CVE-2026-34881 Overview
OpenStack Glance before 29.1.1, 30.x before 30.1.1, and 31.0.0 is affected by a Server-Side Request Forgery (SSRF) vulnerability. By leveraging HTTP redirects, an authenticated user can bypass URL validation checks and redirect requests to internal services. This vulnerability specifically affects the Glance image import functionality, including the web-download and glance-download import methods, as well as the optional ovf_process image import plugin (not enabled by default).
Critical Impact
Authenticated attackers can bypass URL validation through HTTP redirects, potentially accessing internal services and sensitive resources not intended to be publicly accessible.
Affected Products
- OpenStack Glance versions before 29.1.1
- OpenStack Glance 30.x versions before 30.1.1
- OpenStack Glance version 31.0.0
Discovery Timeline
- 2026-03-31 - CVE-2026-34881 published to NVD
- 2026-04-02 - Last updated in NVD database
Technical Details for CVE-2026-34881
Vulnerability Analysis
This Server-Side Request Forgery (SSRF) vulnerability in OpenStack Glance allows authenticated users to manipulate the image import process to access internal resources. The flaw exists in how Glance handles HTTP redirects during image import operations. While Glance implements URL validation to prevent access to internal services, the validation occurs only on the initial URL provided by the user. When the remote server responds with an HTTP redirect (such as 301 or 302 status codes), Glance follows the redirect without re-validating the destination URL against the same security policies.
This bypass mechanism enables attackers to craft malicious URLs that initially pass validation but redirect to internal network addresses, cloud metadata services, or other restricted endpoints. The vulnerability is classified under CWE-918 (Server-Side Request Forgery), reflecting its core nature as an SSRF attack vector.
Root Cause
The root cause of this vulnerability lies in insufficient validation of redirect destinations in the Glance image import process. The URL validation logic applies security checks only to the original user-supplied URL, not to subsequent redirect targets. This architectural oversight allows attackers to use a legitimate external URL that redirects to internal resources, effectively bypassing the intended security controls.
Attack Vector
The attack requires network access and authenticated credentials to the OpenStack Glance service. An attacker can exploit this vulnerability through the following mechanism:
- The attacker creates or controls an external web server configured to return HTTP redirect responses
- The attacker initiates an image import using the web-download or glance-download method, providing the URL of their controlled server
- Glance validates the initial URL, which points to the external server and passes security checks
- Glance fetches the URL, and the attacker's server responds with an HTTP redirect to an internal service (e.g., http://169.254.169.254/ for cloud metadata)
- Glance follows the redirect without re-validating the destination, accessing the internal resource
This attack can be used to probe internal network services, access cloud provider metadata endpoints containing sensitive credentials, or interact with internal APIs that should not be accessible from the Glance service context.
Detection Methods for CVE-2026-34881
Indicators of Compromise
- Unexpected outbound connections from Glance services to internal IP ranges (e.g., 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16)
- Image import requests targeting known redirect services or URL shorteners
- Access logs showing requests to cloud metadata endpoints (e.g., 169.254.169.254) from Glance worker processes
- Anomalous patterns in image import failures followed by internal service access attempts
Detection Strategies
- Monitor Glance API logs for image import requests with suspicious external URLs, particularly those using URL shorteners or known redirect services
- Implement network segmentation monitoring to detect unexpected traffic from Glance services to internal network segments
- Configure alerting for any access attempts to cloud metadata services from OpenStack service accounts
- Review Glance import job logs for patterns indicating redirect chains or failed imports after following redirects
Monitoring Recommendations
- Enable verbose logging for Glance image import operations to capture full request chains including redirects
- Deploy network monitoring to track all outbound connections from Glance services and alert on connections to RFC 1918 addresses or link-local addresses
- Implement egress filtering with logging to detect and record attempts to access internal resources
- Monitor for unusual spikes in image import failures that may indicate exploitation attempts
How to Mitigate CVE-2026-34881
Immediate Actions Required
- Upgrade OpenStack Glance to version 29.1.1, 30.1.1, or later depending on your deployment branch
- If immediate patching is not possible, consider temporarily disabling the web-download and glance-download import methods
- Review and audit recent image import activities for signs of exploitation
- Implement network-level controls to restrict Glance service access to internal networks
Patch Information
OpenStack has released security patches addressing this vulnerability. Users should upgrade to Glance version 29.1.1 or later for the 29.x branch, version 30.1.1 or later for the 30.x branch, or wait for the patched release following 31.0.0. For detailed patch information, refer to the OpenStack Security Advisory OSSA-2026-004 and the Launchpad Bug Report.
Workarounds
- Disable the web-download and glance-download import methods if they are not essential to your operations by modifying the Glance configuration
- If the ovf_process import plugin is enabled, consider disabling it until patches can be applied
- Implement strict egress firewall rules for Glance services to prevent access to internal networks and cloud metadata endpoints
- Deploy a reverse proxy or web application firewall in front of external URLs accessed by Glance to validate and sanitize redirect destinations
The vulnerability can be mitigated at the network level by configuring firewall rules to prevent Glance worker processes from accessing internal IP ranges. This should be configured on network firewalls or through cloud provider security groups to ensure Glance can only reach legitimate external image sources.
Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.


