1903928 - Permanent failure /resource-timing/no-entries-for-cross-origin-css-fetched.sub.html | single tracking bug

Status: RESOLVED FIXED

(Core :: DOM: Performance APIs, defect)

Description

[Treeherder Bug Filer]

Filed by: istorozhko [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=462066925&repo=try
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/C0n3u_bKQKSW3EIYvckZrA/runs/0/artifacts/public/logs/live_backing.log

---

This test fails on all Android Fission builds,
with both `isolate-everything` and `isolate-high-value` Fission site isolation strategies.

Filed using Intermittent Filer, but this failure is permanent.

Blocks Fission on Android

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

The severity field is not set for this bug.
:fdoty, could you have a look please?

For more information, please visit BugBot documentation.

Comment 2

[Frank Doty [:fdoty]]

:sefeng can you please review for DOM perf?

Comment 3

[[:owlish] 🦉 PST]

Hi Frank, can this bug be prioritized? We are planning on running Fission experiment in 135 Nightly, so it would be great to get this done by then. Thank you!

Comment 4

[Sean Feng [:sefeng211]]

I think this should be handled at https://searchfox.org/mozilla-central/rev/5717a81fa87a3a4cc1f8cac16e7b0de52ebebcea/layout/style/Loader.cpp#1591 where we don't allow this resource to be added for CORS failed stylesheet. Somehow this doesn't work on Android?

Comment 5

[[:owlish] 🦉 PST]

Looking closer to this one, the failure is UNEXPECTED-PASS. Should the test actually pass? Should we simply adjust the expectation, or is the test expected to fail? The FAIL annotation seems to be Android-specific

Comment 6

[Sean Feng [:sefeng211]]

Looks like the job is expired, so I can't check..

wpt.fyi still shows it as failure though https://wpt.fyi/results/resource-timing/no-entries-for-cross-origin-css-fetched.sub.html?label=master&label=experimental&aligned&q=no-entries-for-cross-origin-css-fetched.sub.html

I mean, if some changes make it start to pass, I think we should find which commit is that first and then change the expectation.

Comment 7

[[:owlish] 🦉 PST]

Hm, I might have confused it with something yesterday, here is a recent failure https://treeherder.mozilla.org/jobs?repo=try&revision=99fcd8f83b5e6a4e9b2334b47170e8480b77f9fa&selectedTaskRun=bLkHYxqqRZGlFJxVQWJ8yA.0 (it is UNEXPECTED-FAIL)

Is it something you can take a look at? Thanks!

Comment 8

[Sean Feng [:sefeng211]]

that's for bug 1903924, right? I just commented there

Comment 9

[Intermittent Failures Robot]

1 failures in 4227 pushes (0.0 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• try: 1

Platform and build breakdown:

• android-em-7-0-x86_64-qr: 1
  • debug: 1

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1903928&startday=2024-11-11&endday=2024-11-17&tree=all

Comment 10

[Chris Peterson [:cpeterson]]

Irene says she will look into changing this test's expectation from FAIL to PASS.

This bug summary says "Permanent failure", but the Intermittent Failures Robot says in comment 9 that this test had 1 failure (i.e. TEST-UNEXPECTED-PASS) in 4227 pushes in the last 7 days. What happened there?

Comment 11

[Olivia Hall [:olivia]]

I'm inclined to think this test should either be updated to [PASS] or [PASS, FAIL] as well.

The other metadata has it as [PASS, FAIL] for Linux. Android seems to often (consistently?) give a PASS response now, but I'll try and verify that it isn't sometimes [PASS, FAIL].

I'll likely update this via bug 1925417. Possible this was fixed by the TIMEOUT bug being fixed and allowing tests to proceede as expected in bug 1903929. The TIMEOUT bug was blocking a lot of tests from running normally. It is possible this bug is still intermitent, but now intermittently passing as well, since the test harness is more robust now.

This bug summary says "Permanent failure", but the Intermittent Failures Robot says in comment 9 that this test had 1 failure (i.e. TEST-UNEXPECTED-PASS) in 4227 pushes in the last 7 days. What happened there?

Really strange, might be evidence that the metadata needs to be [PASS, FAIL] like as set on Linux. I'm seeing more PASSes lately, so maybe it was the start of that?

wpt.fyi still shows it as failure though https://wpt.fyi/results/resource-timing/no-entries-for-cross-origin-css-fetched.sub.html?label=master&label=experimental&aligned&q=no-entries-for-cross-origin-css-fetched.sub.html

IIUC, isn't this link for the Linux Desktop build? Android does have its own build/harness and a lot of times things can be different. For example, SHIP and Fission are just now on the way to shipping on Android. (Both the W and W-fis suites run fission and we have a few nofis suites.)

I'm sending up trys testing some things on bug 1925417.

Please join the conversation over on bug 1925417, especially if you think this test should not be updated to [PASS] or [PASS, FAIL].

Comment 12

[Olivia Hall [:olivia]]

Fixed by bug 1925417.