MEDIUM 5.4

CVE-2026-10601: Grafana Tempo & Loki Path Traversal Credential Exposure

A vulnerability in Grafana's Tempo and Loki data source plugins allows users with basic Viewer permissions to bypass intended access controls and reach backend endpoints they shouldn't be able to access. By crafting specially designed requests, an authenticated attacker can potentially steal database credentials, extract sensitive internal information, or execute administrative commands on the connected backend system. While the attack requires valid user credentials, the low barrier to entry (Viewer level is a basic permission tier) and the potential exposure of sensitive infrastructure credentials make this a meaningful risk for most organizations.

Source data · NVD / CISA · public domain

CVSS
3.1 · 5.4 MEDIUM · CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:L/A:L
Weaknesses (CWE)
CWE-22
Affected products
1 configuration(s)
Published / Modified
2026-06-22 / 2026-07-10

NVD description (verbatim)

A user with Viewer permissions can use specially crafted requests to the Tempo and Loki data source plugins to reach unintended backend endpoints. Depending on the backend configuration this can expose data source credentials, leak internal responses, or trigger administrative actions on the configured backend.

1 reference(s) · View on NVD →

SEC.co analysis · AI-assisted, reviewed against source

Technical summary

CVE-2026-10601 is a path traversal vulnerability (CWE-22) affecting Grafana's Tempo and Loki data source plugins. The flaw allows a user with Viewer-level permissions to construct requests that traverse intended API boundaries and reach unintended backend endpoints. The vulnerability exists because the plugins do not properly validate or restrict the endpoints a Viewer can target through the data source interface. Depending on the backend configuration and network architecture, successful exploitation can result in credential exposure, information disclosure of internal API responses, or execution of unintended administrative actions on the backend system. The CVSS 3.1 score of 5.4 (Medium) reflects a network-accessible vulnerability requiring low privileges, with limited integrity and availability impact but no confidentiality impact in the base vector—though actual impact can be amplified by specific backend configurations.

Business impact

If exploited, this vulnerability could lead to compromise of data source credentials stored within Grafana, which often include database passwords or API keys for critical monitoring infrastructure. Exposure of these credentials enables lateral movement to backend systems (databases, time-series backends, or APIs). In organizations where Grafana is widely used for observability, the loss of credential secrecy could affect multiple downstream systems. Additionally, triggering unintended administrative actions on backends could cause data manipulation, service disruption, or misconfiguration of monitoring systems themselves, potentially masking other security incidents.

Affected systems

Grafana is affected, specifically the Tempo and Loki data source plugins. The vulnerability applies to any Grafana instance with Tempo or Loki configured as a data source, where users with Viewer permissions or higher exist. Organizations with Grafana deployed in multi-tenant or shared environments face elevated risk, as Viewer is a low-privilege role often assigned broadly. The actual exploitability and impact depend on what backends are connected (Tempo, Loki, or other systems these plugins interface with) and their network exposure.

Exploitability

Exploitation requires valid Grafana credentials at the Viewer permission level—the lowest standard role in Grafana's access model. No user interaction is needed; exploitation is purely programmatic. The attack is practical for insiders or attackers who have obtained valid user credentials. The impact scales with backend configuration: if backends are internet-facing or contain sensitive data/credentials, the risk is substantially higher. At present, this vulnerability is not listed on the CISA KEV catalog, indicating either limited active exploitation in the wild or recent disclosure with monitoring still underway.

Remediation

Upgrade Grafana to a patched version that includes fixes for the Tempo and Loki data source plugins. Verify against the official Grafana security advisory for exact affected version ranges and patch availability. In parallel, audit Grafana user permissions and reduce the number of users with Viewer or higher access to only those who require it. Review data source credentials and rotate any that may have been exposed. Implement network segmentation to limit the backends that Grafana can reach, and restrict Grafana's outbound connectivity to only necessary endpoints. Monitor Grafana audit logs for suspicious requests to data source endpoints.

Patch guidance

Consult the official Grafana security advisory to identify which versions of Grafana and the Tempo/Loki plugins are affected, and which patch versions resolve the issue. Patch as soon as feasible, prioritizing Grafana instances that handle sensitive data or run in environments with broad Viewer access. Test patches in a non-production environment first, particularly if your monitoring stack depends on Grafana for critical observability. After patching, validate that Tempo and Loki data sources continue to function correctly.

Detection guidance

Review Grafana API access logs for unusual requests originating from Viewer-level accounts targeting Tempo or Loki data source endpoints. Look for repeated failed requests to unfamiliar paths, requests to backend management endpoints, or rapid enumeration of endpoints. Enable detailed audit logging in Grafana if not already active. Monitor backend systems (Loki, Tempo servers) for unexpected API calls or authentication attempts from Grafana's IP range. Search for any suspicious credentials or environment variables that may have been extracted via this vulnerability.

Why prioritize this

Although scored Medium (5.4 CVSS), this vulnerability merits prioritization because: (1) it requires only Viewer permissions—a commonly assigned role; (2) successful exploitation directly exposes data source credentials, which are prime targets for lateral movement; (3) Grafana is often deployed in security-critical observability roles where compromise could mask other attacks; (4) the Tempo and Loki plugins are widely used in cloud-native and Kubernetes environments where backend systems may be high-value targets. The lack of KEV listing does not diminish internal risk, as limited public exploit availability may reflect recency rather than low risk.

Risk score, explained

The CVSS 3.1 score of 5.4 reflects a network-accessible, low-privilege vulnerability with limited direct confidentiality impact per the base vector. However, real-world risk is contextual and often higher: credential exposure (a likely outcome) amplifies impact beyond the base score, and in multi-tenant Grafana environments, a single compromised Viewer account could unlock access to shared backend infrastructure. The score should be treated as a floor; organizations with broad Viewer access, internet-facing backends, or sensitive monitoring data should treat this as a higher-priority issue.

Frequently asked questions

Can this vulnerability be exploited without valid Grafana credentials?

No. Exploitation requires an authenticated user account with at least Viewer permissions. This is a privilege-escalation vulnerability rather than an unauthenticated entry point. However, Viewer is a minimal privilege role, and in many organizations it is assigned to a large user base.

What is the difference between Tempo and Loki, and why are both plugins affected?

Loki is a log aggregation system, and Tempo is a distributed tracing backend. Both are observability platforms compatible with Grafana as data sources. Both plugins are vulnerable to the same path-traversal issue, allowing crafted requests to reach unintended backend endpoints. If your Grafana instance uses one or both, you are at risk.

If our backend credentials are not stored in Grafana, are we still at risk?

Yes. Even if credentials are not retrievable, the vulnerability can leak internal API responses or trigger administrative actions. For example, crafted requests could modify dashboard metadata, alter alerting rules, or retrieve internal system information. The severity depends on what the backend exposes and what actions are available.

Why is this not on the CISA KEV list if it is so serious?

The CISA KEV catalog focuses on vulnerabilities with evidence of active, widespread exploitation in the wild. Recent disclosures, limited public exploit code, or exploitation primarily by targeted threat actors may not yet meet KEV criteria. This does not mean the vulnerability is low-risk; it reflects current exploit availability and threat-landscape data.

This analysis is based on the official CVE description and CVSS 3.1 scoring. Actual exploit impact depends on your Grafana configuration, backend architecture, and access controls. Always verify patch version numbers and affected product versions against the official Grafana security advisory before deploying remediations. SEC.co makes no warranty regarding the completeness or accuracy of this intelligence; use it alongside your own security assessments and vendor guidance. Proof-of-concept code or weaponized exploit details are not provided, and are outside the scope of this advisory. Source: NVD (public-domain), retrieved 2026-07-28. Analysis generated by SEC.co (claude-haiku-4-5).