-
Bug
-
Resolution: Unresolved
-
Normal
-
None
-
None
-
1
-
False
-
-
False
-
CY25Q1-S3
-
Very Likely
-
0
Description
We are currently working on parsing CSAF VEX for the Trivy security scanner and have encountered some uncertainty regarding the interpretation of RPM PURLs, specifically concerning their namespace usage.
If this isn't the appropriate forum for this question, I would be grateful if you could redirect me to the proper channel.
Context and Examples
Looking at *CVE-2023-39325*, there are two product versions for the same package `foreman_ygg_worker`:
```json
{
"category": "product_version",
"name": "foreman_ygg_worker",
"product": {
"name": "foreman_ygg_worker",
"product_id": "foreman_ygg_worker",
"product_identification_helper":
}
},
{
"category": "product_version",
"name": "satellite_client_6/foreman_ygg_worker",
"product": {
"name": "satellite_client_6/foreman_ygg_worker",
"product_id": "satellite_client_6/foreman_ygg_worker",
"product_identification_helper":
}
}
```
One version has an empty namespace, while the other includes `satellite_client_6` in the namespace. Seeing the website, Both appear to be components of Red Hat Satellite 6, but with different severity ratings (one being Moderate).
I've noticed some inconsistencies across different CVEs:
1. CVE-2023-44487: Only includes `satellite_client_6/foreman_ygg_worker` as a Red Hat Satellite 6 component
2. CVE-2022-41717: Conversely, only includes `foreman_ygg_worker` as a Red Hat Satellite 6 component
-
- Questions
1. What does the namespace (e.g., `satellite_client_6`) signify in this context? Does it represent repository information?
2. How should these differences across CVEs be interpreted when implementing vulnerability scanning?
I appreciate any guidance on correctly interpreting these PURL formats and their namespaces.