42a0bffec26736bbdb95edf0cd6470f1a62808fe |
|
19-Dec-2017 |
Prathmesh Prabhu <pprabhu@chromium.org> |
Support running test_that via a symlink BUG=b:70390932 TEST=Manually patch test_that.py in lab-tools checkout; test_that works. Change-Id: I5eceebae232d8e791ddd6380c4a419bae1ed94cb
/external/autotest/site_utils/test_that.py
|
f0dda1b3ab3941adeaf7c4e5b567227d4f970bd9 |
|
31-Oct-2017 |
C Shapiro <shapiroc@chromium.org> |
test_that support for model parameter Add pass through support for model going to the test_suite BUG=b:65673486 TEST=test_that -b coral -i coral-release/R63-10000.0.0 :lab: dummy_Pass & test_that -b coral -m robo360 -i coral-release/R63-10000.0.0 :lab: dummy_Pass Change-Id: I00721d8d300a3647ec590d80f132d51695dfe55d Reviewed-on: https://chromium-review.googlesource.com/747622 Commit-Ready: C Shapiro <shapiroc@google.com> Tested-by: C Shapiro <shapiroc@google.com> Reviewed-by: Prathmesh Prabhu <pprabhu@chromium.org>
/external/autotest/site_utils/test_that.py
|
02b46c55690dcbccaf2687ce3f0a6f6440b959db |
|
11-Sep-2017 |
Allen Li <ayatane@chromium.org> |
[autotest] Add old RPC support to test_that BUG=chromium:763207 TEST=trybot TEST=test_that :lab: -b reks -i reks-release/R63-9927.0.0 -p suites dummy_Pass -w cautotest TEST=Run test_that against old moblab (verify bug fix) Change-Id: I18ee13e8b561b0595358aba4f5cb3e8cb148c140 Reviewed-on: https://chromium-review.googlesource.com/661218 Commit-Ready: Allen Li <ayatane@chromium.org> Tested-by: Allen Li <ayatane@chromium.org> Reviewed-by: Prathmesh Prabhu <pprabhu@chromium.org> Reviewed-by: Laurence Goodby <lgoodby@chromium.org>
/external/autotest/site_utils/test_that.py
|
3c0d6cba982b138e228c8f912028cbbc2db7b20e |
|
22-Aug-2017 |
Allen Li <ayatane@chromium.org> |
[autotest] Clean up new suite_args API This is a scrub of the repo, making sure the API is correct at all interfaces and backward compatibility is maintained at the control file interface. In detail, the changes are: AFE create_suite_job RPC made less defensive. An invalid argument will cause the RPC call to error instead of replacing suite_args with an empty dict, which probably would not have produced the desired behavior anyway, so erroring out is better. Inject suite_args into the control file. This allows new Autotest to run old tests. Make the relevant new tests support being run by old Autotest. The referenced bug contains a diagram listing all of the interface points between the new and old API and the compatibility between them. BUG=chromium:758427 TEST=site_utils/test_that.py -b reks -i reks-release/R60-9592.71.0 -w localhost :lab: dummy_Pass TEST=site_utils/run_suite.py -b reks -i reks-release/R60-9592.71.0 -p suites -s dummy Change-Id: I352491b4ad72818ce288144a0380e441a3f65b40 Reviewed-on: https://chromium-review.googlesource.com/631236 Commit-Ready: Allen Li <ayatane@chromium.org> Tested-by: Allen Li <ayatane@chromium.org> Reviewed-by: Allen Li <ayatane@chromium.org>
/external/autotest/site_utils/test_that.py
|
429fd801a01a1f0b9fc8be09cf293b95d314088b |
|
08-Aug-2017 |
Allen Li <ayatane@chromium.org> |
[autotest] Make test_that pass suite_args as a dict BUG=chromium:753072 TEST=Trybot TEST=Run test_that Change-Id: I296a20bd7bc56567254fae7663ddc74178f5caf9 Reviewed-on: https://chromium-review.googlesource.com/604737 Commit-Ready: Allen Li <ayatane@chromium.org> Tested-by: Allen Li <ayatane@chromium.org> Reviewed-by: Prathmesh Prabhu <pprabhu@chromium.org>
/external/autotest/site_utils/test_that.py
|
68ee1c54e49751ba4d8278586e121ceee97ef134 |
|
05-Jul-2017 |
Prathmesh Prabhu <pprabhu@chromium.org> |
test_that: Fix argument validation. A test_droid CL broke argument validation so that test_that started accepting :lab: runs without a --build argument. This would lead to cryptic error later from the run_suite RPC. FixIt, and add unittests to protect this argument validation. BUG=None TEST=(new) unittest fails before the change, pass afterwards. Change-Id: Iffcd8005c62665ea9c0a2f1c45d43b8564c67abf Reviewed-on: https://chromium-review.googlesource.com/559950 Commit-Ready: Prathmesh Prabhu <pprabhu@chromium.org> Tested-by: Prathmesh Prabhu <pprabhu@chromium.org> Reviewed-by: Xixuan Wu <xixuan@chromium.org>
/external/autotest/site_utils/test_that.py
|
910c6042cd2c93e7cfa8e253f96092792600b5be |
|
25-Oct-2016 |
Allen Li <ayatane@chromium.org> |
[autotest] Fix argparse handling of arguments starting with dashes BUG=chromium:658013 TEST=Run test_that against lab Change-Id: I98c8682ac6d46e7a3bc0c2aa85e5024915989aef Reviewed-on: https://chromium-review.googlesource.com/402710 Commit-Ready: Allen Li <ayatane@chromium.org> Tested-by: Allen Li <ayatane@chromium.org> Reviewed-by: Richard Barnette <jrbarnette@google.com>
/external/autotest/site_utils/test_that.py
|
47b68eeb258a34a9c3f7138dfe4f737f81431b1d |
|
06-Jun-2016 |
Kevin Cheng <kevcheng@chromium.org> |
[autotest] test_that error message correction. Error message pointed out typo in flag to use: '--autotest-dir' instead of '--autoest_dir' BUG=None TEST=Check that it prints correct message now Change-Id: I55915c5075654ff4514a29e6c2304c7c2caab6ed Reviewed-on: https://chromium-review.googlesource.com/350030 Commit-Ready: Kevin Cheng <kevcheng@chromium.org> Tested-by: Kevin Cheng <kevcheng@chromium.org> Reviewed-by: Dan Shi <dshi@google.com>
/external/autotest/site_utils/test_that.py
|
57740ca3b8ceb4bf4ff682e7fe8f208d4f7357d9 |
|
19-Feb-2016 |
Simran Basi <sbasi@google.com> |
[autotest] fix test_{droid,that} --pretend. BUG=b:27043786,chromium:586833 TEST=test_that and test_droid runs. Change-Id: I0de93498e3d05db8bd30753d166bf7b7e7e3f486 Reviewed-on: https://chromium-review.googlesource.com/328713 Commit-Ready: Simran Basi <sbasi@chromium.org> Tested-by: Simran Basi <sbasi@chromium.org> Reviewed-by: Dan Shi <dshi@chromium.org> Reviewed-by: Kevin Cheng <kevcheng@chromium.org>
/external/autotest/site_utils/test_that.py
|
60dfcefd5a8d726118d95297d1250909e834435b |
|
13-Feb-2016 |
Laurence Goodby <lgoodby@google.com> |
[autotest] add max_runtime_mins option to test_that. BUG=b:27174274 TEST=ran tests in Talyn lab. Change-Id: I8eef3e53f339f1346e249f104933fb63740caa32 Reviewed-on: https://chromium-review.googlesource.com/326928 Commit-Ready: Laurence Goodby <lgoodby@chromium.org> Tested-by: Laurence Goodby <lgoodby@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org>
/external/autotest/site_utils/test_that.py
|
f067d9c7e8dc08426a16edc89d2d33f17646a50f |
|
22-Dec-2015 |
Simran Basi <sbasi@google.com> |
test_droid: Basic remote support. Allows test_droid to run tests against Android devices connected to a remote host. This is useful if the test station is in the lab and unaccesible. BUG=b:26186193 TEST=test_droid.py ZX1G22JHXN android_DummyTest --debug \ --remote android1758-wifi-test-station-2.cros Change-Id: I7f98440540fcf2e81e8715b81591c1d64ac60f7a Reviewed-on: https://chromium-review.googlesource.com/319811 Commit-Ready: Simran Basi <sbasi@chromium.org> Tested-by: Simran Basi <sbasi@chromium.org> Reviewed-by: Dan Shi <dshi@chromium.org>
/external/autotest/site_utils/test_that.py
|
14622bb8249c2200e5258ea6d40b3362229e90a7 |
|
25-Nov-2015 |
Simran Basi <sbasi@google.com> |
[autotest] test_droid V1. This CL introduces test_droid, a script to launch autotest tests against local Android DUTs connected to the host computer via USB. Much of the logic is similar to test_that, so this has all been pulled into a new script test_utils.py Pending work: * Moving of generate_test_report to somewhere it is accesible outside of the chroot. * Add in support for remote adb hosts (i.e. the DUT is connected to a remote computer). * Sequences support. BUG=b:25748759 TEST=./test_droid.py 7d52318 brillo_WhitelistedGtests CQ-DEPEND=CL:315230 Change-Id: Idab3805f7921173fbc706354b322f37244240fa2 Reviewed-on: https://chromium-review.googlesource.com/314870 Commit-Ready: Simran Basi <sbasi@chromium.org> Tested-by: Simran Basi <sbasi@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
54ec4f3c0ae231c6177174603b448919dd1fa420 |
|
10-Dec-2015 |
Fang Deng <fdeng@chromium.org> |
[autotest] test_that checks autoserv return code. If autoserv dies for unexpected reasons, test_that will exit with code 1. TEST=test_that dummy_Pass HOST_IP test_that dummy_Fail HOST_IP force abstract_ssh to raise an exception so that autoserv will fail at crashinfo collection step. Check exit code is 1. BUG=chromium:555073 Change-Id: I85eae412cb07d13f2f1eda49460faacf11bf393f Reviewed-on: https://chromium-review.googlesource.com/317593 Commit-Ready: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
7c55662fad7b6fa0d837e2a5dfff115af8b232be |
|
07-Dec-2015 |
J. Richard Barnette <jrbarnette@chromium.org> |
[autotest] Validate --board and --build with test_that :lab:. A board and build are both required with `test_that :lab:`. This adds a check to enforce that with a useful error message. BUG=chromium:418379 TEST=run test_that using all combinations of --board and --build Change-Id: I9f39fd751482db836c186d5a2024521d73c8e9a7 Reviewed-on: https://chromium-review.googlesource.com/316513 Commit-Ready: Richard Barnette <jrbarnette@chromium.org> Tested-by: Richard Barnette <jrbarnette@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
61078ebc0efc39d5ff57dd1513460ea0947f21b8 |
|
18-Nov-2015 |
Aviv Keshet <akeshet@chromium.org> |
test_that: remove obsolete help text There is no :vm: mode to test that (and I believe there never was implemented). Runs against a local vm are just specified by using the local host and port for REMOTE. BUG=None TEST=None Change-Id: Ibf0e2f5c102deb76d22f6b8dbf8e434199d32aea Reviewed-on: https://chromium-review.googlesource.com/312988 Commit-Ready: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Dan Shi <dshi@chromium.org>
/external/autotest/site_utils/test_that.py
|
cbb5ba051619fc10b0607abc369c11592202a0a9 |
|
11-Aug-2015 |
Laurence Goodby <lgoodby@google.com> |
Fix for test_that_wrapper call to test_that.parse_arguments. test_that_wrapper expects one return argument from parse_arguments. BUG=chromium:446321 TEST=Ran unittest, tested with moblab. Change-Id: I82bde8832e3f7213405b3f98efd703be0d153587 Reviewed-on: https://chromium-review.googlesource.com/292810 Reviewed-by: Simran Basi <sbasi@chromium.org> Tested-by: Laurence Goodby <lgoodby@chromium.org> Commit-Queue: Laurence Goodby <lgoodby@chromium.org>
/external/autotest/site_utils/test_that.py
|
079312e32942bdf5d72721d0bcfd81bb87f927b0 |
|
05-Aug-2015 |
Laurence Goodby <lgoodby@google.com> |
test_that: adds a --web option. Similar to run_suite's --web option, this option is used for remote :lab: runs to direct the test to a specific server (for example, a moblab). The --web option is not passed to remote test_that invocations. BUG=chromium:446321 TEST='--web' is used for run_suite but is not passed to remote test_that. Change-Id: I80081d751709b2f3d82fa8248139b43778962342 Reviewed-on: https://chromium-review.googlesource.com/290811 Trybot-Ready: Laurence Goodby <lgoodby@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Laurence Goodby <lgoodby@chromium.org> Tested-by: Laurence Goodby <lgoodby@chromium.org>
/external/autotest/site_utils/test_that.py
|
48e51174a21255fef9904d642f5c01d1457a9b0d |
|
24-Jun-2015 |
Daniel Wang <wonderfly@google.com> |
Take private key from a cmdline flag BUG=brillo:1200 TEST=unit tests Change-Id: Ic0f153c1905bacaac199d5ba6bfbd9c04a8cce6e Reviewed-on: https://chromium-review.googlesource.com/286914 Commit-Queue: Daniel Wang <wonderfly@google.com> Trybot-Ready: Daniel Wang <wonderfly@google.com> Tested-by: Daniel Wang <wonderfly@google.com> Reviewed-by: Prathmesh Prabhu <pprabhu@chromium.org>
/external/autotest/site_utils/test_that.py
|
36cfd831af781eed114337efa5b90c103a49b502 |
|
10-Oct-2014 |
Dan Shi <dshi@chromium.org> |
[autotest] Support updating both firmware and CrOS from run_suite Add firmware_build option for run_suite to pass in firmware build. Save the build containing server-side test to job_keyvals. Design doc: https://docs.google.com/a/google.com/document/d/115aAHyZaatFDzuKWm61Sj2fImH4NDWikXqifvWKbbW0/edit# BUG=chromium:270258 TEST=local run_suite: make sure it works with older build: ./run_suite.py -b veyron_jerry -i trybot-veyron_jerry-paladin/R45-7086.0.0-b8 -p suites -s dummy_server Call failed as --firmware_ro_build not supported yet. ./run_suite.py -b veyron_jerry -i trybot-veyron_jerry-paladin/R45-7122.0.0-b11 --firmware_build veyron_jerry-firmware/R41-6588.9.0 --test_source_build trybot-veyron_jerry-paladin/R45-7122.0.0-b11 --firmware_ro_build veyron_jerry-firmware/R41-6588.9.0 -p suites -s dummy_server Call failed as the firmware build does not have test_suites package built: ./run_suite.py -b veyron_jerry -i trybot-veyron_jerry-paladin/R45-7122.0.0-b11 --firmware_build veyron_jerry-firmware/R41-6588.9.0 --test_source_build veyron_jerry-firmware/R41-6588.9.0 -p suites -s dummy_server make sure it works with new build: ./run_suite.py -b veyron_jerry -i trybot-veyron_jerry-paladin/R45-7122.0.0-b11 --firmware_build veyron_jerry-firmware/R41-6588.9.0 --test_source_build trybot-veyron_jerry-paladin/R45-7122.0.0-b11 -p suites -s dummy_server DEPLOY=apache,scheduler Change-Id: I50d23a7e81e4d6224b3483111110f910eb407074 Reviewed-on: https://chromium-review.googlesource.com/272793 Reviewed-by: Dan Shi <dshi@chromium.org> Commit-Queue: Dan Shi <dshi@chromium.org> Trybot-Ready: Dan Shi <dshi@chromium.org> Tested-by: Dan Shi <dshi@chromium.org>
/external/autotest/site_utils/test_that.py
|
a6b2ce9a320ac947873805c413a757e2d377e46f |
|
28-May-2015 |
Dan Shi <dshi@chromium.org> |
[autotest] Fix provision job in test_that --provision arg in autoserv was changed, which breaks test_that's run_provisioning_job. This CL updates the method with correct autoserv command line args: ['--provision', '--job-labels', build] BUG=chromium:492817 TEST=test_that --build veyron_jerry-release/R44-7058.0.0 172.27.215.232 dummy_Pass Change-Id: I6f67a854e897234396d2087a2b6417fbd6cbcbed Reviewed-on: https://chromium-review.googlesource.com/273863 Tested-by: Dan Shi <dshi@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Dan Shi <dshi@chromium.org>
/external/autotest/site_utils/test_that.py
|
8bd6d482f849d9e015f0a74c588e5135e3119dcb |
|
22-Apr-2015 |
Simran Basi <sbasi@google.com> |
[test_that] Remove --force arg when calling autotest_quickmerge autotest_quickmerge's --force arg will soon be deleted so remove the call from test_that. BUG=brillo:908 TEST=test_that still works. CQ-DEPEND=CL:266755 Change-Id: I7851dcbc14b3a6e1b1b9bfd766b7efdb02bf3851 Reviewed-on: https://chromium-review.googlesource.com/266763 Reviewed-by: Dan Shi <dshi@chromium.org> Tested-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Simran Basi <sbasi@chromium.org>
/external/autotest/site_utils/test_that.py
|
2252db34500a41ecbc8db0972ad91adce3030ab6 |
|
08-Apr-2015 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] remove unneeded legacy support from test_that This CL removes code that allows test_that to work with builds that pre-date the relocation of the build-time autotest tree in the sysroot. Given that that change is over a year old, it's unlikely that anyone still needs this support. BUG=chromium:256906 TEST=Unit tests Change-Id: I6f4854c3dc4bff7a242a0b34bfc661376e209023 Reviewed-on: https://chromium-review.googlesource.com/264750 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
1767613dbf75dc28299b9420cbffe0b2589a7783 |
|
08-Apr-2015 |
Mussa <mussa@chromium.org> |
Always perform quickmerge step when running test_that. This scenario has come up multiple times. You are modifying your test on your local machine but when you run test_that the code you added isn't getting run. Test_that is running some other previous version of your code. You could chase down packages on the DUT or on your local box and look for sentinel files or checksum files and delete them to trick test_that to pickup your changes. Or, you can use a flag (--force) that autotest_quickmerge.py' already supports. Pass --force flag to autotest_quickmerge.py by default. BUG=None. TEST=I ran a local test. Will run test_that unit tests when I find out how to. Change-Id: I0f8f6f3c6763defe45cb31ef64e815fbd75bc0ae Reviewed-on: https://chromium-review.googlesource.com/264497 Tested-by: Mussa Kiroga <mussa@chromium.org> Reviewed-by: Katherine Threlkeld <kathrelkeld@chromium.org> Commit-Queue: Mussa Kiroga <mussa@chromium.org>
/external/autotest/site_utils/test_that.py
|
b669cbdea7f4bfc9cbfcea48d9cf9ce62f0deff5 |
|
13-Sep-2013 |
Dan Shi <dshi@chromium.org> |
[autotest] Disable install autotest from job_repo_url for test_that use_packing option for autotest._install method is True by default. When using test_that to run test, DUT will try to install autotest package from job_repo_url with server configured in global_config (cautotest). For DUT not located in lab, there is no job_repo_url existed for such DUT in cautotest, and therefore autotest will try to copy over the local files. However, when a user try to use a DUT in the lab to run test_that, autotest package is downloaded and installed from job_repo_url. That leads to a problem that developers can't use test_that to verify autotest code. The fix is to add a new option no_use_packaging to autoserv, and thread the option to autotest._install. The option is set to False by default, and set to True in test_that. Also, when this argument is set to True, any package installation will use AutoservFetcher. BUG=chromium:265673 TEST=run following command and confirm local autotest code is copied over to the DUT: test_that -b lumpy 172.22.75.225 dummy_Pass Change-Id: I365f5f7dc79c8e65cddaa0af77f1f1a1e4b0012a Reviewed-on: https://chromium-review.googlesource.com/169343 Reviewed-by: Mungyung Ryu <mkryu@google.com> Commit-Queue: Mungyung Ryu <mkryu@google.com> Tested-by: Mungyung Ryu <mkryu@google.com>
/external/autotest/site_utils/test_that.py
|
789f87c87dcacee0584033b21cbf920210b8d336 |
|
24-Mar-2015 |
J. Richard Barnette <jrbarnette@chromium.org> |
Remove references to the try-bot and pfq pools. The try-bot pool is poorly loved, and generally a nuisance. Remove references to the pool. Shift test_that to use the suites pool by default. This also deletes a stray, dead reference to the pfq pool, as that pool has been unused for some time. BUG=chromium:469926 TEST=None Change-Id: Iaa449c5d5efab5ee98c63e44d38f1a51dfed5d9a Reviewed-on: https://chromium-review.googlesource.com/262061 Tested-by: Richard Barnette <jrbarnette@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Richard Barnette <jrbarnette@chromium.org>
/external/autotest/site_utils/test_that.py
|
b723b1b17967f7f12f782b1d2d1b48854194d218 |
|
04-Feb-2015 |
Simran Basi <sbasi@google.com> |
test_that: Be able to specify pool. BUG=chromium:446321 TEST='--pool' is passed to test_that_wrapper Change-Id: I591ca76419aa6cf796dfa986816f542ba5000963 Reviewed-on: https://chromium-review.googlesource.com/246185 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Simran Basi <sbasi@chromium.org>
/external/autotest/site_utils/test_that.py
|
f5f2ba2abff1d5831ac8ead266a7a99c69681984 |
|
03-Feb-2015 |
Christopher Wiley <wiley@chromium.org> |
test_that: Enhance help text Make it clear that you don't get local changes when you run in the lab. BUG=None TEST=None Change-Id: Ic4aa0660fc5d913d668985307b6b8e7e058f09cc Reviewed-on: https://chromium-review.googlesource.com/245590 Tested-by: Christopher Wiley <wiley@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Christopher Wiley <wiley@chromium.org>
/external/autotest/site_utils/test_that.py
|
5783f8a94ce208df0cf46547e8b50b7f74a9638e |
|
22-Dec-2014 |
Dan Shi <dshi@chromium.org> |
[autotest] Provide suggestions when test_that failed to locate any test. Make it easier for user to locate the exact test to run. When test_that failed to locate any test matching the test argument, provide a list of suggestions. For example: test_that -b peppy --fast 172.27.213.193 e:dummy_Pass_123 leads to following exception message: ValueError: Found no tests. Check your suite name, test name, or test matching wildcard. Did you mean any of following tests? dummy_Pass dummy_PassServer dummy_Pass.wifichaos dummy_Pass.actionable dummy_Pass.experimental dummy_Fail dummy_Pass.bluetooth dummy_RepeatArgs dummy_FailServer dummy_FailServer_Flaky BUG=chromium:441455 TEST=local test runs with test_that: test_that -b peppy --fast 172.27.213.193 e:dummy_Pass_123 test_that -b peppy --fast 172.27.213.193 f:dummy test_that -b peppy --fast 172.27.213.193 suite:bvt_123 Change-Id: I22a8b566580e492c68b8729afc22d6950a2166a3 Reviewed-on: https://chromium-review.googlesource.com/237300 Trybot-Ready: Dan Shi <dshi@chromium.org> Tested-by: Dan Shi <dshi@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Dan Shi <dshi@chromium.org>
/external/autotest/site_utils/test_that.py
|
8a89e9b22f6edcd987020a9aefa74eb21571606f |
|
08-Oct-2014 |
Fang Deng <fdeng@chromium.org> |
[autotest] Use realpath for sysroot in test_that Use realpath for sysroot when checking whether it is already inside sysroot. TEST=run test_tat BUG=chromium:416905 Change-Id: Id08ab0cba78ee1493e1b79e1433974327e2b7a0b Reviewed-on: https://chromium-review.googlesource.com/222102 Tested-by: Fang Deng <fdeng@chromium.org> Reviewed-by: Dan Shi <dshi@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
f6e4024f58da69d578dede9b1d3f4456fc2a28df |
|
03-Oct-2014 |
Aviv Keshet <akeshet@chromium.org> |
autotest: improve test_that.py error message when no tests found BUG=chromium:419931 TEST=None Change-Id: Ic3b71c7a23183fc89436a757064dfb7852a9c34f Reviewed-on: https://chromium-review.googlesource.com/221315 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Reviewed-by: Olof Johansson <olofj@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
8e555337e8ee40ac235f4d860f69497a9967e0e5 |
|
27-Jun-2014 |
Josh Triplett <josh@joshtriplett.org> |
test_that: Point to latest results directory When done, print the path to the "latest" symlink, not just to the directory itself. Move the message to after the creation of that symlink. BUG=None TEST=test_that $machine dummy_Pass Change-Id: I086641962ea7e91aedea93f1dbe31753bdc2991f Reviewed-on: https://chromium-review.googlesource.com/205964 Tested-by: Josh Triplett <josh@joshtriplett.org> Reviewed-by: Ilja Friedel <ihf@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Josh Triplett <josh@joshtriplett.org>
/external/autotest/site_utils/test_that.py
|
6303341b1b7c1447344712903be1f3f2517ceb82 |
|
27-Jun-2014 |
Josh Triplett <josh@joshtriplett.org> |
test_that: Use a relative symlink for test_that_latest for chroot-friendliness Linking to the absolute path of the latest test results only works from within the chroot; making the symlink relative makes it easy to access from outside the chroot as chroot/tmp/test_that_latest/ . BUG=None TEST=test_that $machine dummy_Pass Change-Id: I26c4c276028d5b7424126d3f4aa800e071fe0f35 Reviewed-on: https://chromium-review.googlesource.com/205963 Tested-by: Josh Triplett <josh@joshtriplett.org> Reviewed-by: Ilja Friedel <ihf@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Josh Triplett <josh@joshtriplett.org>
/external/autotest/site_utils/test_that.py
|
89dc43dd18fe29c73c901f670dc8eeceda144a69 |
|
29-May-2014 |
Jorge Lucangeli Obes <jorgelo@chromium.org> |
Fix typo in 'test_that' error message. BUG=None TEST=None Change-Id: I3b40e2a5b7382d6ec503d0fb3ad8a57ef1fb01ec Reviewed-on: https://chromium-review.googlesource.com/202160 Commit-Queue: Jorge Lucangeli Obes <jorgelo@chromium.org> Tested-by: Jorge Lucangeli Obes <jorgelo@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
04be2bd5e4666a5c253e9c30ab20555e04286032 |
|
08-May-2014 |
Ilja H. Friedel <ihf@chromium.org> |
Autotest: Change logging.warn() to logging.warning(). logging.warn() is deprecated. See http://bugs.python.org/issue13235 Substitution was performed via ~/cros/src/third_party/autotest/files$ find ./ -type f | xargs sed -i 's/logging.warn(/logging.warning(/' BUG=None. TEST=There should be one-- and preferably only one --obvious way to do it. Change-Id: Ie5665743121a49f7fbd5d1f47896a7c65e87e489 Reviewed-on: https://chromium-review.googlesource.com/198793 Commit-Queue: Ilja Friedel <ihf@chromium.org> Tested-by: Ilja Friedel <ihf@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org>
/external/autotest/site_utils/test_that.py
|
f81917c3692293e7b72a4fa467813cd1a43f3126 |
|
18-Apr-2014 |
Jorge Lucangeli Obes <jorgelo@chromium.org> |
test_that: fix typo in usage string. BUG=None TEST=test_that --help shows no typo. Change-Id: I78c1b3619095842c38a4abeb39c4acb95850dbed Reviewed-on: https://chromium-review.googlesource.com/195500 Commit-Queue: Jorge Lucangeli Obes <jorgelo@chromium.org> Tested-by: Jorge Lucangeli Obes <jorgelo@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
20b20c73d91bf3411da5b5fa69b8cee592e8088f |
|
02-Apr-2014 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that reuse a created suite when scheduling iterations This CL pulls turns test_that's schedule_local_suite into fetch_local_suite and moves the scheduling call into perform_local_run. When scheduling multiple iterations of a test or suite, this allows the created suite object to be re-used, which saves a lot of duplicate searches through all the control files in the tree when scheduling duplicate iterations. NOTE: This does not really fix the linked-to bug (test_that fails when more than 999 tests are schedule). However, it does make it less painful, and I don't feel like filing a new bug, so I'm linking to that one. TEST=Unit tests pass `git grep schedule_local_suite` -> no results `test_that --board=lumpy --iterations=1000 localhost dummy_Pass` -> 1000 jobs are scheduled in under a minute, rather than the snails pace before. BUG=chromium:356879 Change-Id: I50e3c70a95c829a2d97333bee4472f25c298da66 Reviewed-on: https://chromium-review.googlesource.com/193095 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Reviewed-by: Christopher Wiley <wiley@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
5c50833d70f220440303fc1c1eea36fb7798dfc6 |
|
19-Mar-2014 |
Fang Deng <fdeng@chromium.org> |
[autotest] dynamic suite emits sub-job's dir as subdir to status.log It is enforced by client/common_lib/base_job.py that, within a single job, mutiple tests with the same name can not run in the same sub-directory. That is to say, (test_name, subdir) should be unique within the scope of a job. Tko parser hold this assumption and it get confused if this is not the case. This cl makes dynamic suite emit a subdir to status.log for each test status. The subdir is composed of job_tag and subdir retrieved from tko_test_view_2. This makes sure of two things: a) if a job has been retried, we will not end up with having two statuses that look exactly the same. For example, in the case a retry has been run, without changes in this cl the status log looks like START ---- dummy_Fail.Fail timestamp=.. ... START ---- dummy_Fail.Fail timestamp=.. ... With changes in this cl, it will look like START 116-fdeng/172.2.3.4/dummy_Fail.Fail dummy_Fail.Fail timestamp=.. ... START 117-fdeng/172.5.6.7/dummy_Fail.Fail dummy_Fail.Fail timestamp=.. ... b) This also makes sure that if someone runs a test twice with different subdir_tag, (test_name, subdir) is still unique within the scope of the suite job. For example, one could do in control file job.run_test('dummy_Fail', tag='Fail', subdir_tag='subdir_1') job.run_test('dummy_Fail', tag='Fail', subdir_tag='subdir_2') With changes in this cl, the status log looks like START 116-fdeng/172.2.3.4/dummy_Fail.Fail.subdir_1 dummy_Fail.Fail timestamp=.. ... START 116-fdeng/172.2.3.4/dummy_Fail.Fail.subdir_2 dummy_Fail.Fail timestamp=.. ... TEST=unittest; dummy suite runs ok with both run_suite and test_that; Works fine with test retry changes: - Cherry-pick test retry changes crosreview.com/188365 and crosreview.com/190585. - Added JOB_RETRIES = 2 to dummy_Fail/control. Created a trybot image. - Ran dummy suite and observed that dummy_Fail was retried twice without errors. BUG=chromium:354654 DEPLOY=apache Change-Id: Ib939a8d95b3a4db06d408257b1be8608cce5fed2 Reviewed-on: https://chromium-review.googlesource.com/191002 Reviewed-by: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
2ce92d5a16b7842d2312258289747ebb3f7937ac |
|
14-Mar-2014 |
Christopher Wiley <wiley@chromium.org> |
Add a --iterations flag to test_that This is useful when we want to run a test N times to gauge how flaky a test is. TEST=Run test_that with --iterations=2. BUG=chromium:352719 Change-Id: Id787d9599f5ef222bafbea0ff6945ea9d0cef8f9 Reviewed-on: https://chromium-review.googlesource.com/190062 Reviewed-by: Christopher Wiley <wiley@chromium.org> Commit-Queue: Christopher Wiley <wiley@chromium.org> Tested-by: Christopher Wiley <wiley@chromium.org>
/external/autotest/site_utils/test_that.py
|
ad7309b61b322548ed3a5edd18fec8e42f47ff5b |
|
12-Feb-2014 |
Jorge Lucangeli Obes <jorgelo@chromium.org> |
test_that: fix spacing in help message. BUG=None TEST=test_that -h Change-Id: Iaa629765ccf5ca66b8dbc1b0363b6018d4f3eea8 Reviewed-on: https://chromium-review.googlesource.com/186112 Commit-Queue: Jorge Lucangeli Obes <jorgelo@chromium.org> Tested-by: Jorge Lucangeli Obes <jorgelo@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
f860bd1617b5c200a0378cc379d83585c5ecff63 |
|
24-Jan-2014 |
beeps <beeps@chromium.org> |
[Autotest] Download sonic extension in the server test. Though the canonical way to handle extensions is to install them as a dep, our builders seem to have trouble contacting the omaha server to get the extension. Since our drones don't have the same problem, download the extension in the server test instead. TEST=Ran the test. BUG=chromium:321179, chromium:321177: Change-Id: I4bcea636eb3a469e415af28630005a86196b7800 Reviewed-on: https://chromium-review.googlesource.com/183743 Tested-by: Prashanth B <beeps@chromium.org> Reviewed-by: Dan Shi <dshi@chromium.org> Commit-Queue: Prashanth B <beeps@chromium.org>
/external/autotest/site_utils/test_that.py
|
2113774da25f7e6b74e4e8c801e72d581d770ed8 |
|
10-Jan-2014 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add an image name format example to test_that BUG=None TEST=None Change-Id: I7c13ea174f84a207775f9713518824193163da11 Reviewed-on: https://chromium-review.googlesource.com/182213 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
3f6c99a531bdc4481499157f554c666af52377ba |
|
31-Oct-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that use legacy_path less aggressively The current logic is incorrectly using legacy_path in certain cases where the new autotest path is already available. This fixes that. Instead of running in legacy path mode if the existence of /build/[board]/usr/local/autotest/site_utils is detected, we will now run in legacy mode if: 1) /build/[board]/usr/local/autotest exists and 2) /build/[board]/usr/local/build/autotest does not exist. BUG=None TEST=unit tests pass, local test_that run passes. Change-Id: I142c41b57f802675733b993a8f734930e3b2ef3c Reviewed-on: https://chromium-review.googlesource.com/175320 Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
f0229428c6412d5bdebb46c962cc52d21bd8c3ae |
|
24-Oct-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] fail test_that if autotest_quickmerge fails BUG=chromium:310849 TEST=Verified that test_that dies if autoserv fails. Change-Id: I24e628e5d6c3ce374627183b0dd2e35508f5fe83 Reviewed-on: https://chromium-review.googlesource.com/174451 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
1a6ce6e485e44536f705dfc9a0bd66c3b0b294eb |
|
22-Oct-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] Check that test_that.py exists before bootstrapping to it BUG=chromium:310173 TEST=Unit tests pass. Intentionally ran test_that with a non-existant autotest-dir, saw informative exception raised. Change-Id: Ie2a18e032fd3e42dfc7098b55412a4de50f762ae Reviewed-on: https://chromium-review.googlesource.com/173763 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
6c0283984ec221edae7f446109bd7764f72e89dd |
|
22-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] updates to autotest repo to reflect usr/local/build/autotest CQ-DEPEND=CL:I836d933e8a828177304dd10a74841138a9502aff BUG=chromium:256906 TEST=See CQ-DEPEND'ed CL. Also, tested that test_that works against both pre- and post- CL:I836d933e8a828177304dd10a74841138a9502aff builds. Change-Id: I501c61c0a9c0198f3e15256c7e0cf03452b0ca93 Reviewed-on: https://chromium-review.googlesource.com/62879 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
a6d597a916d2dc8ae8e25e45d5839bb093190ece |
|
10-Oct-2013 |
Fang Deng <fdeng@chromium.org> |
[autotest] Add --enforce-deps option to test_that Previously, if run a suite using test_that, dependencies are enforced by default. However, some labels are not auto-detectable, such as 'wificell' and thus can never be satisfied. This cl puts the dependency enforcement logic behind --enforce-deps, so that users can choose to not enforce dependencies. TEST=locally run dummy suite with test_that with and without the option; run unit test. BUG=chromium:265715 Change-Id: Ibbc31f0b8b7959cb98e78583daee55391583e002 Reviewed-on: https://chromium-review.googlesource.com/172670 Reviewed-by: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
9525124a90ad9dd8b6cbdeb269ad2aa78bf73709 |
|
10-Oct-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] show quickmerge messages at INFO level The recent breakage of autotest_quickmerge went unnoticed for a full day, in part becase the quickmerge output is suppressed by default when running test_that. I think devs would like to see the quickmerge output by default, so this CL upgrades those messages to the INFO level. BUG=None TEST=None Change-Id: I4882fb7ee9af023c7167547617d80cc087eddc0e Reviewed-on: https://chromium-review.googlesource.com/172610 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
a222727818d6194b59f0b83fd35e2580d9dcd974 |
|
03-Oct-2013 |
Fang Deng <fdeng@chromium.org> |
[autotest] Experimental tests do not influence the exit code of test_that Use --ignore_experimental_tests option when calling generate_test_report script. CQ-DEPEND=CL:Idd7b0fafc9ffe081504b4cf8c1e3e771878ee06e TEST=Confirmed that when only experimental tests failed, test_that exits with code 0. BUG=chromium:265989 Change-Id: I30e6a18be6cdda04b2e38c14ddfcb394c5bf6d56 Reviewed-on: https://chromium-review.googlesource.com/171731 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
2db9676a1fa09fe9eb4d4350712353889a2f0588 |
|
04-Oct-2013 |
Fang Deng <fdeng@chromium.org> |
[autotest] Dynamic stuie stores `experimental` keyval in job object server/cros/dynamcis_suite/suite.py:create_job now stores `experimental` keyval in the job object. test_that retrieves this keyval via job.keyvals['experimental'] and drops it in the keyval file under the job result directory. This keyval will be further used by generate_test_report script. TEST=Run test_that, confirmed that 'experimental' keval is correctly written to the keyval file; unit tests passed. BUG=chromium:265989 Change-Id: Ib2f99c7d1c1125c431970c8ef0b9b5241cbf0bc7 Reviewed-on: https://chromium-review.googlesource.com/170944 Reviewed-by: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
b1da8306a7932cdbb5c4c10e25cac6e83a965a1c |
|
24-Sep-2013 |
Fang Deng <fdeng@chromium.org> |
[autotest] test_that respects a test's DEPENDENCIES. test_that now auto-detects a host's labels and assign appropriate labels to the DUT in afe. When the suite is scheduled locally, only test with satisfiable DEPENDIENCIES will be scheduled. There is no need to check with each job's dependencies. Because only one DUT is added to the AFE. If no host(in our case, only one host) can statisfy a test's dependencies, Suite.schedule() will not create a job for that test. BUG=chromium:265715 TEST=Run dummy suite, confirm dummy_BT is not run because it requests bluetooth; Once manually add 'bluetooth' label to the dut, dummy_BT is scheduled and run. Run dummy_BT, confirm it is not run. Change-Id: Icc653f79b69ea87247082662306b6af64b1b970b Reviewed-on: https://chromium-review.googlesource.com/171444 Reviewed-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
6f74e1962ba83aa9c42b98992183a7327d70772b |
|
23-Sep-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that :lab: runs no longer require sysroot This CL reorganizes test_that's main(...) a bit so that lab runs no longer go through the sysroot checking / bootstrapping logic. BUG=chromium:297037 TEST=unit tests. local runs of test_that still work. Lab runs of test that work even when there is no sysroot present for that board. Change-Id: I12e2c97aabac326934259abfb08b9c223f7d4246 Reviewed-on: https://chromium-review.googlesource.com/170303 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
89f1e06ce5198d847f1a1aae22a538fde85c7b35 |
|
18-Sep-2013 |
beeps <beeps@chromium.org> |
[autotest] Test_that raises exception for syntax errors in the control file. Currently, if we run into a syntax error while parsing a control file through test_that, we fail the test run with a cryptic ValueError. Since test_that is mostly used for development purposes, control files often contain syntax errors. This cl raises a ControlVariableException if we run into a syntax error while parsing a control file. This cl also adds a unittest that actually parses all contol files looking for syntax errors. This is beneficial because unittests run during pre-cq try-bots and can catch any gramattical transgressions before the cq run. TEST=Raised exception when control file had syntax error. Did the same when an unrelated control file had a syntax error. Ran sucessful dummy_suite to sanity check. BUG=chromium:282108 Change-Id: I1016a7a9fc318293ac7873e25468fddfefd2908a Reviewed-on: https://chromium-review.googlesource.com/169883 Tested-by: Prashanth B <beeps@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Prashanth B <beeps@chromium.org>
/external/autotest/site_utils/test_that.py
|
00f7a92eeda5a3530e850bc194e19f29bdc0584e |
|
20-Sep-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] fix merge conflict in test_that.py Minor fix -- a referenced variable had its name changed in another CL. BUG=None TEST=Ran a test_that run targetting :lab:, and did not see an error due to undefined variable sysroot_autotest_path Change-Id: I3e155da343b932506e854e2f5be849d00b511287 Reviewed-on: https://chromium-review.googlesource.com/170190 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Prashanth B <beeps@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
7cd12315f095d45c697267df9a37427ed87f9ea3 |
|
25-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add ability for test_that to run jobs in lab This CL adds the ability for test_that to run jobs in the lab. It accomplishes this thusly: 1) A new optional |suite_args| argument to afe.create_suite_job, whose value gets injected into the suite control file in much the same way as job args are injected into job control files. 2) Plumbing to thread #1 through from run_suite.py. 3) A new suite control file, control.test_that_wrapper, that is responsible for kicking of test_that-initiated suites. It expects to see a suite_args string giving the command line that test_that was run with, and re-uses test_that's argument parser to figure out which tests (and therefore which suite predicates) are needed. 4) A small refactor to dynamic_suite, so that it uses suite's create_from_predicates rather than create_from_name. 5) A new optional |predicate| argument to dynamic_suite's reimage_and_run, which if specified causes dynamic suite to use this job-matching predicate rather than its default suite-name predicate. 6) Code path in test_that to pass :lab: runs on to run_suite.py. So, when a developer runs, at their desk: $ test_that -b board -i image :lab: dummy_Pass Then test_that will, kick off a test_that_wrapper suite in the lab, using run_suite, passing along test_that's args into the suite control file so that it can figure out which tests to run: $ run_suite -b board -i image -s test_that_wrapper --pool try-bot \ --suite_args="-b board -i image :lab: dummy_Pass" This CL also fixes a bug in dynamic_suite/tools.py in which strings that contained quotes were being incorrectly injected into control files in inject_var. CQ-DEPEND=CL:I4b1483b977d9032c7adf8117cf32e83c65eed007 BUG=chromium:262984 TEST=Unit tests pass. Ran some test_that jobs against my local autotest instance, using a trybot image that was build with this CL (so that control.test_that_wrapper would be available). $ test_that -b lumpy -i ${image} :lab: dummy_Pass $ test_that -b lumpy -i ${image} :lab: e:login_.* $ test_that -b lumpy -i ${image} :lab: suite:smoke dummy_Fail DEPLOY=apache Change-Id: Ifa73d7de7aac9c6efebd5f559708623804ad3691 Reviewed-on: https://chromium-review.googlesource.com/63378 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
9a84123a76582c977047031c1d69c2772d94985f |
|
18-Sep-2013 |
Fang Deng <fdeng@chromium.org> |
[autotest] test_that handles -i/--build option. Provision the DUT to the requested build specified by -i/--build. If provisioning fails, it will not run the tests. TEST=1) Run with -i option, confirm that the DUT is successfully provisioned. 2) Run without -i/--board option, confirm that the script still works. 3) Pass in an invalid value for -i, confirm that the script fails without continuing to run the tests. 4) Confirm that sigint is still handled properly. 5) test_that_unittest passes BUG=chromium:265713 Change-Id: I0a002a55127b3af4c6949934ce67f54c454e0a94 Reviewed-on: https://chromium-review.googlesource.com/169833 Reviewed-by: Fang Deng <fdeng@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
4efdf03d1057f9d38e043b7c2affe842109805d2 |
|
17-Sep-2013 |
beeps <beeps@chromium.org> |
[autotest] Modify test_that to figure out the board of the remote host. Test_that currently relies on the user for the --board argument. This cl teaches test_that to ssh into a DUT and find the board, if one isn't specified. TEST=Invoked test_that dummy_Pass with and without a board, with and without a resutls dir, on DUTs with different boards. BUG=chromium:265705 CQ-DEPEND=CL:I50811eb8b0e6b145a190211391c2409913bc8a33 Change-Id: I0b12bb456c8f032266c7530a39f87cefc5d9182b Reviewed-on: https://chromium-review.googlesource.com/169767 Tested-by: Prashanth B <beeps@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Prashanth B <beeps@chromium.org>
/external/autotest/site_utils/test_that.py
|
b3ee20bce0aa6a649f93c438a50ae364f6f781d9 |
|
17-Sep-2013 |
Fang Deng <fdeng@chromium.org> |
[autotest] Add --autotest_dir to test_that --autotest_dir allows one to specify a path to a desired autotest folder instead of using the autotest directory in the board specific sysroot. BUG=chromium:272569 TEST=1) Make a copy of autotest directory at /tmp/my_autotest, 2) Insert some logging messages to autoserv and cros_host, 3) Run ./site_utils/test_that.py --autotest_dir=/tmp/my_autotest DUT_IP_ADDR dummy_Pass from the source tree. 4) Confirm that the inserted messages are in the log. confirm that autotest quickmerge is not run. 5) Run it again from the source tree without --autotest_dir option, confirm that autotest quickmerge is run and the sysroot version of autotest is used. 6) Run directly from sysroot without --autotest_dir, confirm that it works and autotest quickmerge is not run. Change-Id: I4da8ed0cb04a696eb4127fbe6d12d34551ded81d Reviewed-on: https://chromium-review.googlesource.com/169764 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Fang Deng <fdeng@chromium.org> Tested-by: Fang Deng <fdeng@chromium.org>
/external/autotest/site_utils/test_that.py
|
0842c9825ccca49502c791d177eb8e4c76bb4210 |
|
16-Sep-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] improve test_that failure message when no board specified BUG=None TEST=None Change-Id: Ib3170d75e701029361ebf4f4a8176c6f35dd1910 Reviewed-on: https://chromium-review.googlesource.com/169543 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
49db6aa3c902bfc87e3a7d3c442aa85e5f5067c0 |
|
13-Sep-2013 |
Dan Shi <dshi@chromium.org> |
[autotest] Delete results_directory if it already exists. when --results_dir is specified and the directory already exists, test_that does not start the test. This CL changes the behavior to delete existing result directory. BUG=chromium:280747 TEST=run following command twice, confirm the file contents are created in the second run: test_that --results_dir /tmp/test1 -b lumpy 172.22.75.225 dummy_Pass Change-Id: I7bb612390af085f87d28e0402afbc0c94ae8394b Reviewed-on: https://chromium-review.googlesource.com/169327 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Dan Shi <dshi@chromium.org> Tested-by: Dan Shi <dshi@chromium.org>
/external/autotest/site_utils/test_that.py
|
e10e8d04e08451ae0c9e5a10c0e8d768636c2105 |
|
10-Sep-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] refactor test_that's main for better readability This CL refactors test_that's main(...), separating out test_that's bootstrap-into-sysroot behavior from its run-some-tests-from-sysroot behavior. BUG=None TEST=unit tests pass; Manually verified that test_that still works (`test_that -b lumpy my.dut suite:smoke`) Change-Id: I17c238d2817216900237ce72aeab7a5361abd742 Reviewed-on: https://chromium-review.googlesource.com/168666 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
bebd62154f7e782793bab4092762b31af286272d |
|
05-Sep-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] log the test_that command line to debug output BUG=None TEST=Manually verified Change-Id: I4f8b08af7f19d39cd03c6715ed9ebc652ef4ce59 Reviewed-on: https://chromium-review.googlesource.com/168127 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
6a704f7a70401f88599390f18c188b6c8010ef9a |
|
04-Sep-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] thread --ssh_options parameter through test_that BUG=chromium:285378 TEST=Unit tests pass. Manually verified that parameter threads to autoserv. Change-Id: Ib19fe8c378ae71f5c5432f82f919fd761a10ca80 Reviewed-on: https://chromium-review.googlesource.com/168072 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
a6adc7a1ec6c1ebb5db420da12ab9f6e5314db05 |
|
30-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add pattern matching ability to test_that's tests arg This CL adds two type of test pattern matching to test_that. Patterns can now be specified with e:[test name pattern] or f:[file name pattern] as the TEST argument to test_that. It also demotes the complaint "Getting control files for a specific suite has not been implemented for FileSystemGetter.' to DEBUG level, so it doesn't cause test_that user confusion. BUG=chromium:214066 TEST=Unit tests pass. Also... Suite names still work the same way: $ test_that [remote] suite:smoke # smoke suite scheduled correctly Test name patterns work, like this: $ test_that [remote] e:cantelope # No tests scheduled, test_that errors out $ test_that [remote] e:dummy # No tests scheduled $ test_that [remote] e:dummy.* # 6 tests whose name begin with dummy # scheduled File name pattern work, like this: $ test_that [remote] f:dummy # # No tests scheduled. $ test_that [remote] f:.*dummy.* # 8 tests scheduled $ test_that [remote] f:.*server.*dummy.* # 2 tests scheduled $ test_that [remote] f:.*client.*dummy.* # 6 tests scheduled Change-Id: I104847bdf7956e020f2581128946471d1794b8d2 Reviewed-on: https://chromium-review.googlesource.com/167668 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Prashanth Balasubramanian <beeps@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
2ebbd3588686d29f984bde75b6725f88db52eb9b |
|
28-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] have test_that's --debug flag also turn on autoserv verbosity BUG=None TEST=Manually tested. Unit tests pass. Change-Id: I7e91c89cbee138a57015b4b8578778afda090ca0 Reviewed-on: https://chromium-review.googlesource.com/167174 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
6a7c7f16c643ad4a77550bf5eb0cd9f96a94e680 |
|
30-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that create a log of the test pass/fail report Creating a log file of the output of generate_test_report is particularly helpful in diagnosing VMTest failures. BUG=None TEST=Manually verified that test_report.log file is created in results directory. Change-Id: I2878e22d0f379e88a26f5c44354ab6ad691dba5a Reviewed-on: https://chromium-review.googlesource.com/167700 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Chris Sosa <sosa@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
f2b6910112d0bb6fe28e2b93099e8b0e9540deeb |
|
28-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] fix test_that breakage due to provisioning changes est_that was broken by a recent change to create_job which enforces that there are hosts actually able to run the desired job. This simple fix just ensures that test_that's mock afe contains a host that satisfies the newly enforced constraints. BUG=chromium:280567 TEST=Unit tests pass; manually verified that test_that can run tests. Change-Id: I097407c29ff8a1663da2755fab61e560a5f61c7e Reviewed-on: https://chromium-review.googlesource.com/167319 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Alexander Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
cbee2dc77cea843b7da34b348bf2a35f375d8a09 |
|
23-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] remove some accidentally committed debug prints Oops. Nobody wanted to see that. BUG=None TEST=None Change-Id: I86ec5151dc94c5a1a63ebb13b4498a04d9cf1b2d Reviewed-on: https://gerrit.chromium.org/gerrit/66839 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Christopher Wiley <wiley@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
1de5bc60d475098e13920dc4bd9d8862f9c59760 |
|
19-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add testing key to the ssh-agent during a test_that local run CQ-DEPEND=CL:If48fb7e9e0aa1e9744a172db42f1a03d9b22aa56 BUG=chromium:242987, chromium:274047 TEST=test_that can now run tests against a DUT from a chroot that does not already have a ssh-agent set up. Unit tests pass. Change-Id: I83fe442cdd5a3ef3594cefa2a818d56c576d5969 Reviewed-on: https://gerrit.chromium.org/gerrit/66232 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
ad7e34ebd5620d7b7eb17e187517ec22611689da |
|
20-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] include test name in test_that results directory Before this CL, if you ran a test suite with test_that, your result directories would just be named results-0, results-1, etc. After this CL, the results directories will include the name of the test, ie results-01-logging_CrashSender. TEST=Ran locally to verify behavior. Also, unit tests pass. BUG=None Change-Id: Id270aacf231aa0a9585eeecc952598f540305167 Reviewed-on: https://gerrit.chromium.org/gerrit/66290 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
e43bccf088d41592440e088c2871e621edc56919 |
|
14-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] improve test_that's logging This CL accomplishes a few things: 1) Makes the output format of logging messages in test_that more in line with the rest of autotest. 2) Makes test_that drop log files (test_that.DEBUG, .ERROR, .INFO, and .WARNING) into the debug/ subdirectory of its logging directory. 3) Adds a --debug flag which allows the user to optionally see DEBUG level messages in stdout. 4) In order to make the logs more readable, test_that is now calling autoserv with its --no-console-prefix flag. BUG=chromium:273256 TEST=Unit tests pass Change-Id: I8b2c4594f22de40ddd25a2ba84214f65021f5c29 Reviewed-on: https://gerrit.chromium.org/gerrit/65893 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
4dd3480a0f4fdde97d697993446cbe3d376129ca |
|
14-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] do not run test_that as root CQ-DEPEND=CL:Iedf70c3bc891f639e396c2a12dfb275923d8f8e9 BUG=chromium:272575 TEST=See CQ-DEPENDED CL; Unit tests pass. Change-Id: If5e4293bfc88255ee3eff278763634006d22a8d6 Reviewed-on: https://gerrit.chromium.org/gerrit/65790 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
c14951af46e88db7574b9e01ff77d2e419e8058e |
|
13-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add ssh_verbosity to test_that and autoserv_utils BUG=chromium:271621 TEST=unit tests pass. Manually verified that flag works as expected. Change-Id: I31c59a80a7a2151c27d4a41ff8c6cfbf51d259cb Reviewed-on: https://gerrit.chromium.org/gerrit/65720 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
fd77591378849955562d4c95d9fc73f5cd1d26b0 |
|
06-Aug-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add a --whitelist-chrome-crashes flag to test_that This flag is required in order for test_that to replace run_remote_tests on builders' VMTest stage. BUG=chromium:268678 TEST=Unit tests pass; ran test_that with new flag, ensured that it was passed to and understood by generate_test_report tool. Change-Id: I7b65561d96cd26ca496699161b8fd9e81a087a34 Reviewed-on: https://gerrit.chromium.org/gerrit/64801 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org>
/external/autotest/site_utils/test_that.py
|
15782defe390f87bf17c41a8bace97d6f56caf83 |
|
31-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add --results_dir flag to test_that This CL adds a new flag to test_that which allows you to specify a results directory (other than the default behavior of creating a new directory in /tmp). It works analagously to the --results_dir_root argument of run_remote_tests BUG=chromium:265919 TEST=Ran test_that both with and without new flag. Without new flag, results were stored in /tmp subdir as before. With new flag, new results directory was created if necessary and results were stored therein. Change-Id: I86657e9cece38eeecd8b08d3a7c616c449e0ecfe Reviewed-on: https://gerrit.chromium.org/gerrit/63936 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Prathmesh Prabhu <pprabhu@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
63f00aa79396ffb0d320907000ff14ef75b18e90 |
|
26-Jul-2013 |
Prathmesh Prabhu <pprabhu@chromium.org> |
test_that: Use default board saved by setup_board if none is provided. BUG=chromium:264879 TEST=Ran test_that without the --board flag to verify default board is detected. Change-Id: I2703e7ddf64fb8a9fc80cf90b933b3ec03ecd685 Reviewed-on: https://gerrit.chromium.org/gerrit/63493 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Prathmesh Prabhu <pprabhu@chromium.org> Commit-Queue: Prathmesh Prabhu <pprabhu@chromium.org>
/external/autotest/site_utils/test_that.py
|
e9170d908d522788e664bc69a12d1c56d8583cec |
|
19-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add a --no-experimental flag to test_that This CL adds a --no-experimental flag to test_that, which will cause it to skip tests marked as experimental when scheduling a suite. A minor change to suite.py was required in order for test_that's count of scheduled tests to be correct when skipping experimental tests. BUG=chromium:262135 TEST=Manually verified that experimental tests do not get scheduled when flag is used. Unit tests pass. Change-Id: I33247446e95915f372ced59f46b78c36fa82c1a7 Reviewed-on: https://gerrit.chromium.org/gerrit/62657 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
8ea71df7ccf65d49df1cc0f83a1ea57d1d3d8c21 |
|
19-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add a --no-quickmerge flag to test_that This CL adds a --no-quickmerge flag to test_that, which will force it to skip the quickmerge step. This option will be used for test_that runs on the builder (as part of VMtest), since on the builder we will have a fresh sysroot and there is no need to run quickmerge. BUG=chromium:262135 TEST=Unit tests pass. Manually verified that flag works. Change-Id: Ia1e12257f5c1288e352f0d3e0078ed764688b0fd Reviewed-on: https://gerrit.chromium.org/gerrit/62648 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
c5e4609c81f5b009211aea34dcc288ea68e9190e |
|
19-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add a --pretend flag to test_that This CL adds a --pretend flag to test_that which causes it to print out the autoserv commands that it would have run, rather than actually running them. BUG=None TEST=Manually tested that --pretend flag works as expected. Unit tests pass. Change-Id: I492c67ee31319ca4c7c157f182b4108a213f047c Reviewed-on: https://gerrit.chromium.org/gerrit/62646 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
c9e7462ef6fae0dba65e37c2d14d7920c08b7e5b |
|
18-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] test_that makes a symlink to latest results directory With this CL, after each local run of test_that, just before generating the test report, test_that will create a symlink from /tmp/test_that_latest to the latest results directory. BUG=chromium:258204 TEST=Ran test that a few times, verified that symlink was updated to point to latest results directory. Change-Id: I4718c919082eff177ba9671140104109c2dc63b7 Reviewed-on: https://gerrit.chromium.org/gerrit/62492 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
30322f95d61f6a2cffada92472729d3f2a926f6b |
|
18-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add --args support to test_that TEST=new unit test of behavior BUG=None Change-Id: I41674d4b6fcf52ab1a888a43f2c2c3047e1bfc60 Reviewed-on: https://gerrit.chromium.org/gerrit/62523 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
0819a61ebdab865250e42af55398f9bbdf413595 |
|
19-Jul-2013 |
Prathmesh Prabhu <pprabhu@chromium.org> |
autotest: Print the result directory at the end of report from test_that The result directory path will be printed after the test report. This saves the user from having to dig out the directory name from the output. BUG=chromium:261835 TEST=(1) Run python test_that_unittest.py (2) Run some autotest test and make sure that the result directory is printed at the end and is correct. Change-Id: If8d7f7b64b35c9e087187782e8dd28c3f9df83fb Reviewed-on: https://gerrit.chromium.org/gerrit/62559 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Prathmesh Prabhu <pprabhu@chromium.org> Tested-by: Prathmesh Prabhu <pprabhu@chromium.org>
/external/autotest/site_utils/test_that.py
|
5e33c1775e6af56e3b193f91b537c46843de1148 |
|
16-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that 0-pad the job id in results directory BUG=chromium:258674 TEST=Ran smoke suite, saw that jobs were 0-padded to be of length 2. Change-Id: Ia234a7ba9a7926e4c7c07ebef81fc96ad63ab007 Reviewed-on: https://gerrit.chromium.org/gerrit/62087 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Simran Basi <sbasi@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
f6b5aaeef62ea9eb77f7c351db1ebf079c87dc3f |
|
09-Jul-2013 |
Christopher Wiley <wiley@chromium.org> |
autotest: Add --fast flag to test_that test_that will turn off some slow autoserv functionality that developers are unlikely to miss if this flag is passed. TEST=Ran test_that with this flag, noted that tests ran faster and appropriate logging was emited/omited. BUG=chromium:258509 Change-Id: I470c1e2392ed20d775f97459d0d52bf29ea92396 Reviewed-on: https://gerrit.chromium.org/gerrit/61304 Tested-by: Christopher Wiley <wiley@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Christopher Wiley <wiley@chromium.org>
/external/autotest/site_utils/test_that.py
|
ba4992a7d4bbd3d4e22803606d2e51c49aefe303 |
|
02-Jul-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] test_that summarizes results via generate_test_report script This CL gets test_that to use the generate_test_report script (supplied in the chroot by crostest_utils) to summarize the results of a run. This is a temporary measure to give some nice result summarizing while sidestepping the bigger problem of re-using the TKO / run_suite result summarizing code. BUG=chromium:243431 TEST=Ran tests in the chroot using test_that. A results summary was displayed at the end. Change-Id: Ia5a4f00e9c27a696887eea358df6d27886cf7526 Reviewed-on: https://gerrit.chromium.org/gerrit/60820 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
c8824406c766ed55e2bb49b5a078595778b9b138 |
|
30-Jun-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] test_that pull test results to subdirectory of /tmp This CL causes test_that to pull logs to a unique-per-run subdirectory of /tmp. In a future CL, this will allow test_that to parse the result logs and summarize the success/failure of tests and suites. BUG=chromium:243431 TEST=unit tests pass; Ran a suite with test_that, and saw that /tmp/test_that_results_[unique string]/ was created, contained test result logs, and had appropriate permissions to allow non-root user to examine directory and results. Change-Id: Id44d85946657b9d570cbf728b19018643512c9dd Reviewed-on: https://gerrit.chromium.org/gerrit/60783 Commit-Queue: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
10711967437405d2b338bfd9426b56efb9126099 |
|
24-Jun-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] fix test_that breakage by adding BOARD label to afe BUG=chromium:253594 TEST=Ran test_that. No longer had ValidationError during suite run. test_that unit tests pass. Change-Id: I0b4eefcd1c8f1c5f2c6f58d0188e10fa1991b037 Reviewed-on: https://gerrit.chromium.org/gerrit/59830 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
a0913079c5cc26f115ccdfb4063022b7daee9d39 |
|
12-Jun-2013 |
Alex Miller <milleral@chromium.org> |
[autotest] Make board always be the metahost. When the scheduler goes to pull the list of hosts to consider for a HQE to run against, it pulls all hosts with the label specified by metahost. Currently, we use the cros-version label, if available, and pool otherwise. In preperation for preemptive scheduling, we need the scheduler to consider all hosts within any given board, since it can (plus or minus some DEPENDENCIES) choose any of them to provision for the job. It's worth noting that this CL might also speed up the scheduler tick, as we're going to be considering ~30 hosts per job, instead of ~100. (Average number of hosts per board vs average number of hosts per pool.) There's also some rework hidden in here to add |board| as a required argument to Suite. BUG=chromium:249437 TEST=local run_suite, unittest Change-Id: I4afd8ef958b96c9fb9bc4a7f838a360d508683d6 Reviewed-on: https://gerrit.chromium.org/gerrit/58386 Commit-Queue: Alex Miller <milleral@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Tested-by: Alex Miller <milleral@chromium.org>
/external/autotest/site_utils/test_that.py
|
2dc98936c15686db6576af05267a9239c48d6e61 |
|
18-Jun-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] run quickmerge as part of test_that This CL causes autotest_quickmerge to be invoked at the start of each test_that local run, right before test_that re-executes its sysroot version. This saves a step for developers, so they need not remember to merge their changes before running test_that. Now that autotest_quickmerge quickly no-ops if the sysroot is already newer than the source tree, this is safe to do with each test_that run. BUG=chromium:248545 TEST=Ran test_that suite, saw that quickmerge ran before suite was scheduled. CQ-DEPEND=CL:I75d4cd058282c3ecee279849da53cdcafade2cf1 Change-Id: I08f2003ef2555f9436df630e08b48961f4489a01 Reviewed-on: https://gerrit.chromium.org/gerrit/58924 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
1d991ea598db9165543b57ad84c70056b9d6ec24 |
|
13-Jun-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that exit cleanly on ctrl+C, SIGINT, SIGTERM Prior to this CL, and historically in run_remote_tests, a keyboard interrupt (CTRL+C) or other form of kill signal sent to test_that would leave orphaned autoserv processes around. This CL addresses that in several ways. It is instructive to look at the process tree that gets created when a user starts a local test_that run at their desk, in the chroot, since it contains several layers of test_that bootstrapping before the autoserv processes: bash -> test_that [shell script, in path] -> sudo -> test_that.py [source tree version] -> test_that.py [sysroot tree version] -> autoserv This CL does the following: 1) SIGINT or SIGTERM signals sent to test_that.py [source tree] will get forwarded to test_that.py [sysroot]. 2) SIGINT or SIGTERM signals sent to test_that.py [sysroot] will cause a SIGINT signal to be sent to the currently running autoserv process, if one is running. After 5 seconds, if autoserv has not finished cleaning up and exiting, then it will be killed. With these changes, sending a SIGINT or SIGTERM to any process from the sudo process down results in no orphaned processes. Sending a SIGINT or SIGERM to the test_that shell script is addressed by a separate CL (see the CQ-DEPEND line below). BUG=chromium:213051 TEST=CTRL+C after kicking of a local suite with test_that results in autoserv processes exiting. Sending SIGINT or SIGKILL signals to any processes up the tree as far as the sudo command results in the same. CQ-DEPEND=CL:I2391bccbb9086582ee530a129703912690396134 Change-Id: I6bba284f28736e26e0685c9911dbaa4e82a90472 Reviewed-on: https://gerrit.chromium.org/gerrit/58475 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
69ebb6c5761f0ccab13c977f6b1facde60214642 |
|
11-Jun-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that able to schedule individual tests BUG=chromium:246750 TEST=new unittest in suite_unittest, passes. Manually ran test_that with a mix of various test / suite names in combination, and tests ran correctly. Change-Id: Ia7bd73fc916398d6e63a031dc6244b6fc6f4e0f8 Reviewed-on: https://gerrit.chromium.org/gerrit/58242 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
712128f3bafef58e1a7b036f483a6fc25df9b2e1 |
|
11-Jun-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] fix test_that root logger BUG=None TEST=INFO logging messages emitted by suite and test scheduling methods now actually appear. Change-Id: Ibe24137c59f45e21c1d1967ed1c7de7ad1f0a4da Reviewed-on: https://gerrit.chromium.org/gerrit/58241 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Alex Miller <milleral@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
d4a0430e484047965e6e2426d84a3ffd2539b2cb |
|
01-May-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] add suite-running to test_that.py This CL adds basic suite-running capability to test_that. A single or multiple suites can be specified at the command line, and run against a local DUT (see TEST for usage). CQ-DEPEND=CL:Id57038b001f18a589e665b4d044cfa03ccebc954 BUG=chromium:236471 TEST=From anywhere in the chroot, assuming you have run build_packages for the board in question: `test_that -b <board> <DUT> suite:smoke` # see suite run # run, suite, run Also, new unit tests, which pass. Change-Id: Ibdc284ab59f50c84d5deaf39e6e8c5406188b81c Reviewed-on: https://gerrit.chromium.org/gerrit/49990 Tested-by: Aviv Keshet <akeshet@chromium.org> Reviewed-by: Scott Zawalski <scottz@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
0e5d525bc44ba567dc6572644c8982495db9314c |
|
27-Apr-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] make test_that re-execute its board-sysroot-installed version In the chroot path we will be adding a script which just sudo executes test_that from the autotest source tree. However, we want to be running test_that from within the autotest tree installed on the sysroot for the board being tested. When test_that is run from elsewhere, it should re-execute the board-installed version. Down the road, there will be a few bootstrapping steps added before re-executing the sysroot version. Specifically, board-autodetection, will take place first, followed by any necessary emerge/quickmerge operations, followed by re-executing the sysroot version of the script. BUG=chromium:236471 TEST=(in chroot): 1) `test_that --board=frumpy asdf asdf` ERROR:root:/build/frumpy/ does not exist. Have you run setup_board? 2) after setup_board for lumpy, but before build packages `test_that --board=lumpy asdf asdf` ERROR:root:/build/lumpy/usr/local/autotest/ does not exist. Have you run build_packages? 3) after running build_packages `test_that --board=lumpy asdf asdf` This script does not do anything yet. Reached end of execution for script /build/lumpy/usr/local/autotest/site_utils/test_that.py with args ['--board', 'lumpy', 'asdf', 'asdf']. Change-Id: I0624984ee8d6a1b6fa0221a64d0e5ef96fd438c3 Reviewed-on: https://gerrit.chromium.org/gerrit/49384 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|
021c19f1d9feac7dc5ca63861dd09ef3d5c551d3 |
|
22-Feb-2013 |
Aviv Keshet <akeshet@chromium.org> |
[autotest] test_that script initial commit, with argument parsing First commit of test_that.py, with command line args: -------- usage: test_that.py [-h] [-b BOARD] [-i BUILD] [--args [ARGS]] REMOTE TEST [TEST ...] Run remote tests. positional arguments: REMOTE hostname[:port] for remote device. Specify :lab: to run in test lab, or :vm:PORT_NUMBER to run in vm. TEST Run given test(s). Use suite:SUITE to specify test suite. optional arguments: -h, --help show this help message and exit -b BOARD, --board BOARD Board for which the test will run. -i BUILD, --build BUILD Build to test. Device will be reimaged if necessary. Omit flag to skip reimage and test against already installed DUT image. --args [ARGS] Argument string to pass through to test. -------- CQ-DEPEND=CL:I940e09d6986fe049eb4c94a91e3e484aabccd7a4 BUG=chromium:236471 TEST=None Change-Id: Iaeb425cb5575ceb67f868d3c8d8343a58255cae8 Reviewed-on: https://gerrit.chromium.org/gerrit/49255 Reviewed-by: Aviv Keshet <akeshet@chromium.org> Tested-by: Aviv Keshet <akeshet@chromium.org> Commit-Queue: Aviv Keshet <akeshet@chromium.org>
/external/autotest/site_utils/test_that.py
|