery.
This is more reliable and cloud-native.
Recommended Enterprise Architecture
Most enterprise customers use:
- Separate MID clusters per cloud/region
- Dedicated cloud service accounts
- API-based cloud discovery
- Local network discovery within each cloud
rather than routing all discovery traffic from one cloud to another.
Final Recommendation
- Possible? → Yes
- Supported? → Yes
- Recommended for enterprise scale? → Usually No
Best practice is distributed MID architecture aligned with each cloud environment.
✅ Issue resolved? → Mark as Correct
Found value? → Mark as Helpful
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
2 weeks ago
@ChaitanyaG26248 Wanted to check whether did you get chance to review this?
✅ Issue resolved? → Mark as Correct
Found value? → Mark as Helpful
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
2 weeks ago
Best Practise
Instead of routing everything through AWS, you should deploy local resources in each cloud:
-
Deploy Local MID Servers: Place dedicated MID Server clusters inside each cloud environment (e.g., an AWS MID cluster for AWS, an Azure MID cluster for Azure, and a GCP MID cluster for GCP). Keep the heavy discovery traffic local to that specific cloud's network.
and @shubhamseth already mentioned point like network letancy , Schedule may fail,Security Concerns etc
So to avoid that Issue use seprate mid for seprate cloud
If my answer helps please mark helpful and accept solution