ghsa

An information disclosure vulnerability was found in containers/podman in versions before 2.0.5. When using the deprecated Varlink API or the Docker-compatible REST API, if multiple containers are created in a short duration, the environment variables from the first container will get leaked into subsequent containers. An attacker who has control over the subsequent containers could use this flaw to gain access to sensitive information stored in such variables.

A security-check flaw was found in the way the Heketi 5 server API handled user requests. An authenticated Heketi user could send specially crafted requests to the Heketi server, resulting in remote command execution as the user running Heketi server and possibly privilege escalation.

### Summary An authenticated user can change the download folder and upload a crafted template to the specified folder lead to remote code execution ### Details example version: 0.5 file:src/pyload/webui/app/blueprints/app_blueprint.py ```python @bp.route("/render/<path:filename>", endpoint="render") def render(filename): mimetype = mimetypes.guess_type(filename)[0] or "text/html" data = render_template(filename) return flask.Response(data, mimetype=mimetype) ``` So, if we can control file in the path "pyload/webui/app/templates" in latest version and path in "module/web/media/js"(the difference is the older version0.4.20 only renders file with extension name ".js"), the render_template func will works like SSTI(server-side template injection) when render the evil file we control. in /settings page and the choose option general/general, where we can change the download folder. ![image](https://github.com/pyload/pyload/assets/48705773/0b239138-9aaa-45c4-bf84-c1c3103c452a...

A vulnerability was discovered in Rancher versions 2.0 through the aforementioned fixed versions, where users were granted access to resources regardless of the resource's API group. For example Rancher should have allowed users access to `apps.catalog.cattle.io`, but instead incorrectly gave access to `apps.*`. Resource affected include: **Downstream clusters:** apiservices clusters clusterrepos persistentvolumes storageclasses **Rancher management cluster** apprevisions apps catalogtemplates catalogtemplateversions clusteralertgroups clusteralertrules clustercatalogs clusterloggings clustermonitorgraphs clusterregistrationtokens clusterroletemplatebindings clusterscans etcdbackups nodepools nodes notifiers pipelineexecutions pipelines pipelinesettings podsecuritypolicytemplateprojectbindings projectalertgroups projectalertrules projectcatalogs projectloggings projectmonitorgraphs projectroletemplatebindings projects secrets sourcecodeproviderconfigs There is not a direct mitigati...

A vulnerability was discovered in Rancher 2.0.0 through the aforementioned patched versions, where a malicious Rancher user could craft an API request directed at the proxy for the Kubernetes API of a managed cluster to gain access to information they do not have access to. This is done by passing the "Impersonate-User" or "Impersonate-Group" header in the Connection header, which is then correctly removed by the proxy. At this point, instead of impersonating the user and their permissions, the request will act as if it was from the Rancher management server and incorrectly return the information. The vulnerability is limited to valid Rancher users with some level of permissions on the cluster. There is not a direct mitigation besides upgrading to the patched Rancher versions.

### Impact A flaw discovered in Rancher versions from 2.5.0 up to and including 2.5.9 allows an authenticated user to impersonate any user on a cluster through the Steve API proxy, without requiring knowledge of the impersonated user's credentials. This is due to the Steve API proxy not dropping the impersonation header before sending the request to the Kubernetes API. A malicious user with authenticated access to Rancher could use this to impersonate another user with administrator access in Rancher, receiving, then, administrator level access in the cluster. ### Patches Patched versions include releases 2.5.10, 2.6.0 and later versions. ### Workarounds Limit access in Rancher to trusted users. There is not a direct mitigation besides upgrading to the patched Rancher versions. ### For more information If you have any questions or comments about this advisory: * Reach out to [SUSE Rancher Security team](https://github.com/rancher/rancher/security/policy) for security related inquir...

### Impact This vulnerability only affects customers using group based authentication in Rancher versions up to and including 2.4.17, 2.5.11 and 2.6.2. When removing a Project Role associated to a group from a project, the bindings that grant access to cluster scoped resources for those subjects do not get deleted. This happens due to an incomplete authorization logic check. A user who is a member of an affected group with authenticated access to Rancher could use this to access resources they should no longer have access to. The exposure level will depend on the original permission level granted to the affected project role. ### Patches Patched versions include releases 2.4.18, 2.5.12, 2.6.3 and later versions. ### Workarounds Limit access in Rancher to trusted users. There is not a direct mitigation besides upgrading to the patched Rancher versions. ### References Cluster and project roles documentation for Rancher [2.6](https://rancher.com/docs/rancher/v2.6/en/admin-settings/rba...

Gitea 0.9.99 through 1.12.x before 1.12.6 does not prevent a git protocol path that specifies a TCP port number and also contains newlines (with URL encoding) in ParseRemoteAddr in modules/auth/repo_form.go.

Stack buffer overflow vulnerability in gitea 1.9.0 through 1.13.1 allows remote attackers to cause a denial of service (crash) via vectors related to a file path.

A flaw was found in kubevirt 0.29 and earlier. Virtual Machine Instances (VMIs) can be used to gain access to the host's filesystem. Successful exploitation allows an attacker to assume the privileges of the VM process on the host system. In worst-case scenarios an attacker can read and modify any file on the system where the VMI is running. The highest threat from this vulnerability is to data confidentiality and integrity as well as system availability.