Security details of IPython Parallel#

IPython Parallel exposes the full power of the Python interpreter over a TCP/IP network (or BSD socket) for the purposes of parallel computing. This feature raises the important question of IPython’s security model. This document gives details about this model and how it is implemented in IPython’s architecture.

Process and network topology#

To enable parallel computing, IPython has a number of different processes that run. These processes are discussed at length in the IPython documentation and are summarized here:

  • The IPython engine. This process is a full blown Python interpreter in which user code is executed. Multiple engines are started to make parallel computing possible.

  • The IPython hub. This process monitors a set of engines and schedulers, and keeps track of the state of the processes. It listens for registration connections from engines and clients, and monitor connections from schedulers.

  • The IPython schedulers. This is a set of processes that relay commands and results between clients and engines. They are typically on the same machine as the hub, and listen for connections from engines and clients, but connect to the Hub.

  • The IPython client. This process is typically an interactive Python process that is used to coordinate the engines to get a parallel computation done.

Collectively, these processes are called an IPython cluster, and the hub and schedulers together are referred to as the controller.

These processes communicate over any transport supported by ZeroMQ (tcp, pgm, ipc) with a well-defined topology. The IPython controller processes listen on sockets. Upon starting, an engine connects to a hub and sends a message requesting registration, to which the hub replies with connection information for the schedulers, and the engine then connects to the schedulers. These engine->hub and engine->scheduler connections persist for the lifetime of each engine.

The IPython client also connects to the controller processes using a number of sockets. This is one socket per scheduler. These connections persist for the lifetime of the client only.

A given IPython controller and set of engines typically has a relatively short lifetime, such as the duration of a single parallel computation performed by a single user. Finally, the hub, schedulers, engines, and client processes typically execute with the permissions of that same user. More specifically, the controller and engines are not executed as root or with any other superuser or shared-user permissions.

Application logic#

When running the IPython kernel to perform a parallel computation, a user connects an IPython client to send Python commands and data through the IPython schedulers to the IPython engines, where those commands are executed and the data processed.

Via the client, a user can instruct the IPython engines to execute arbitrary Python commands. These Python commands can include calls to the system shell, access the filesystem, etc., as required by the user’s application. From this perspective, when a user runs an IPython engine on a host, that engine has the same capabilities and permissions as the user themselves (as if they were logged onto the engine’s host with a terminal).

ZeroMQ and Connection files#

IPython uses ZeroMQ for networking. By default, no IPython connections are encrypted. Open ports listen only on localhost. When no encryption is used, messages are signed via HMAC digest using a shared key for authentication.

As of IPython 7.1, all connections can be authenticated and encrypted using CurveZMQ.

The key (whether the CurveZMQ key or HMAC digest key) is distributed to engines and clients via connection files.

TCP connections can be tunneled over SSH. IPython supports both shell (openssh) and paramiko based tunnels for connections.

In our architecture, the controller is the only process that listens on network ports, and is thus the main point of vulnerability. The standard model for secure connections is to designate that the controller listen on localhost, and use ssh-tunnels to connect clients and/or engines, or connect over a trusted private network.

To connect and authenticate to the controller an engine or client needs some information that the controller has stored in a JSON file. The JSON files may need to be copied to a location where the clients and engines can find them. Typically, this is the ~/.ipython/profile_default/security directory on the host where the client/engine is running, which could be on a different filesystem than the controller. Once the JSON files are copied over, everything should work fine.

Currently, there are two JSON files that the controller creates:

ipcontroller-engine.json

This JSON file has the information necessary for an engine to connect to a controller.

ipcontroller-client.json

The client’s connection information. Similar to the engine file, but lists a different collection of ports.

ipcontroller-client.json will look something like this, under default localhost circumstances:

{
  "ssh": "",
  "interface": "tcp://127.0.0.1",
  "registration": 54886,
  "control": 54888,
  "mux": 54890,
  "hb_ping": 54891,
  "hb_pong": 54892,
  "task": 54894,
  "iopub": 54896,
  "broadcast": [
    54900,
    54901
  ],
  "key": "7e99e423-c437d4daf7cf23ee84cae803",
  "location": "mylaptop",
  "pack": "json",
  "unpack": "json",
  "signature_scheme": "hmac-sha256"
}

If, however, you are running the controller on a work node on a cluster, you will likely need to use ssh tunnels to connect clients from your laptop to it. You will also probably need to instruct the controller to listen for engines coming from other work nodes on the cluster. An example of ipcontroller-client.json, as created by:

$> ipcontroller --ip=* --ssh=login.mycluster.com
{
  "ssh": "login.mycluster.com",
  "interface": "tcp://*",
  "registration": 55836,
  "control": 55837,
  "mux": 55839,
  "task": 55843,
  "task_scheme": "lru",
  "iopub": 55845,
  "notification": 55852,
  "broadcast": [
    55847,
    55848,
    55849
  ],
  "key": "70bc97ac-e66ac5143885ca8b376d4cb7",
  "location": "mylaptop",
  "pack": "json",
  "unpack": "json",
  "signature_scheme": "hmac-sha256"
}

More details of how these JSON files are used are given below.

Secure network connections#

Overview#

ZeroMQ supports encryption and authentication via a mechanism called CurveZMQ. IPython Parallel supports CurveZMQ as of version 7.1.

To enable CurveZMQ, set

c.IPController.enable_curve = True

in ipcontroller_config.py, or set the environment variable IPP_ENABLE_CURVE=1.

When using CurveZMQ, all connections are authenticated using the controller’s server key, which is distributed to engines and clients via connection files. This key is all that is needed to connect to an IPython cluster and execute code.

Additionally, when using CurveZMQ, all communication is encrypted using the controller’s server key. Each client typically generates its own unique, short-lived key pair for its side of encrypted communication. When not using CurveZMQ, messages are not encrypted.

Security without CurveZMQ#

When not using CurveZMQ, all ZeroMQ connections are not authenticated and not encrypted. For this reason, users of IPython must be very careful in managing connections, because an open TCP/IP socket presents access to arbitrary execution as the user on the engine machines. As a result, the default behavior of controller processes is to only listen for clients on the loopback interface, and remote clients must establish SSH tunnels to connect to the controller processes.

Warning

If the controller’s loopback interface is untrusted, then IPython should be considered vulnerable without CurveZMQ, and this extends to the loopback of all connected clients, which have opened a loopback port that is redirected to the controller’s loopback port.

SSH#

Without CurveZMQ, ZeroMQ sockets themselves provide no security. SSH tunnels can be used to encrypt traffic across the network, but at least loopback traffic will be unencrypted. A connection file file, such as ipcontroller-client.json, will contain information for connecting to the controller, possibly including the address of an ssh server through which the client is to tunnel. The Client object then creates tunnels using either openssh or paramiko, depending on the platform. If users do not wish to use OpenSSH or Paramiko, or the tunneling utilities are insufficient, then they may construct the tunnels themselves, and connect clients and engines as if the controller were on loopback on the connecting machine.

Authentication#

IPython uses a key-distribution model of authentication. Whether you are using CurveZMQ or message-digest signatures. In both cases, a single key is used for authentication, and the same key is distributed in ipcontroller-{client|engine}.json.

There is exactly one shared key per cluster - it must be the same everywhere. Typically, the controller creates this key, and stores it in the private connection files ipython-{engine|client}.json. These files are typically stored in the ~/.ipython/profile_<name>/security directory, and are maintained as readable only by the owner, as is common practice with a user’s keys in their .ssh directory.

The key distribution model is the same for both security implementations, however the level of authentication is substantially different.

If you are using CurveZMQ, connections are authenticated using the server key. This means connections attempted without the key will be rejected, and no messages can be sent or received by unauthenticated clients.

Authentication without CurveZMQ#

If not using CurveZMQ, connections are not authenticated, only messages are authenticated. This means that clients with access to the controller’s ports, but without the key will be able to connect and send and receive messages, but the requests will be rejected. To protect users of shared machines, HMAC digests are used to sign messages, using the shared key.

The Session object that handles the message protocol uses a unique key to verify valid messages. This can be any value specified by the user, but the default behavior is a pseudo-random 128-bit number, as generated by uuid.uuid4(). This key is used to initialize an HMAC object, which digests all messages, and includes that digest as a signature and part of the message. Every message that is unpacked (on Controller, Engine, and Client) will also be digested by the receiver, ensuring that the sender’s key is the same as the receiver’s. No messages that do not contain this key are acted upon in any way. The key itself is never sent over the network.

Warning

It is important to note that the signatures protect against unauthorized messages, but, as there is no encryption, provide exactly no protection of data privacy. It is possible, however, to use a custom serialization scheme (via Session.packer/unpacker traits) that does incorporate your own encryption scheme.

Encryption#

Messages are only encrypted when using CurveZMQ, which provides perfect-forward security by issuing short-lived keys for each session. Knowing the distributed CurveZMQ key is not enough to decrypt communication from another client.

Specific security vulnerabilities#

There are a number of potential security vulnerabilities present in IPython’s architecture. In this section we discuss those vulnerabilities and detail how the security architecture described above prevents them from being exploited.

Unauthorized clients#

The IPython client can instruct the IPython engines to execute arbitrary Python code with the permissions of the user who started the engines. If an attacker were able to connect their own hostile IPython client to the IPython controller, they could instruct the engines to execute code.

On the first level, this attack is prevented by requiring access to the controller’s ports, which are recommended to only be open on loopback if the controller is on an untrusted local network. If the attacker does have access to the Controller’s ports, then the attack is prevented by the capabilities based client authentication of the execution key. The relevant authentication information is encoded into the JSON file that clients must present to gain access to the IPython controller. By limiting the distribution of those keys, a user can grant access to only authorized persons, as with SSH keys. When using CurveZMQ, the connection will be rejected without the key. When not using CurveZMQ, requests will be rejected if not signed by the key.

It is highly unlikely that an execution key could be guessed by an attacker in a brute force guessing attack. A given instance of the IPython controller only runs for a relatively short amount of time (on the order of hours). Thus an attacker would have only a limited amount of time to test a search space of size 2**128. For added security, users can have arbitrarily long keys.

Warning

If the attacker has gained enough access to intercept loopback connections on either the controller or client, then a duplicate message can be sent. CurveZMQ prevents replay attacks. To protect against this, CurveZMQ uses nonces. recipients only allow each signature once, and consider duplicates invalid. However, the duplicate message could be sent to another recipient using the same key, and it would be considered valid.

Unauthorized engines#

If an attacker were able to connect a hostile engine to a user’s controller, the user might unknowingly send sensitive code or data to the hostile engine. This attacker’s engine would then have full access to that code and data.

This type of attack is prevented in the same way as the unauthorized client attack, by requiring the authentication key to register the engine with the client.

Unauthorized controllers#

It is also possible that an attacker could try to convince a user’s IPython client or engine to connect to a hostile IPython controller. That controller would then have full access to the code and data sent between the IPython client and the IPython engines.

Again, this attack is prevented through the capabilities in a connection file, which ensure that a client or engine connects to the correct controller. It is also important to note that the connection files also encode the IP address and port that the controller is listening on, so there is little chance of mistakenly connecting to a controller running on a different IP address and port.

When starting an engine or client, a user must specify the key to use for that connection. Thus, in order to introduce a hostile controller, the attacker must convince the user to use the key associated with the hostile controller. As long as a user is diligent in only using keys from trusted sources, this attack is not possible.

Other security measures#

A number of other measures are taken to further limit the security risks involved in running the IPython kernel.

First, by default, the IPython controller listens on random port numbers. While this can be overridden by the user, in the default configuration, an attacker would have to do a port scan to find a controller to attack. When coupled with the relatively short running time of a typical controller (on the order of hours), scans would have to be constant.

Second, much of the time, especially when run on supercomputers or clusters, the controller is running behind a firewall. Thus, for engines or client to connect to the controller:

  • The different processes have to all be behind the firewall.

or:

  • The user has to use SSH port forwarding to tunnel the connections through the firewall.

In either case, an attacker is presented with additional barriers that prevent attacking or even probing the system, because they must have access to localhost on either the controller node or the client’s machine.

Summary#

IPython’s architecture has been carefully designed with security in mind. The capabilities-based authentication model, in conjunction with CurveZMQ, ipc file permissions, and and/or SSH tunneled TCP/IP channels, address the core potential vulnerabilities in the system, while still enabling user’s to use the system in open networks.