Dan Morrill | 3cd199f | 2009-11-06 14:04:16 -0800 | [diff] [blame] | 1 | page.title=Life of a Bug |
| 2 | doc.type=source |
| 3 | @jd:body |
| 4 | <p>The Android Open Source project maintains a public issue tracker where you |
| 5 | can report bugs and request features for the Android software stack. (For |
| 6 | details on this issue tracker, please see <a |
| 7 | href="{@docRoot}source/report-bugs.html">the Reporting Bugs page</a>.) |
| 8 | Reporting bugs is great (thank you!), but what happens to a bug report once |
| 9 | you file it? This page describes the Life of a Bug.</p> |
| 10 | <p><i>Please note: the the Android Open Source Project (AOSP) issue tracker is |
| 11 | intended only for bugs & feature requests related to the Android software |
| 12 | stack. Because many users find their way here looking for the Google apps for |
| 13 | Android (such as Gmail and so on), we have components set up for their |
| 14 | convenience. However, these apps are not part of Android, and any issues |
| 15 | reported on these components are not guaranteed to to receive attention. |
| 16 | Most notably, to report issues related to Android Market, you should visit the |
| 17 | <a href="http://www.google.com/support/forum/p/Android+Market?hl=en">Android |
| 18 | Market Support Forum</a>.</i></p> |
| 19 | <p>Here's the Life of a Bug, in a nutshell:</p> |
| 20 | <ol> |
| 21 | <li>A bug is filed, and has the state "New".</li> |
| 22 | <li>An AOSP contributor periodically reviews and triages bugs. Bugs are |
| 23 | triaged into one of four "buckets": New, Open, No-Action, or Resolved.</li> |
| 24 | <li>Each bucket includes a number of states that provide more detail on the |
| 25 | fate of the issue.</li> |
| 26 | <li>Bugs in the "Resolved" bucket will eventually be included in a future |
| 27 | release of the Android software.</li> |
| 28 | </ol> |
| 29 | <h2>Bucket Details</h2> |
| 30 | <p>Here is some additional information on each bucket, what it means, and how |
| 31 | it's handled.</p> |
| 32 | <h3>New Issues</h3> |
| 33 | <p>New issues include bug reports that are not yet being acted upon. The two |
| 34 | states are:</p> |
| 35 | <ul> |
| 36 | <li><b>New</b><p>The bug report has not yet been triaged (that is, reviewed by |
| 37 | an AOSP contributor.)</p></li> |
| 38 | <li><b>NeedsInfo</b><p>The bug report has insufficient information to act |
| 39 | upon. The person who reported the bug needs to provide additional detail |
| 40 | before it can be triaged. If enough time passes and no new information is |
| 41 | provided, the bug may be closed by default, as one of the No-Action |
| 42 | states.</p></li> |
| 43 | </ul> |
| 44 | |
| 45 | <h3>Open Issues</h3> |
| 46 | <p>This bucket contains bugs that need action, but which are still |
| 47 | unresolved, pending a change to the source code.</p> |
| 48 | <ul> |
| 49 | <li><b>Unassigned</b><p>The bug report has been recognized as an adequately |
| 50 | detailed report of a legitimate issue, but has not yet been assigned to an |
| 51 | AOSP contributor to be fixed. Typically, bugs in this state are considered low |
| 52 | priority, at least insofar that if they were high priority, they'd be assigned |
| 53 | to a contributor.</p></li> |
| 54 | <li><b>Reviewed</b><p>Like <code>Unassigned</code>, but the issue |
| 55 | represented is being tracked in a separate bug database. For example, the bug |
Dan Morrill | 55de681 | 2009-11-15 15:46:36 -0800 | [diff] [blame] | 56 | might have been reported via an internal bug-tracking system, |
| 57 | which is considered the "master" copy. (For instance, Google maintains one |
| 58 | such private issue tracker, intended primarily for bugs which contain |
| 59 | sensitive information which can't be revealed publicly.)</p></li> |
Dan Morrill | 3cd199f | 2009-11-06 14:04:16 -0800 | [diff] [blame] | 60 | <li><b>Assigned</b><li>Like <code>Unassigned</code>, but the bug has been |
| 61 | actually assigned to a specific contributor to fix.</p></li> |
| 62 | </ul> |
| 63 | <p>Typically, a given bug will start in <code>Unassigned</code>, where it |
| 64 | will remain until it is associated with a specific upcoming release, at which |
| 65 | point it will enter <code>Reviewed</code> or <code>Assigned</code>. However, |
| 66 | note that this isn't a guarantee, and it's not uncommon for bugs to go from |
| 67 | <code>Unassigned</code> to one of the Resolved states.</p> |
| 68 | <p>In general, if a bug is in one of these Open states, the AOSP team has |
| 69 | recognized it as a legitimate issue and will fix it according to the product |
| 70 | priorities and milestones. However, it's impossible to guarantee a fix in time |
| 71 | for any particular release.</p> |
| 72 | |
| 73 | <h3>No-Action Issues</h3> |
| 74 | <p>This bucket contains bugs that have for one reason or another been |
| 75 | determined to not require any action.</p> |
| 76 | <ul> |
| 77 | <li><b>Spam</b><p>A kind soul sent us some delicious pork products, that we, |
| 78 | regrettably, do not want.</p></li> |
| 79 | <li><b>Question</b><p>Someone mistook the issue tracker for a help forum. |
| 80 | (This is not as uncommon as one might assume: many users whose native language |
| 81 | isn't English can make this mistake.)</p></li> |
| 82 | <li><b>Unreproducible</b><p>An AOSP contributor attempted to reproduce the |
| 83 | behavior described, and was unable to do so. This sometimes means that the bug |
| 84 | is legitimate but simply rare or difficult to reproduce, and sometimes means |
| 85 | that the bug was fixed in a later release.</p></li> |
| 86 | <li><b>WorkingAsIntended</b><p>An AOSP contributor has determined that the |
| 87 | behavior described isn't a bug, but is the intended behavior. This state is |
| 88 | also commonly referred to as "WAI".</b></li> |
| 89 | <li><b>Declined</b><p>This is like <code>WorkingAsIntended</code>, except |
| 90 | typically used for feature requests instead of bugs. That is, an AOSP |
| 91 | contributor has determined that the request is not going to be implemented in |
| 92 | Android.</b></li> |
| 93 | </ul> |
| 94 | |
| 95 | <h3>Resolved Issues</h3> |
| 96 | <p>This bucket contains bugs that have had action taken, and are now |
| 97 | considered resolved.</p> |
| 98 | <ul> |
| 99 | <li><b>FutureRelease</b><p>This bug has been fixed (or feature implemented) in |
| 100 | a source tree, but has not yet been included in a formal Android |
| 101 | platform release. (Note that this may also include fixes that exist in a |
| 102 | private source tree that has not yet been contributed to a public |
| 103 | tree.)</p></li> |
| 104 | <li><b>Released</b><p>This bug has been fixed, and is included in a formal |
| 105 | Android platform release. When this state is set, we try to also set a |
| 106 | property indicating which release it was fixed in.</p></li> |
| 107 | <li><b>Duplicate</b><p>This bug is a duplicate of another, existing bug |
| 108 | report.</p></li> |
| 109 | </ul> |
| 110 | |
| 111 | <h2>Other Stuff</h2> |
| 112 | <p>The states and lifecycle above are how we generally try to track software. |
| 113 | However, Android contains a lot of software and gets a correspondingly large |
| 114 | number of bugs. As a result, sometimes bugs don't make it through all the |
| 115 | states in a formal progression. We do try to keep the system up to date, but |
| 116 | we tend to do so in periodic "bug sweeps" where we review the database and |
| 117 | make updates.</p> |
| 118 | <p>Since the AOSP is essentially constantly evolving, we do make tweaks to |
| 119 | the list of bug states and the lifecycle described above. When we do this, |
| 120 | however, we'll be sure to update this page as well.</p> |
| 121 | <p>Finally, you should be aware that for a variety of reasons, there are |
| 122 | actually multiple issue trackers for Android-related issues. The <a |
| 123 | href="http://code.google.com/p/android/issues/list">Google Code Project |
| 124 | Hosting Issue Tracker</a> is the <b>only</b> official public issue tracker; however, |
Dan Morrill | 55de681 | 2009-11-15 15:46:36 -0800 | [diff] [blame] | 125 | Google also maintains a private issue tracker, own, as do most OEMs. We try to |
| 126 | keep the public issue tracker in sync with private issue trackers |
Dan Morrill | 3cd199f | 2009-11-06 14:04:16 -0800 | [diff] [blame] | 127 | wherever possible, but in cases where confidential information and security |
| 128 | issues are involved, this isn't always possible.</p> |