| Andrew Geissler | f034379 | 2020-11-18 10:42:21 -0600 | [diff] [blame] | 1 | .. SPDX-License-Identifier: CC-BY-SA-2.0-UK | 
| Andrew Geissler | c9f7865 | 2020-09-18 14:11:35 -0500 | [diff] [blame] | 2 |  | 
|  | 3 | *********************************** | 
|  | 4 | Project Testing and Release Process | 
|  | 5 | *********************************** | 
|  | 6 |  | 
|  | 7 | .. _test-daily-devel: | 
|  | 8 |  | 
|  | 9 | Day to Day Development | 
|  | 10 | ====================== | 
|  | 11 |  | 
|  | 12 | This section details how the project tests changes, through automation | 
|  | 13 | on the Autobuilder or with the assistance of QA teams, through to making | 
|  | 14 | releases. | 
|  | 15 |  | 
|  | 16 | The project aims to test changes against our test matrix before those | 
|  | 17 | changes are merged into the master branch. As such, changes are queued | 
|  | 18 | up in batches either in the ``master-next`` branch in the main trees, or | 
|  | 19 | in user trees such as ``ross/mut`` in ``poky-contrib`` (Ross Burton | 
|  | 20 | helps review and test patches and this is his testing tree). | 
|  | 21 |  | 
|  | 22 | We have two broad categories of test builds, including "full" and | 
|  | 23 | "quick". On the Autobuilder, these can be seen as "a-quick" and | 
|  | 24 | "a-full", simply for ease of sorting in the UI. Use our Autobuilder | 
|  | 25 | console view to see where me manage most test-related items, available | 
|  | 26 | at: :yocto_ab:`/typhoon/#/console`. | 
|  | 27 |  | 
|  | 28 | Builds are triggered manually when the test branches are ready. The | 
|  | 29 | builds are monitored by the SWAT team. For additional information, see | 
|  | 30 | :yocto_wiki:`/wiki/Yocto_Build_Failure_Swat_Team`. | 
|  | 31 | If successful, the changes would usually be merged to the ``master`` | 
|  | 32 | branch. If not successful, someone would respond to the changes on the | 
|  | 33 | mailing list explaining that there was a failure in testing. The choice | 
|  | 34 | of quick or full would depend on the type of changes and the speed with | 
|  | 35 | which the result was required. | 
|  | 36 |  | 
|  | 37 | The Autobuilder does build the ``master`` branch once daily for several | 
|  | 38 | reasons, in particular, to ensure the current ``master`` branch does | 
|  | 39 | build, but also to keep ``yocto-testresults`` | 
|  | 40 | (:yocto_git:`/cgit.cgi/yocto-testresults/`), | 
|  | 41 | buildhistory | 
|  | 42 | (:yocto_git:`/cgit.cgi/poky-buildhistory/`), and | 
|  | 43 | our sstate up to date. On the weekend, there is a master-next build | 
|  | 44 | instead to ensure the test results are updated for the less frequently | 
|  | 45 | run targets. | 
|  | 46 |  | 
|  | 47 | Performance builds (buildperf-\* targets in the console) are triggered | 
|  | 48 | separately every six hours and automatically push their results to the | 
|  | 49 | buildstats repository at: | 
|  | 50 | :yocto_git:`/cgit.cgi/yocto-buildstats/`. | 
|  | 51 |  | 
|  | 52 | The 'quick' targets have been selected to be the ones which catch the | 
|  | 53 | most failures or give the most valuable data. We run 'fast' ptests in | 
|  | 54 | this case for example but not the ones which take a long time. The quick | 
|  | 55 | target doesn't include \*-lsb builds for all architectures, some world | 
|  | 56 | builds and doesn't trigger performance tests or ltp testing. The full | 
|  | 57 | build includes all these things and is slower but more comprehensive. | 
|  | 58 |  | 
|  | 59 | Release Builds | 
|  | 60 | ============== | 
|  | 61 |  | 
|  | 62 | The project typically has two major releases a year with a six month | 
|  | 63 | cadence in April and October. Between these there would be a number of | 
|  | 64 | milestone releases (usually four) with the final one being stablization | 
|  | 65 | only along with point releases of our stable branches. | 
|  | 66 |  | 
|  | 67 | The build and release process for these project releases is similar to | 
|  | 68 | that in `Day to Day Development <#test-daily-devel>`__, in that the | 
|  | 69 | a-full target of the Autobuilder is used but in addition the form is | 
|  | 70 | configured to generate and publish artefacts and the milestone number, | 
|  | 71 | version, release candidate number and other information is entered. The | 
|  | 72 | box to "generate an email to QA"is also checked. | 
|  | 73 |  | 
|  | 74 | When the build completes, an email is sent out using the send-qa-email | 
|  | 75 | script in the ``yocto-autobuilder-helper`` repository to the list of | 
|  | 76 | people configured for that release. Release builds are placed into a | 
|  | 77 | directory in https://autobuilder.yocto.io/pub/releases on the | 
|  | 78 | Autobuilder which is included in the email. The process from here is | 
|  | 79 | more manual and control is effectively passed to release engineering. | 
|  | 80 | The next steps include: | 
|  | 81 |  | 
|  | 82 | -  QA teams respond to the email saying which tests they plan to run and | 
|  | 83 | when the results will be available. | 
|  | 84 |  | 
|  | 85 | -  QA teams run their tests and share their results in the yocto- | 
|  | 86 | testresults-contrib repository, along with a summary of their | 
|  | 87 | findings. | 
|  | 88 |  | 
|  | 89 | -  Release engineering prepare the release as per their process. | 
|  | 90 |  | 
|  | 91 | -  Test results from the QA teams are included into the release in | 
|  | 92 | separate directories and also uploaded to the yocto-testresults | 
|  | 93 | repository alongside the other test results for the given revision. | 
|  | 94 |  | 
|  | 95 | -  The QA report in the final release is regenerated using resulttool to | 
|  | 96 | include the new test results and the test summaries from the teams | 
|  | 97 | (as headers to the generated report). | 
|  | 98 |  | 
|  | 99 | -  The release is checked against the release checklist and release | 
|  | 100 | readiness criteria. | 
|  | 101 |  | 
|  | 102 | -  A final decision on whether to release is made by the YP TSC who have | 
|  | 103 | final oversight on release readiness. |