Bug 172030

It looks like this may have started with https://trac.webkit.org/changeset/216711/webkit

(In reply to Ryan Haddad from comment #1) > It looks like this may have started with > https://trac.webkit.org/changeset/216711/webkit If 216711 caused this, then I think this test was utterly broken in the first place.

(In reply to Ryan Haddad from comment #1) > It looks like this may have started with > https://trac.webkit.org/changeset/216711/webkit You say *MAY* have started with this... how sure are you?

(In reply to Brady Eidson from comment #3) > (In reply to Ryan Haddad from comment #1) > > It looks like this may have started with > > https://trac.webkit.org/changeset/216711/webkit > > You say *MAY* have started with this... how sure are you? I said may because I did not reproduce it locally, but regression range is between https://trac.webkit.org/changeset/216709 and https://trac.webkit.org/changeset/216712. None of the other commits look related.

Cannot reproduce this locally no matter what I try. Looking at changes to the test to make it more resilient.

Created attachment 309927 [details] Patch for suppression/exploration

Comment on attachment 309927 [details] Patch for suppression/exploration Rejecting attachment 309927 [details] from commit-queue. Failed to run "['/Volumes/Data/EWS/WebKit/Tools/Scripts/webkit-patch', '--status-host=webkit-queues.webkit.org', '--bot-id=webkit-cq-02', 'validate-changelog', '--check-oops', '--non-interactive', 309927, '--port=mac']" exit_code: 1 cwd: /Volumes/Data/EWS/WebKit Nobody - Test gardening, in effect found in /Volumes/Data/EWS/WebKit/Tools/ChangeLog does not appear to be a valid reviewer according to contributors.json. /Volumes/Data/EWS/WebKit/Tools/ChangeLog neither lists a valid reviewer nor contains the string "Unreviewed" or "Rubber stamp" (case insensitive). Full output: http://webkit-queues.webkit.org/results/3727676

Created attachment 309928 [details] Patch for suppression/exploration

Comment on attachment 309928 [details] Patch for suppression/exploration Clearing flags on attachment: 309928 Committed r216777: <http://trac.webkit.org/changeset/216777>

All reviewed patches have been landed. Closing bug.

New failure on bots: /Volumes/Data/slave/elcapitan-release/build/Tools/TestWebKitAPI/Tests/WebKit2Cocoa/LocalStorageClear.mm:170 Value of: 1ul Actual: 1 Expected: websiteDataRecords.count Which is: 2 /Volumes/Data/slave/elcapitan-release/build/Tools/TestWebKitAPI/Tests/WebKit2Cocoa/LocalStorageClear.mm:186 Value of: originalWebsiteDataRecordCount - 1 Actual: 1 Expected: websiteDataRecords.count Which is: 0

(In reply to Brady Eidson from comment #11) > New failure on bots: > > /Volumes/Data/slave/elcapitan-release/build/Tools/TestWebKitAPI/Tests/ > WebKit2Cocoa/LocalStorageClear.mm:170 > Value of: 1ul > Actual: 1 > Expected: websiteDataRecords.count > Which is: 2 Since we wiped the entire data store directory *and* installed only a single record of app cache ness, we expect exactly *one* record here. Getting 2 is entirely unexpected. > /Volumes/Data/slave/elcapitan-release/build/Tools/TestWebKitAPI/Tests/ > WebKit2Cocoa/LocalStorageClear.mm:186 > Value of: originalWebsiteDataRecordCount - 1 > Actual: 1 > Expected: websiteDataRecords.count > Which is: 0 This is actually right - At the end of the test, there should absolutely be 0 results left for getting all data of all types.

Created attachment 309945 [details] Get more info from the bots

The commit-queue encountered the following flaky tests while processing attachment 309945 [details]: The commit-queue is continuing to process your patch.

The commit-queue encountered the following flaky tests while processing attachment 309945 [details]: http/tests/misc/acid3.html bug 165734 (authors: abecsi@webkit.org, arv@chromium.org, cdumez@apple.com, rniwa@webkit.org, and tonyg@chromium.org) The commit-queue is continuing to process your patch.

Comment on attachment 309945 [details] Get more info from the bots Clearing flags on attachment: 309945 Committed r216800: <http://trac.webkit.org/changeset/216800>

All reviewed patches have been landed. Closing bug.

Reopening since the landed patch will not fix this, but rather hopefully just tell us what's broken.

Finally got the unexpected results: /Volumes/Data/slave/sierra-release/build/Tools/TestWebKitAPI/Tests/WebKit2Cocoa/LocalStorageClear.mm:172 Value of: [record.displayName UTF8String] Actual: "Local documents on your computer" Expected: "127.0.0.1" /Volumes/Data/slave/sierra-release/build/Tools/TestWebKitAPI/Tests/WebKit2Cocoa/LocalStorageClear.mm:174 Value of: [type UTF8String] Actual: "WKWebsiteDataTypeIndexedDBDatabases" Expected: [WKWebsiteDataTypeOfflineWebApplicationCache UTF8String] Which is: "WKWebsiteDataTypeOfflineWebApplicationCache" So, bizarrely, there are IndexedDB remnants around. No clue why, and why nobody can reproduce locally instead of on the bots.

Also quite bizarre that deleting the IDB directory at the start doesn't resolve this.

I see, app cache doesn't go into that directory. Okay, fine. Let's resolve this once and for all.

Created attachment 309969 [details] Patch to fix

Attachment 309969 [details] did not pass style-queue: ERROR: Tools/TestWebKitAPI/Tests/WebKit2Cocoa/LocalStorageClear.mm:139: This { should be at the end of the previous line [whitespace/braces] [4] Total errors found: 1 in 2 files If any of these errors are false positives, please file a bug against check-webkit-style.

Comment on attachment 309969 [details] Patch to fix Clearing flags on attachment: 309969 Committed r216805: <http://trac.webkit.org/changeset/216805>

All reviewed patches have been landed. Closing bug.