1220246 - Intermittent test_geodefaults.js | should_get_geo_defaults_only_once - [should_get_geo_defaults_only_once : 104] "undefined" == "number"

Status: RESOLVED FIXED

(Firefox :: Search, defect, P3)

Description

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://treeherder.mozilla.org/logviewer.html#?job_id=5497912&repo=fx-team

Comment 1

[Intermittent Failures Robot]

27 automation job failures were associated with this bug in the last 7 days.

Repository breakdown:
* mozilla-inbound: 16
* fx-team: 7
* mozilla-central: 2
* try: 1
* b2g-inbound: 1

Platform breakdown:
* windows8-64: 22
* windows7-32: 5

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2015-10-26&endday=2015-11-01&tree=all

Comment 2

[Intermittent Failures Robot]

26 automation job failures were associated with this bug yesterday.

Repository breakdown:
* mozilla-inbound: 21
* fx-team: 3
* mozilla-central: 1
* b2g-inbound: 1

Platform breakdown:
* windows8-64: 24
* windows7-32: 2

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2015-11-02&endday=2015-11-02&tree=all

Comment 3

[Intermittent Failures Robot]

16 automation job failures were associated with this bug yesterday.

Repository breakdown:
* mozilla-inbound: 8
* fx-team: 4
* mozilla-central: 3
* try: 1

Platform breakdown:
* windows8-64: 15
* windows7-32: 1

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2015-11-04&endday=2015-11-04&tree=all

Comment 4

[Florian Quèze [:florian]]

Attachment: Patch to reproduce the failure

The failure occurs when it takes more than 1000ms for the local test HTTP server to reply to the /lookup_defaults requests. The test slaves that showed this failure must be pretty overloaded. The attached patch forces a 1000ms wait to reproduce locally and debug the test.

Comment 5

[Florian Quèze [:florian]]

https://treeherder.mozilla.org/#/jobs?repo=try&revision=de466242fe4a

Comment 6

[Intermittent Failures Robot]

32 automation job failures were associated with this bug yesterday.

Repository breakdown:
* mozilla-inbound: 18
* b2g-inbound: 6
* fx-team: 5
* mozilla-central: 3

Platform breakdown:
* windows8-64: 28
* windows7-32: 4

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2015-11-05&endday=2015-11-05&tree=all

Comment 7

[Florian Quèze [:florian]]

Attachment: fix test_geodefaults.js to avoid failing when the http server takes more than 1000ms to reply, r=?

Try looks green, so I would like to try landing this simple patch.

The problem here is that whenever something related to search is changed, the search service caches things to disk with a 1000ms delay. The test currently expects one notification of the file being written to disk, but when the HTTP server takes more than a second to respond, the file gets written to disk twice, and the promise only waits for the first notification. The patch addresses this by creating the promise after the search service is already done initializing. There's no risk of the (relevant) notification having been already fired at that point, because the file is written to disk with a delay.

Comment 8

[Florian Quèze [:florian]]

Attachment: Patch v2

Looks like the reason why we are writing the file to disk twice during the should_get_geo_defaults_only_once test is that the promiseAfterCache from the no_request_if_prefed_off test can be resolved due to writing the initial cache rather than due to the currentEngine/resetToOriginalDefaultEngine code block. And also, this code is pointless now that the metadata is no longer stored in a separate file.

Comment 9

[Florian Quèze [:florian]]

https://hg.mozilla.org/integration/fx-team/rev/c921afa9fc3e164ab52dc4564349a92e16663317
Bug 1220246 - fix test_geodefaults.js to avoid failing when the http server takes more than 1000ms to reply, r=Gijs.

Comment 10

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/c921afa9fc3e

Comment 11

[Intermittent Failures Robot]

22 automation job failures were associated with this bug yesterday.

Repository breakdown:
* mozilla-inbound: 14
* b2g-inbound: 5
* mozilla-central: 2
* fx-team: 1

Platform breakdown:
* windows8-64: 20
* windows7-32: 2

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2015-11-06&endday=2015-11-06&tree=all

Comment 12

[Intermittent Failures Robot]

129 automation job failures were associated with this bug in the last 7 days.

Repository breakdown:
* mozilla-inbound: 77
* fx-team: 19
* b2g-inbound: 18
* mozilla-central: 13
* try: 2

Platform breakdown:
* windows8-64: 118
* windows7-32: 11

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2015-11-02&endday=2015-11-08&tree=all

Comment 13

[Intermittent Failures Robot]

5 automation job failures were associated with this bug in the last 7 days.

Repository breakdown:
* mozilla-inbound: 5

Platform breakdown:
* windows8-64: 2
* linux64: 2
* windows7-32: 1

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2016-03-07&endday=2016-03-13&tree=all

Comment 14

[Ryan VanderMeulen [:RyanVM]]

Looks like this never really went away.

Comment 15

[Intermittent Failures Robot]

9 automation job failures were associated with this bug in the last 7 days.

Repository breakdown:
* mozilla-inbound: 8
* fx-team: 1

Platform breakdown:
* linux64: 7
* windows8-64: 2

For more details, see:
https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1220246&startday=2016-05-23&endday=2016-05-29&tree=all

Comment 16

[BMO Automation]

Bulk assigning P3 to all open intermittent bugs without a priority set in Firefox components per bug 1298978.