Command execution in VMs

(This page is about qrexec v3. For qrexec v2, see here.)

The qrexec framework is used by core Qubes components to implement communication between domains. Qubes domains are isolated by design, but there is a need for a mechanism to allow the administrative domain (dom0) to force command execution in another domain (VM). For instance, when user selects an application from the KDE menu, it should be started in the selected VM. Also, it is often useful to be able to pass stdin/stdout/stderr from an application running in a VM to dom0 (and the other way around). In specific circumstances, Qubes allows VMs to be initiators of such communications (so, for example, a VM can notify dom0 that there are updates available for it).

Qrexec basics

Qrexec is built on top of vchan (a library providing data links between VMs). During domain creation a process named qrexec-daemon is started in dom0, and a process named qrexec-agent is started in the VM. They are connected over vchan channel. qrexec-daemon listens for connections from dom0 utility named qrexec-client. Typically, the first thing that a qrexec-client instance does is to send a request to qrexec-daemon to start a process (let’s name it VMprocess) with a given command line in a specified VM (someVM). qrexec-daemon assigns unique vchan connection details and sends them both to qrexec-client (in dom0) and qrexec-agent (in someVM). qrexec-client starts a vchan server which qrexec-agent connects to. Since then, stdin/stdout/stderr from the VMprocess is passed via vchan between qrexec-agent and the qrexec-client process.

So, for example, executing in dom0:

qrexec-client -d someVM user:bash

allows to work with the remote shell. The string before the first semicolon specifies what user to run the command as. Adding -e on the qrexec-client command line results in mere command execution (no data passing), and qrexec-client exits immediately after sending the execution request and receiving status code from qrexec-agent (whether the process creation succeeded). There is also the -l local_program flag – with it, qrexec-client passes stdin/stdout of the remote process to the (spawned for this purpose) local_program, not to its own stdin/stdout.

The qvm-run command is heavily based on qrexec-client. It also takes care of additional activities, e.g. starting the domain if it is not up yet and starting the GUI daemon. Thus, it is usually more convenient to use qvm-run.

There can be almost arbitrary number of qrexec-client processes for a domain (so, connected to the same qrexec-daemon, same domain) – their data is multiplexed independently. Number of available vchan channels is the limiting factor here, it depends on the underlying hypervisor.

Qubes RPC services

Some tasks (like inter-vm file copy) share the same rpc-like structure: a process in one VM (say, file sender) needs to invoke and send/receive data to some process in other VM (say, file receiver). Thus, the Qubes RPC framework was created, facilitating such actions.

Obviously, inter-VM communication must be tightly controlled to prevent one VM from taking control over other, possibly more privileged, VM. Therefore the design decision was made to pass all control communication via dom0, that can enforce proper authorization. Then, it is natural to reuse the already-existing qrexec framework.

Also, note that bare qrexec provides VM <-> dom0 connectivity, but the command execution is always initiated by dom0. There are cases when VM needs to invoke and send data to a command in dom0 (e.g. to pass information on newly installed .desktop files). Thus, the framework allows dom0 to be the rpc target as well.

Thanks to the framework, RPC programs are very simple – both rpc client and server just use their stdin/stdout to pass data. The framework does all the inner work to connect these processes to each other via qrexec-daemon and qrexec-agent. Additionally, disposable VMs are tightly integrated – rpc to a DisposableVM is identical to rpc to a normal domain, all one needs is to pass $dispvm as the remote domain name.

Qubes RPC administration

(TODO: fix for non-linux dom0)

In dom0, there is a bunch of files in /etc/qubes-rpc/policy directory, whose names describe the available rpc actions. Their content is the rpc access policy database. Currently defined actions are:


These files contain lines with the following format:

srcvm destvm (allow|deny|ask)[,user=user_to_run_as][,target=VM_to_redirect_to]

You can specify srcvm and destvm by name, or by one of $anyvm, $dispvm, dom0 reserved keywords (note string dom0 does not match the $anyvm pattern; all other names do). Only $anyvm keyword makes sense in srcvm field (service calls from dom0 are currently always allowed, $dispvm means “new VM created for this particular request,” so it is never a source of request). Currently there is no way to specify source VM by type. Whenever a rpc request for action X is received, the first line in /etc/qubes-rpc/policy/X that match srcvm/destvm is consulted to determine whether to allow rpc, what user account the program should run in target VM under, and what VM to redirect the execution to. Note that if the request is redirected (target= parameter), policy action remains the same - even if there is another rule which would otherwise deny such request. If the policy file does not exist, user is prompted to create one; if still there is no policy file after prompting, the action is denied.

In the target VM, the /etc/qubes-rpc/RPC_ACTION_NAME must exist, containing the file name of the program that will be invoked, or being that program itself

  • in which case it must have executable permission set (chmod +x).

In the src VM, one should invoke the client via:

/usr/lib/qubes/qrexec-client-vm target_vm_name RPC_ACTION_NAME rpc_client_path client arguments

Note that only stdin/stdout is passed between rpc server and client – notably, no command line arguments are passed. Source VM name is specified by QREXEC_REMOTE_DOMAIN environment variable. By default, stderr of client and server is logged to respective /var/log/qubes/qrexec.XID files. It is also possible to call service without specific client program - in which case server stdin/out will be connected with the terminal:

/usr/lib/qubes/qrexec-client-vm target_vm_name RPC_ACTION_NAME

Be very careful when coding and adding a new rpc service. Unless the offered functionality equals full control over the target (it is the case with e.g. qubes.VMShell action), any vulnerability in an rpc server can be fatal to Qubes security. On the other hand, this mechanism allows to delegate processing of untrusted input to less privileged (or disposable) AppVMs, thus wise usage of it increases security.

For example, this command will run the firefox command in a DisposableVM based on work:

$ qvm-run --dispvm=work firefox

By contrast, consider this command:

$ qvm-run --dispvm=work --service qubes.StartApp+firefox

This will look for a firefox.desktop file in a standard location in a DisposableVM based on work, then launch the application described by that file. The practical difference is that the bare qvm-run command uses the qubes.VMShell service, which allows you to run an arbitrary command with arbitrary arguments, essentially providing full control over the target VM. By contrast, the qubes.StartApp service allows you to run only applications that are advertised in /usr/share/applications (or other standard locations) without control over the arguments, so giving a VM access to qubes.StartApp is much safer. While there isn’t much practical difference between the two commands above when starting an application from dom0 in Qubes 4.0, there is a significant security risk when launching applications from a domU (e.g., from a separate GUI domain). This is why qubes.StartApp uses our standard qrexec argument grammar to strictly filter the permissible grammar of the Exec= lines in .desktop files that are passed from untrusted domUs to dom0, thereby protecting dom0 from command injection by maliciously-crafted .desktop files.

Extra keywords available in Qubes 4.0 and later

This section is about a not-yet-released version, some details may change

In Qubes 4.0, target VM can be specified also as $dispvm:DISP_VM, which is very similar to $dispvm but forces using a particular VM (DISP_VM) as a base VM to be started as DisposableVM. For example:

anon-whonix $dispvm:anon-whonix-dvm allow

Adding such policy itself will not force usage of this particular DISP_VM - it will only allow it when specified by the caller. But $dispvm:DISP_VM can also be used as target in request redirection, so it is possible to force particular DISP_VM usage, when caller didn’t specify it:

anon-whonix $dispvm allow,target=$dispvm:anon-whonix-dvm

Note that without redirection, this rule would allow using default Disposable VM (default_dispvm VM property, which itself defaults to global default_dispvm property). Also note that the request will be allowed (allow action) even if there is no second rule allowing calls to $dispvm:anon-whonix-dvm, or even if there is a rule explicitly denying it. This is because the redirection happens after considering the action.

In Qubes 4.0 there are also additional methods to specify source/target VM:

  • $tag:some-tag - meaning a VM with tag some-tag
  • $type:type - meaning a VM of type (like AppVM, TemplateVM etc)

Target VM can be also specified as $default, which matches the case when calling VM didn’t specified any particular target (either by using $default target, or empty target).

In Qubes 4.0 policy confirmation dialog (ask action) allow the user to specify target VM. User can choose from VMs that, according to policy, would lead to ask or allow actions. It is not possible to select VM that policy would deny. By default no VM is selected, even if the caller provided some, but policy can specify default value using default_target= parameter. For example:

work-mail work-archive allow
work-mail $tag:work ask,default_target=work-files
work-mail $default  ask,default_target=work-files

The first rule allow call from work-mail to work-archive, without any confirmation. The second rule will ask the user about calls from work-mail VM to any VM with tag work. And the confirmation dialog will have work-files VM chosen by default, regardless of the VM specified by the caller (work-mail VM). The third rule allow the caller to not specify target VM at all and let the user choose, still - from VMs with tag work (and work-archive, regardless of tag), and with work-files as default.

Service argument in policy

Sometimes just service name isn’t enough to make reasonable qrexec policy. One example of such a situation is qrexec-based USB passthrough - using just service name isn’t possible to express the policy “allow access to device X and deny to others”. It also isn’t feasible to create a separate service for every device…

For this reason, starting with Qubes 3.2, it is possible to specify a service argument, which will be subject to policy. Besides the above example of USB passthrough, a service argument can make many service policies more fine-grained and easier to write precise policy with “allow” and “deny” actions, instead of “ask” (offloading additional decisions to the user). And generally the less choices the user must make, the lower the chance to make a mistake.

The syntax is simple: when calling a service, add an argument to the service name separated with + sign, for example:

/usr/lib/qubes/qrexec-client-vm target_vm_name RPC_ACTION_NAME+ARGUMENT

Then create a policy as usual, including the argument (/etc/qubes-rpc/policy/RPC_ACTION_NAME+ARGUMENT). If the policy for the specific argument is not set (file does not exist), then the default policy for this service is loaded (/etc/qubes-rpc/policy/RPC_ACTION_NAME).

In target VM (when the call is allowed) the service file will searched as:

  • /etc/qubes-rpc/RPC_ACTION_NAME+ARGUMENT
  • /etc/qubes-rpc/RPC_ACTION_NAME

In any case, the script will receive ARGUMENT as its argument and additionally as QREXEC_SERVICE_ARGUMENT environment variable. This means it is also possible to install a different script for a particular service argument.

See below for an example service using an argument.

Revoking “Yes to All” authorization

Qubes RPC policy supports “ask” action. This will prompt the user whether given RPC call should be allowed. That prompt window also has a “Yes to All” option, which will allow the action and add a new entry to the policy file, which will unconditionally allow further calls for the given service-srcVM-dstVM tuple.

In order to remove such authorization, issue this command from a dom0 terminal (for qubes.Filecopy service):

sudo nano /etc/qubes-rpc/policy/qubes.Filecopy

and then remove the first line(s) (before the first ## comment) which are the “Yes to All” results.

Qubes RPC example

We will show the necessary files to create an rpc call that adds two integers on the target and returns back the result to the invoker.

  • rpc client code (/usr/bin/our_test_add_client):

    echo $1 $2    # pass data to rpc server
    exec cat >&$SAVED_FD_1 # print result to the original stdout, not to the other rpc endpoint
  • rpc server code (/usr/bin/our_test_add_server)

    read arg1 arg2 # read from stdin, which is received from the rpc client
    echo $(($arg1+$arg2)) # print to stdout - so, pass to the rpc client
  • policy file in dom0 (/etc/qubes-rpc/policy/test.Add )

    $anyvm $anyvm ask
  • server path definition ( /etc/qubes-rpc/test.Add)

  • invoke rpc via

    /usr/lib/qubes/qrexec-client-vm target_vm test.Add /usr/bin/our_test_add_client 1 2

and we should get “3” as answer, after dom0 allows it.

Note: For a real world example of writing a qrexec service, see this blog post.

Qubes RPC example - with argument usage

We will show the necessary files to create an rpc call that reads a specific file from a predefined directory on the target. Besides really naive storage, it may be a very simple password manager. Additionally, in this example a simplified workflow will be used - server code placed directly in the service definition file (in /etc/qubes-rpc directory). And no separate client script will be used.

  • rpc server code (/etc/qubes-rpc/test.File)

    argument="$1" # service argument, also available as $QREXEC_SERVICE_ARGUMENT
    if [ -z "$argument" ]; then
      echo "ERROR: No argument given!"
      exit 1
    # service argument is already sanitized by qrexec framework and it is
    # guaranteed to not contain any space or /, so no need for additional path
    # sanitization
    cat "/home/user/rpc-file-storage/$argument"
  • specific policy file in dom0 (/etc/qubes-rpc/policy/test.File+testfile1 )

    source_vm1 target_vm allow
  • another specific policy file in dom0 (/etc/qubes-rpc/policy/test.File+testfile2 )

    source_vm2 target_vm allow
  • default policy file in dom0 (/etc/qubes-rpc/policy/test.File )

    $anyvm $anyvm deny
  • invoke rpc from source_vm1 via

    /usr/lib/qubes/qrexec-client-vm target_vm test.File+testfile1

    and we should get content of /home/user/rpc-file-storage/testfile1 as answer.

  • also possible to invoke rpc from source_vm2 via

    /usr/lib/qubes/qrexec-client-vm target_vm test.File+testfile2

    But when invoked with other argument or from different VM, it should be denied.

Qubes RPC internals

(This is about the implementation of qrexec v3. For the implementation of qrexec v2, see here.)

Qrexec framework consists of a number of processes communicating with each other using common IPC protocol (described in detail below). Components residing in the same domain (qrexec-client-vm to qrexec-agent, qrexec-client to qrexec-daemon) use pipes as the underlying transport medium, while components in separate domains (qrexec-daemon to qrexec-agent, data channel between qrexec-agents) use vchan link. Because of vchan limitation, it is not possible to establish qrexec connection back to the source domain.

Dom0 tools implementation

  • /usr/lib/qubes/qrexec-daemon: One instance is required for every active domain. Responsible for:
    • Handling execution and service requests from dom0 (source: qrexec-client).
    • Handling service requests from the associated domain (source: qrexec-client-vm, then qrexec-agent).
  • Command line: qrexec-daemon domain-id domain-name [default user]
  • domain-id: Numeric Qubes ID assigned to the associated domain.
  • domain-name: Associated domain name.
  • default user: Optional. If passed, qrexec-daemon uses this user as default for all execution requests that don’t specify one.
  • /usr/lib/qubes/qrexec-policy: Internal program used to evaluate the RPC policy and deciding whether a RPC call should be allowed.
  • /usr/lib/qubes/qrexec-client: Used to pass execution and service requests to qrexec-daemon. Command line parameters:
    • -d target-domain-name: Specifies the target for the execution/service request.
    • -l local-program: Optional. If present, local-program is executed and its stdout/stdin are used when sending/receiving data to/from the remote peer.
    • -e: Optional. If present, stdout/stdin are not connected to the remote peer. Only process creation status code is received.
    • -c <request-id,src-domain-name,src-domain-id>: used for connecting a VM-VM service request by qrexec-policy. Details described below in the service example.
    • cmdline: Command line to pass to qrexec-daemon as the execution/service request. Service request format is described below in the service example.

Note: None of the above tools are designed to be used by users directly.

VM tools implementation

  • qrexec-agent: One instance runs in each active domain. Responsible for:
    • Handling service requests from qrexec-client-vm and passing them to connected qrexec-daemon in dom0.
    • Executing associated qrexec-daemon execution/service requests.
  • Command line parameters: none.
  • qrexec-client-vm: Runs in an active domain. Used to pass service requests to qrexec-agent.
  • Command line: qrexec-client-vm target-domain-name service-name local-program [local program arguments]
  • target-domain-name: Target domain for the service request. Source is the current domain.
  • service-name: Requested service name.
  • local-program: local-program is executed locally and its stdin/stdout are connected to the remote service endpoint.

Qrexec protocol details

Qrexec protocol is message-based. All messages share a common header followed by an optional data packet.

/* uniform for all peers, data type depends on message type */
struct msg_header {
   uint32_t type;           /* message type */
   uint32_t len;            /* data length */

When two peers establish connection, the server sends MSG_HELLO followed by peer_info struct:

struct peer_info {
   uint32_t version; /* qrexec protocol version */

The client then should reply with its own MSG_HELLO and peer_info. The lower of two versions define protocol used for this connection. If either side does not support this version, the connection is closed.

Details of all possible use cases and the messages involved are described below.

dom0: request execution of some_command in domX and pass stdin/stdout

  • dom0: qrexec-client is invoked in dom0 as follows:

    qrexec-client -d domX [-l local_program] user:some_command

    • user may be substituted with the literal DEFAULT. In that case, default Qubes user will be used to execute some_command.
  • dom0: qrexec-client sets QREXEC_REMOTE_DOMAIN environment variable to domX.
  • dom0: If local_program is set, qrexec-client executes it and uses that child’s stdin/stdout in place of its own when exchanging data with qrexec-agent later.
  • dom0: qrexec-client connects to domX’s qrexec-daemon.
  • dom0: qrexec-daemon sends MSG_HELLO header followed by peer_info to qrexec-client.
  • dom0: qrexec-client replies with MSG_HELLO header followed by peer_info to qrexec-daemon.
  • dom0: qrexec-client sends MSG_EXEC_CMDLINE header followed by exec_params to qrexec-daemon.

      /* variable size */
      struct exec_params {
         uint32_t connect_domain; /* target domain id */
         uint32_t connect_port;   /* target vchan port for i/o exchange */
         char cmdline[0];         /* command line to execute, size = msg_header.len - sizeof(struct exec_params) */

    In this case, connect_domain and connect_port are set to 0.

  • dom0: qrexec-daemon replies to qrexec-client with MSG_EXEC_CMDLINE header followed by exec_params, but with empty cmdline field. connect_domain is set to Qubes ID of domX and connect_port is set to a vchan port allocated by qrexec-daemon.
  • dom0: qrexec-daemon sends MSG_EXEC_CMDLINE header followed by exec_params to the associated domX qrexec-agent over vchan. connect_domain is set to 0 (dom0), connect_port is the same as sent to qrexec-client. cmdline is unchanged except that the literal DEFAULT is replaced with actual user name, if present.
  • dom0: qrexec-client disconnects from qrexec-daemon.
  • dom0: qrexec-client starts a vchan server using the details received from qrexec-daemon and waits for connection from domX’s qrexec-agent.
  • domX: qrexec-agent receives MSG_EXEC_CMDLINE header followed by exec_params from qrexec-daemon over vchan.
  • domX: qrexec-agent connects to qrexec-client over vchan using the details from exec_params.
  • domX: qrexec-agent executes some_command in domX and connects the child’s stdin/stdout to the data vchan. If the process creation fails, qrexec-agent sends MSG_DATA_EXIT_CODE to qrexec-client followed by the status code (int) and disconnects from the data vchan.
  • Data read from some_command’s stdout is sent to the data vchan using MSG_DATA_STDOUT by qrexec-agent. qrexec-client passes data received as MSG_DATA_STDOUT to its own stdout (or to local_program’s stdin if used).
  • qrexec-client sends data read from local stdin (or local_program’s stdout if used) to qrexec-agent over the data vchan using MSG_DATA_STDIN. qrexec-agent passes data received as MSG_DATA_STDIN to some_command’s stdin.
  • MSG_DATA_STDOUT or MSG_DATA_STDIN with data len field set to 0 in msg_header is an EOF marker. Peer receiving such message should close the associated input/output pipe.
  • When some_command terminates, domX’s qrexec-agent sends MSG_DATA_EXIT_CODE header to qrexec-client followed by the exit code (int). qrexec-agent then disconnects from the data vchan.

domY: invoke execution of qubes service qubes.SomeRpc in domX and pass stdin/stdout

  • domY: qrexec-client-vm is invoked as follows:

    `qrexec-client-vm domX qubes.SomeRpc local_program [params]`
  • domY: qrexec-client-vm connects to qrexec-agent (via local socket/named pipe).
  • domY: qrexec-client-vm sends trigger_service_params data to qrexec-agent (without filling the request_id field):

       struct trigger_service_params {
          char service_name[64];
          char target_domain[32];
          struct service_params request_id; /* service request id */
       struct service_params {
          char ident[32];
  • domY: qrexec-agent allocates a locally-unique (for this domain) request_id (let’s say 13) and fills it in the trigger_service_params struct received from qrexec-client-vm.
  • domY: qrexec-agent sends MSG_TRIGGER_SERVICE header followed by trigger_service_params to qrexec-daemon in dom0 via vchan.
  • dom0: domY’s qrexec-daemon executes qrexec-policy: qrexec-policy domY_id domY domX qubes.SomeRpc 13.
  • dom0: qrexec-policy evaluates if the RPC should be allowed or denied. If the action is allowed it returns 0, if the action is denied it returns 1.
  • dom0: domY’s qrexec-daemon checks the exit code of qrexec-policy.
    • If qrexec-policy returned not 0: domY’s qrexec-daemon sends MSG_SERVICE_REFUSED header followed by service_params to domY’s qrexec-agent. service_params.ident is identical to the one received. domY’s qrexec-agent disconnects its qrexec-client-vm and RPC processing is finished.
    • If qrexec-policy returned 0, RPC processing continues.
  • dom0: if qrexec-policy allowed the RPC, it executed qrexec-client -d domX -c 13,domY,domY_id user:QUBESRPC qubes.SomeRpc domY.
  • dom0: qrexec-client sets QREXEC_REMOTE_DOMAIN environment variable to domX.
  • dom0: qrexec-client connects to domX’s qrexec-daemon.
  • dom0: domX’s qrexec-daemon sends MSG_HELLO header followed by peer_info to qrexec-client.
  • dom0: qrexec-client replies with MSG_HELLO header followed by peer_info to domX’sqrexec-daemon.
  • dom0: qrexec-client sends MSG_EXEC_CMDLINE header followed by exec_params to domX’sqrexec-daemon

      /* variable size */
      struct exec_params {
         uint32_t connect_domain; /* target domain id */
         uint32_t connect_port;   /* target vchan port for i/o exchange */
         char cmdline[0];         /* command line to execute, size = msg_header.len - sizeof(struct exec_params) */

    In this case, connect_domain is set to id of domY (from the -c parameter) and connect_port is set to 0. cmdline field contains the RPC to execute, in this case user:QUBESRPC qubes.SomeRpc domY.

  • dom0: domX’s qrexec-daemon replies to qrexec-client with MSG_EXEC_CMDLINE header followed by exec_params, but with empty cmdline field. connect_domain is set to Qubes ID of domX and connect_port is set to a vchan port allocated by domX’s qrexec-daemon.
  • dom0: domX’s qrexec-daemon sends MSG_EXEC_CMDLINE header followed by exec_params to domX’s qrexec-agent. connect_domain and connect_port fields are the same as in the step above. cmdline is set to the one received from qrexec-client, in this case user:QUBESRPC qubes.SomeRpc domY.
  • dom0: qrexec-client disconnects from domX’s qrexec-daemon after receiving connection details.
  • dom0: qrexec-client connects to domY’s qrexec-daemon and exchanges MSG_HELLO as usual.
  • dom0: qrexec-client sends MSG_SERVICE_CONNECT header followed by exec_params to domY’s qrexec-daemon. connect_domain is set to ID of domX (received from domX’s qrexec-daemon) and connect_port is the one received as well. cmdline is set to request ID (13 in this case).
  • dom0: domY’s qrexec-daemon sends MSG_SERVICE_CONNECT header followed by exec_params to domY’s qrexec-agent. Data fields are unchanged from the step above.
  • domY: qrexec-agent starts a vchan server on the port received in the step above. It acts as a qrexec-client in this case because this is a VM-VM connection.
  • domX: qrexec-agent connects to the vchan server of domY’s qrexec-agent (connection details were received before from domX’s qrexec-daemon).
  • After that, connection follows the flow of the previous example (dom0-VM).