-
Task
-
Resolution: Obsolete
-
Major
-
None
-
None
-
None
- Background:
- IBM is our (ACSCS) "friendly tenant", they're willing to go through some pain and let us try new features
- IBM is the biggest tenant. They will become even bigger (scalability?) soon.
- We want to first test ScannerV4 on IBM, then move everybody to ScannerV4
- We understand that there is some OOMing issue going on, related to ScannerV4
- We are trying to understand the risks
- So far, we(ACSCS)'ve gathered that
- Switching back to ScannerV2 is impossible if node scanning was enabled
- Switching back to ScannerV2, regardless, causes some "pain, broken policies and confusion"
- ScannerV4 components have started OOMed recently, and we wanted to make sure that the memory usage was expected.
- We have only a very piecemeal/partial understanding of the consequences of enabling Scanner V4:
- ScannerV4 and/or ScannerV4 interactions/cohabitation
- Node scanning
- Delegated Scanning
- Others? Maybe SBOM generation ?
- Some more background: in a nutshell, our objective is to assess the risk of enabling ScannerV4 for IBM on the Cloud Service.
- clones
-
-
- Closed
-
