An improper limitation of path name flaw was found in containernetworking/cni in versions before 0.8.1. When specifying the plugin to load in the 'type' field in the network configuration, it is possible to use special elements such as "../" separators to reference binaries elsewhere on the system. This flaw allows an attacker to execute other existing binaries other than the cni plugins/types, such as 'reboot'. The highest threat from this vulnerability is to confidentiality, integrity, as well as system availability.
In containerd (an industry-standard container runtime) before versions 1.3.10 and 1.4.4, containers launched through containerd's CRI implementation (through Kubernetes, crictl, or any other pod/container client that uses the containerd CRI service) that share the same image may receive incorrect environment variables, including values that are defined for other containers. If the affected containers have different security contexts, this may allow sensitive information to be unintentionally shared. If you are not using containerd's CRI implementation (through one of the mechanisms described above), you are not vulnerable to this issue. If you are not launching multiple containers or Kubernetes pods from the same image which have different environment variables, you are not vulnerable to this issue. If you are not launching multiple containers or Kubernetes pods from the same image in rapid succession, you have reduced likelihood of being vulnerable to this issue This vulnerability has been fixed in containerd 1.3.10 and containerd 1.4.4. Users should update to these versions.
Hyperledger Besu is an open-source, MainNet compatible, Ethereum client written in Java. In Besu before version 1.5.1 there is a denial-of-service vulnerability involving the HTTP JSON-RPC API service. If username and password authentication is enabled for the HTTP JSON-RPC API service, then prior to making any requests to an API endpoint the requestor must use the login endpoint to obtain a JSON web token (JWT) using their credentials. A single user can readily overload the login endpoint with invalid requests (incorrect password). As the supplied password is checked for validity on the main vertx event loop and takes a relatively long time this can cause the processing of other valid requests to fail. A valid username is required for this vulnerability to be exposed. This has been fixed in version 1.5.1.
Dex is a federated OpenID Connect provider written in Go. In Dex before version 2.27.0 there is a critical set of vulnerabilities which impacts users leveraging the SAML connector. The vulnerabilities enables potential signature bypass due to issues with XML encoding in the underlying Go library. The vulnerabilities have been addressed in version 2.27.0 by using the xml-roundtrip-validator from Mattermost (see related references).
Hyperledger Indy Node is the server portion of a distributed ledger purpose-built for decentralized identity. In Hyperledger Indy before version 1.12.4, there is lack of signature verification on a specific transaction which enables an attacker to make certain unauthorized alterations to the ledger. Updating a DID with a nym transaction will be written to the ledger if neither ROLE or VERKEY are being changed, regardless of sender. A malicious DID with no particular role can ask an update for another DID (but cannot modify its verkey or role). This is bad because 1) Any DID can write a nym transaction to the ledger (i.e., any DID can spam the ledger with nym transactions), 2) Any DID can change any other DID's alias, 3) The update transaction modifies the ledger metadata associated with a DID.
osquery is a SQL powered operating system instrumentation, monitoring, and analytics framework. In osquery before version 4.6.0, by using sqlite's ATTACH verb, someone with administrative access to osquery can cause reads and writes to arbitrary sqlite databases on disk. This _does_ allow arbitrary files to be created, but they will be sqlite databases. It does not appear to allow existing non-sqlite files to be overwritten. This has been patched in osquery 4.6.0. There are several mitigating factors and possible workarounds. In some deployments, the people with access to these interfaces may be considered administrators. In some deployments, configuration is managed by a central tool. This tool can filter for the `ATTACH` keyword. osquery can be run as non-root user. Because this also limits the desired access levels, this requires deployment specific testing and configuration.
Nolan Ray from Apple Information Security identified a security vulnerability in Spinnaker, all versions prior to version 1.23.4, 1.22.4 or 1.21.5. The vulnerability exists within the handling of SpEL expressions that allows an attacker to read and write arbitrary files within the orca container via authenticated HTTP POST requests.
containerd is an industry-standard container runtime and is available as a daemon for Linux and Windows. In containerd before versions 1.3.9 and 1.4.3, the containerd-shim API is improperly exposed to host network containers. Access controls for the shim’s API socket verified that the connecting process had an effective UID of 0, but did not otherwise restrict access to the abstract Unix domain socket. This would allow malicious containers running in the same network namespace as the shim, with an effective UID of 0 but otherwise reduced privileges, to cause new processes to be run with elevated privileges. This vulnerability has been fixed in containerd 1.3.9 and 1.4.3. Users should update to these versions as soon as they are released. It should be noted that containers started with an old version of containerd-shim should be stopped and restarted, as running containers will continue to be vulnerable even after an upgrade. If you are not providing the ability for untrusted users to start containers in the same network namespace as the shim (typically the "host" network namespace, for example with docker run --net=host or hostNetwork: true in a Kubernetes pod) and run with an effective UID of 0, you are not vulnerable to this issue. If you are running containers with a vulnerable configuration, you can deny access to all abstract sockets with AppArmor by adding a line similar to deny unix addr=@**, to your policy. It is best practice to run containers with a reduced set of privileges, with a non-zero UID, and with isolated namespaces. The containerd maintainers strongly advise against sharing namespaces with the host. Reducing the set of isolation mechanisms used for a container necessarily increases that container's privilege, regardless of what container runtime is used for running that container.