wiki:WebKitGTK/KeepingTheTreeGreen

Version 77 (modified by clopez@igalia.com, 10 years ago) ( diff )

--

(courtesy of Martin Robinson, who introduced the notion of Garderner in WebKit world of fame)

  1. Keep the tree green: the gardener on a given day will diagnose all bot redness and try to form a hypothesis for each failure. In my experience there are several types of failures:
    1. The GTK+ port is lacking some feature or has a bug: In this case the gardener will figure out exactly what the problem is, open a bug with as much information as possible including test diffs and links to relevant bugs. The gardener will CC any people involved in WebKitGTK+ who would might know how to fix the problem.
    2. A test is exposing a general WebKit bug that only GTK+ triggers (often the case with JavaScriptCore and inspector tests): In this case the gardener will leave a comment on the original bug with the test diff or stack trace (for something like an assertion failure) and offer to gather more information for the original patch author (even resorting to debugging directly on the bots if it cannot be reproduced locally).
    3. A test is missing expectations or needs new expectations: The gardener will generate new expectations after verifying that the tests seem to work when run manually or by looking at the expected results.
    4. A DRT feature is missing: The gardener will open a bug. The idea is that we will completely avoid skipped tests with no information about the failure.
  1. If the tree isn't red, the gardener should be able to choose how to spend the day:
    1. Generating results for tests that do not have results (soon this will include pixel results as well).
    2. Diagnosing and opening bugs for failures that do not have open bugs (there's a huge list). This also includes unskipping tests that are passing.
    3. Implementing missing DRT features. There are a ton of these and we need to kill them. Using DRTSupportGtk we can do it without making API decisions.
  1. At the end of the day the gardener should keep a very simple log for tomorrow's gardener, so he/she can follow up with any pending investigations. This will also help us keep track of newly skipped tests. See the Gardening log entry at the bottom of this wiki page.

Check also chrome gardening instructions: ​http://dev.chromium.org/developers/how-tos/webkit-gardening

Gardening schedule

Day Gardener
Wed 02 Jul clopez
Thu 03 Jul
Fri 04 Jul
Sat 05 Jul
Sun 06 Jul ltilve
Mon 07 Jul
Tue 08 Jul
Wed 09 Jul clopez
Thu 10 Jul
Fri 11 Jul
Sat 12 Jul
Sun 13 Jul
Mon 14 Jul
Tue 15 Jul
Wed 16 Jul
Thu 17 Jul
Fri 18 Jul
Sat 19 Jul
Sun 20 Jul
Mon 21 Jul
Tue 22 Jul
Wed 23 Jul
Thu 24 Jul
Fri 25 Jul
Sat 26 Jul
Sun 27 Jul
Mon 28 Jul
Tue 29 Jul
Wed 30 Jul
Thu 31 Jul
Fri 01 Aug
Sat 02 Aug

Gardening log

  • Please put new log entries at the top.

Done:

Pending:

  • Check the remaining failures.

-- ltilve@igalia.com - Mon Jul 07 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Wed Jul 02 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Fri Jun 27 2014

Done:

Pending:

  • Check the remaining failures.

-- eocanha@igalia.com - Fri Jun 13 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Thu Jun 12 2014

Done:

Pending:

  • Check the remaining failures.

-- ltilve@igalia.com - Wed Jun 11 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Wed Jun 04 2014

Done:

-- ltilve@igalia.com - Thu May 21 2014

Done:

Pending:

  • Check the remaining failures.

-- eocanha@igalia.com - Fri May 16 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Thu May 15 2014

Done:

-- ltilve@igalia.com - Thu May 13 2014

Done:

-- ltilve@igalia.com - Thu May 7 2014

Done:

-- clopez@igalia.com - Thu May 01 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Thu May 01 2014

Done:

-- ltilve@igalia.com - Thu Apr 29 2014

Done:

-- calvaris@igalia.com - Thu Apr 28 2014

Done:

Pending:

  • Check the remaining failures.

-- eocanha@igalia.com - Fri Apr 25

Done:

Notes:

  • If you notice that a test started failing on r167754 it could be because
    of the removal of the XVFB_SCREEN_DEPTH=8 variable. Let me know that
    or update ​http://webkit.org/b/132126

-- clopez@igalia.com - Thu Apr 24 2014

Done:

Pending:

  • Check the remaining failures.

-- elima@igalia.com - Thu Apr 24 2014

Done:

Pending:

  • Check the remaining failures.

-- ltilve@igalia.com - Mon Apr 22 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Mon Apr 21 2014

Done:

-- elima@igalia.com - Thu Apr 17 2014

Done:

Pending:

  • Check the remaining failures.

-- aperez@igalia.com - Wed Apr 16 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Wed Apr 16 2014

Done:

Pending:

  • Check the remaining failures.

-- ltilve@igalia.com - Tue Apr 15 2014

Done:

Pending:

  • Check the remaining failures.

-- elima@igalia.com - Fri Apr 14 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Fri Apr 11 2014

Done:

Pending:

  • Check the remaining failures.

-- eocanha@igalia.com - Fri Apr 11 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Thu Apr 10 2014

Done:

  • Checked the status of garden-o-matic:
    • Works fine for WebKitGTK+ port when using a Git clone.
    • Usage instructions in the Rebaseline page.

aperez@igalia.com - Wed 9 Apr 2014

Done:

Pending:

  • Check the remaining failures.

-- clopez@igalia.com - Tue Apr 8 2014

Note: See TracWiki for help on using the wiki.