AI Runtime Security
Known Issues
Table of Contents
Known Issues
A list of known issues in AI Runtime Security.
Review the list of known issues as per the latest release for AI Runtime
Security:
|
Issue ID
|
Description
|
|---|---|
| AIFW-750 | When you destroy the `security_project` Terraform, the Terraform deployment screen (Insights → AI Runtime Security) still shows the firewall with a Deployed status. You can manually delete the Terraform item for the firewall from the deployment screen. |
| AIFW-720 | The onboarding workflow in Strata
Cloud Manager fails if your Strata Logging Service (SLS) has expired.
Verify your SLS license status when creating a deployment profile
in the Customer Support Portal. If your SLS has expired, renew it
before onboarding a cloud account to ensure successful onboarding
Terraform generation. |
| PAN-278537 | Tag-collector CommitALL fails in Strata Cloud Manager (Manage →
Operations → Push Config) when security policies use DLP
profiles in security groups. Workaround: Remove the
DLP profile from the security group in Strata Cloud Manager and push
the configuration to Tag Collector. |
| PAN-278065 | When configuring AI security profiles in Panorama (Objects > Security
Profiles > AI Security), if you edit the advanced settings, the custom
Database Security action values under "AI Data Protection > Database
Security" are reverted to defaults upon saving the
profile. Workaround: Configure AI profile with model
group protections for Data
Protection → Database
Security without exiting the edit window for the
security profile and sub-windows. |
| PAN-280130 | GKE Autopilot clusters do not support Helm deployments due to restrictions on modifying the kube-system namespace. |
| AIFW-680 | During Azure cloud account
onboarding, the virtual network associated with
pre-existing Kubernetes clusters may not be automatically discovered. Workaround: When you create a Kubernetes cluster,
enable the "Bring your own Azure virtual network" option in the
Azure Portal for proper discovery. To enable this:
|
| AIFW-506 | The Application breakdown section on the Applications page () may show a discrepancy in the total number of
applications. The breakdown of VM and Pod applications may not sum up to
the total number of applications displayed at the top of the page. This
is due to some applications being categorized as both VM and Pod types,
leading to potential double-counting in the breakdown. The total
application count remains accurate, reflecting the unique number of
applications across all types.
|
| PAN-280130 | GKE Autopilot clusters do not support Helm deployments due to restrictions on modifying the kube-system namespace. |
| AIFW-690 | GCP account onboarding requires a 10-second wait after a successful Terraform application before clicking "Done" to complete the process. |
| PAN-256741 | Traffic Routing Blocked Between `fw-trust-vpc` and
`app-vpc` Traffic routing between fw-trust-vpc and app-vpc via
VPC peering is currently blocked because the route export from
fw-trust-vpc to app-vpc for 0.0.0.0/0 to ILB is hindered by an
existing default gateway route in the app-vpc.
Workaround: Create a default route in the app-vpc
which uses the AI Runtime Security ILB as the next hop. This ensures
traffic routes correctly through the AI Runtime Security: Network
intercept (AI firewall), enforcing security policies. |
| PLUG-16395 | IPv6 Tags harvesting is not supported. |
| AIFW-421 | Missing CIDR retrieval during AI Runtime Security
deployment While deploying an AI Runtime Security: Network intercept
(Insights
→ AI Runtime Security) in Strata Cloud Manager,
selecting the application namespace does not retrieve the cluster
pod and service CIDR. Workaround: After generating the
Terraform configuration, please whitelist these CIDR values in the
Firewall Trust VPC firewall rule. |
| PAN-263750 | No Detection Logs for GenAI LLM Apps in AI Security Threat
Logs Detection logs for AI threats are missing in
the AI security threat logs under Strata Cloud Manager
(Incidents and Alerts → Log
Viewer) when AI models are targeted by GenAI
prompts. This issue occurs when AI LLM Applications
are defined in the security policy, but the necessary dependent
applications (such as SSL and web browsing) are not included. As a
result, the AI network intercept provides inaccurate threat
verdicts. Workaround: Navigate to
Manage →
Configuration → NGFW and Prisma
Access → Security Services
→ Security Policy and scope it to your AI
network intercept. Edit the policy to include the dependent
apps (web browsing and SSL apps). This will ensure the AI network
intercept detects and logs AI security threats correctly in
the Log Viewer. |
| PAN-266547 | Tag Collector in TC Mode enters Maintenance Mode after upgrade to
`v11.2.3` The issue occurs when you upgrade the AI
Runtime Security tag collector from `v11.2.2-h1` to `v11.2.3`, the
tag-collector enters a rebooting loop and eventually goes to
maintenance mode. Workaround: Don’t upgrade
to `v11.2.3` as the auto-commit feature is not triggered in
`v11.2.3`. |
| PAN-266547 | Tag Collector running `v11.2.2-h1` enters Maintenance Mode with
instance types other than `n2-standard-4` and `Standard_DS3_v2` post
bootstrap. This is due to incorrect capacity file computation and
excessive memory usage. Workaround: Use `n2-standard-4` or
`Standard_DS3_v2` instance sizes for running `v11.2.2-h1` to avoid
this issue. |