Scroll to navigation

autopkgtest(1) General Commands Manual autopkgtest(1)


autopkgtest - test an installed binary package using the source package's tests


autopkgtest [options...] [testbinary...] testsrc -- virt-server [virt-server-arg...]


autopkgtest runs tests on binary Debian packages, as installed on a system (called "testbed"). The tests are supplied in the source package.

autopkgtest runs each test supplied by a particular package and reports the results. It drives the specified virtualisation regime as appropriate, parses the test description metadata, and arranges for data to be copied to and from the testbed as required.

See /usr/share/doc/autopkgtest/README.running-tests.rst.gz for an introduction about how to use autopkgtest.


Positional (non-option) arguments specify exactly one source package (containing the test code) and optionally some binary packages to test.

testsrc can be one of:

.dsc file
Run tests from Debian .dsc source package. By default the package will also be built and the resulting binaries will be used to satisfy test dependencies; to disable that, specify the -B/--no-built-binaries option.

Run tests from a Debian source tree directory. If that is an unbuilt tree, this is very similar to specifying a .dsc. If that is a built tree, all test dependencies get satisfied by archive packages, unless you explicitly specify locally built .debs as well.

Attention: If you just specify a bare directory name which is a legal Debian source package name, it will be interpreted as the latter (see below). In this case, prefix the directory name with ./.

If no source package is specified on the command line and the current directory is a Debian source package, this will be tested.

Downloads the given source package name with apt-get source in the testbed and run its tests. This is similar to specifying a .dsc but avoids copying the source from the host to the testbed. Possibly built binaries (if the test specifies build-needed) will not be used to satisfy dependencies, as usually in this mode you want to test binaries from a real archive.

Git-clones the given URL (which must contain an unbuilt Debian source tree) and runs the tests from that. If branch is given, this branch will be checked out instead of the default (usually "master"). This can also be a more general refspec such as #refs/pull/123/head" for a GitHub pull request.

This is very similar to cloning manually and specifying the checkout directory as test; i. e. this is commonly used with --no-built-binaries. The git package will be installed if necessary.

.changes file
Run tests from the .dsc source package in the given .changes file. If the .changes contains .deb packages, they will be used for the test. Acts as if you had specified the .debs and .dsc from the .changes file as explicit arguments. Note that if the .changes contains only debs, the corresponding .dsc still needs to be specified alongside, or the current directory must be the source package.

All other positional arguments must be .deb binary packages. They will be used for both build and test dependencies of the source package. If any binary package is given, then --no-built-binaries is implied.


Binaries from unbuilt source packages (see above) will not be built or ignored, and dependencies are satisfied with packages from the archive. Note that the source package still gets built if a test requires build-needed.

Read the test metadata from PATH instead of debian/tests/control.

Run only the given test name (from test control file). If this option is used more than once, all the named tests are run. This replaces --testname, which is deprecated.

Skip the given test name (from test control file). If this option is used more than once, all the named tests are skipped.


If you don't specify any option, autopkgtest only writes its output/results to stderr.

Specifies that test artifacts (stderr and stdout from the tests, the log file, built binary packages etc.) should be placed in the given directory. dir must not exist yet or be empty, otherwise autopkgtest will refuse to use it.

Specifies that the trace log should be written to logfile instead of to output-dir.

Specifies that a summary of the outcome should be written to summary. The events in the summary are written to the log in any case.

Do not send a copy of autopkgtest's trace logstream to stderr. This option does not affect the copy sent to logfile or output-dir. Note that without the trace logstream it can be very hard to diagnose problems.


Run commands after opening the testbed. This can be used to do anything that isn't supported by an existing autopkgtest command. If commands is an existing file name, the commands are read from that; otherwise it is a string with the actual commands that gets run as-is. File names without directory will be searched in both the current directory and in /usr/share/autopkgtest/setup-commands/ so you do not need to give the full path for setup scripts shipped with autopkgtest.

Normally, if the setup commands fail, autopkgtest will consider this a transient testbed error (exit code 16). However, if the setup commands exit with code 100, autopkgtest will consider this an "erroneous package" (exit code 12) instead, so this can be used to e. g. detect upgrade errors to a new version. Note that apt exits with exit code 100 in these cases.

This option can be specified multiple times.

If --user is given or the test bed provides a suggested-normal-user capability, the $AUTOPKGTEST_NORMAL_USER environment variable will be set to that user.

If the setup commands affect anything in boot directories (like /boot or /lib/systemd/system) and the testbed supports rebooting, the testbed will be rebooted after the setup commands. This can be suppressed by creating a file /run/autopkgtest_no_reboot.stamp.

Run commands after the --setup-commands, and after every reboot. For example, these commands could be used to add files in a tmpfs.

These commands never cause the testbed to be rebooted (because that could lead to an infinite loop). Otherwise, they are just like the --setup-commands.

This option can be specified multiple times.

Add the given apt source to /etc/apt/sources.list.d and update it, before running any --setup-commands.

This option can be specified multiple times.

Add the given apt RELEASE to /etc/apt/sources.list.d and update it, before running any --setup-commands. The mirror and components to use are copied from the very first existing APT sources.list entry. Both binary ("deb") and source ("deb-src") entries are added.

This option can be specified multiple times.

Run apt-get update and apt-get dist-upgrade -y in the testbed before running the tests. Any --add-apt-source or --apt-pocket options take effect first, so this will upgrade packages from those sources if appropriate.

Set's APT::Default-Release value to the provided value. For apt pinning (related to --apt-pocket, and --pin-packages) to work properly, APT::Default-Release must be set to the release that should provide the packages that are not pinned. For Debian and Ubuntu, this is normally automatically detected from the first entry in /etc/apt/sources.list.

Add apt sources for release-pocket. This finds the first deb line in /etc/apt/sources.list which does not already specify a pocket and adds a deb and deb-src line with that pocket to /etc/apt/sources.list.d/pocket.list. This also calls apt-get update for the new pocket (but not for anything else).

If a package list is given after =, set up apt pinning to use only those packages from pocket. An entry "src:srcname" expands to all binary packages built by that source. This can be used for minimizing dependencies taken from pocket so that package updates in that pocket can be tested independently from each other for better isolation. Attention: This does not currently resolve some situations where dependencies of the given packages can only be resolved in the given pocket. In that case the apt pinning will be removed and package installation will be retried with the entirety of pocket.

Copy file or directory from host into testbed after opening. This happens before --setup-commands thus you can use these files in the setup commands.

Set arbitrary environment variable in the build and test. Can be specified multiple times.

Set up apt pinning to use only those packages from RELEASE. An entry "src:srcname" expands to all binary packages built by that source. This can be used for minimizing dependencies taken from RELEASE so that package updates in that release can be tested independently from each other for better isolation.

Disable the apt-get fallback which is used with --apt-pocket or --pin-packages in case installation of dependencies fails due to strict pinning.

If a test would normally be skipped because it has Restrictions: RESTRICTION, run it anyway. Can be specified multiple times.

For example, you might ignore the restriction isolation-machine when using the null virtualization server if you know that autopkgtest itself is running on an expendable virtual machine. These options also work for unknown restrictions, so they can be used when experimenting with new restrictions.


Run builds and tests as user on the testbed. This needs root on the testbed; if root on the testbed is not available then builds and tests run as whatever user is provided.

Prefixes debian/rules binary with gain-root. The default is not to use anything, except that if --user is supplied or root on the testbed is not available the default is fakeroot.


Include additional debugging information in the trace log. Each additional -d increases the debugging level; the current maximum is -ddd. If you like to see what's going on, -d or -dd is recommended.

Run an interactive shell in the testbed after a failed build, test, or dependency installation.

Run an interactive shell in the testbed after every test.


Use a different timeout for operations on or with the testbed. There are five timeouts affected by five values of which: short: supposedly short operations like setting up the testbed's apt and checking the state (default: 100s); install: installation of packages including dependencies (default: 3,000s); test: test runs (default: 10,000s); copy: copy files/directories between host and testbed (default: 300s); and build: builds (default: 100,000s). The value must be specified as an integer number of seconds.

Multiply all of the default timeouts by the specified factor (see --timeout-which above). Only the defaults are affected; explicit timeout settings are used exactly as specified.


When running commands on the testbed, sets the LANG environment variable to langval. The default in autopkgtest is to set it to C.UTF-8.


Disable automatic test generation with autodep8, even if it is installed. In that case, packages without tests will exit with code 8 ("No tests in this package") just like without autodep8.

Set parallel=N DEB_BUILD_OPTION for building packages. By default this is the number of available processors. This is mostly useful in containers where you can restrict the available RAM, but not restrict the number of CPUs.

Define how to handle the needs-internet restriction. With "try" tests with needs-internet restrictions will be run, but if they fail they will be treated as flaky tests. With "skip" these tests will skipped immediately and will not be run. With "run" the restriction is basically ignored, this is the default.

Validate the test control file and exit without running any tests.

Show command line help and exit.


Specifies the virtualisation regime server, as a command and arguments to invoke. virt-server must be an existing autopkgtest virtualization server such as schroot or qemu.

All the remaining arguments and options after -- are passed to the virtualisation server program. See the manpages of the individual servers for how to use them.


During a normal test run, one line is printed for each test. This consists of a short string identifying the test, some horizontal whitespace, and one of PASS, PASS details, FAIL reason, SKIP reason, or FLAKY reason where the pass/fail indication is separated by any reason or details by some horizontal whitespace.

The string to identify the test consists of a short alphanumeric string invented by autopkgtest to distinguish different command-line arguments, the argid, followed by a hyphen and the test name.

SKIP indicates that a test was not run, or that the test code was started but detected that the test could not complete, for instance because a required resource was not available.

FLAKY indicates that a test would ordinarily have failed, but because this particular test is known to be unreliable, the failure was ignored.

Sometimes a SKIP will be reported when the name of the test is not known or not applicable: for example, when there are no tests in the package, or a there is a test stanza which contains features not understood by this version of autopkgtest. In this case * will appear where the name of the test should be.

If autopkgtest detects that erroneous package(s) are involved, it will print the two lines blame: blamed-thing... and badpkg: message. Here each whitespace-separated blamed-thing is one of arg:argument (representing a pathname found in a command line argument), dsc:package (a source package name), deb:package (a binary package name) or possibly other strings to be determined. This indicates which arguments and/or packages might have contributed to the problem; the ones which were processed most recently and which are therefore most likely to be the cause of a problem are listed last.


If you use lots of options or nontrivial virt server arguments, you can put any part of the command line into a text file, with one line per option. E. g. you can create a file sid.cfg with contents like


and then run

autopkgtest foo_1_amd64.changes @sid.cfg

The contents of the configuration file will be expanded in-place as if you would have given its contents on the command line. Please ensure that you don't place spaces between short options and their values, they would become a part of the argument value.


0 all tests passed
2 at least one test was skipped (or at least one flaky test failed)
4 at least one test failed
6 at least one test failed and at least one test skipped
8 no tests in this package, or all non-superficial tests were skipped
12 erroneous package
14 erroneous package and at least one test skipped
16 testbed failure
20 other unexpected failures including bad usage




This manpage is part of autopkgtest, a tool for testing Debian binary packages. autopkgtest is Copyright (C) 2006-2014 Canonical Ltd.

See /usr/share/doc/autopkgtest/CREDITS for the list of contributors and full copying conditions.

2014 autopkgtest