1891185 - HTTPS-First Telemetry is collecting bad downgrade time

Status: RESOLVED FIXED

(Core :: DOM: Security, defect, P3)

Description

[Malte Jürgens [:maltejur]]

Something seems to be wrong with the httpsfirst.downgrade_time and httpsfirst.downgrade_time_schemeless metrics. We have collected downgrade times as high as 35000s, and downgrade times generally seem to be much higher for schemeless HTTPS-First than for normal HTTPS-First, which we didn't expect.

Comment 1

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

Set release status flags based on info from the regressing bug 1868380

Comment 2

[Malte Jürgens [:maltejur]]

Attachment: Bug 1891185 - Collect different downgrade time on timer for HTTPS-First telemetry r?freddyb

It does not make sense to collect the "downgrade time" as we currently do it when a upgrade was caused by a faster http request on a timer. We can not compare the navigation start timestamp with the current time, as the replacing http channel has started loading earlier than the current time. Instead, collecting the configured value of the timer (3s) makes more sense, as it accurately reflects the time overhead caused by HTTPS-First.

Comment 3

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

Set release status flags based on info from the regressing bug 1868380

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/754af9f77b63
Collect different downgrade time on timer for HTTPS-First telemetry r=freddyb

Comment 5

[Aron Cseh]

Backed out for causing mochitest failures on browser_httpsfirst.js

• Backout link
• Push with failures
• Failure Log

[task 2024-04-29T12:53:09.635Z] 12:53:09     INFO - TEST-PASS | dom/security/test/https-first/browser_httpsfirst.js | undefined assertion name - 
[task 2024-04-29T12:53:09.636Z] 12:53:09     INFO - Buffered messages finished
[task 2024-04-29T12:53:09.637Z] 12:53:09     INFO - TEST-UNEXPECTED-FAIL | dom/security/test/https-first/browser_httpsfirst.js | Summed downgrade time should be above 2 and below 30 seconds (is 0.10s) - 
[task 2024-04-29T12:53:09.637Z] 12:53:09     INFO - Stack trace:
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochikit/content/browser-test.js:test_ok:1592
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochitests/content/browser/dom/security/test/https-first/browser_httpsfirst.js:null:102
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochikit/content/browser-test.js:handleTask:1139
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochikit/content/browser-test.js:_runTaskBasedTest:1211
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochikit/content/browser-test.js:Tester_execTest:1353
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochikit/content/browser-test.js:nextTest/<:1128
[task 2024-04-29T12:53:09.638Z] 12:53:09     INFO - chrome://mochikit/content/tests/SimpleTest/SimpleTest.js:SimpleTest.waitForFocus/<:1058
[task 2024-04-29T12:53:09.639Z] 12:53:09     INFO - TEST-PASS | dom/security/test/https-first/browser_httpsfirst.js | undefined assertion name - 
[task 2024-04-29T12:53:09.639Z] 12:53:09     INFO - Leaving test bound 
[task 2024-04-29T12:53:09.761Z] 12:53:09     INFO - GECKO(2847) | MEMORY STAT | vsize 11322MB | residentFast 353MB | heapAllocated 180MB
[task 2024-04-29T12:53:09.762Z] 12:53:09     INFO - TEST-OK | dom/security/test/https-first/browser_httpsfirst.js | took 7061ms

Comment 6

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/2deb5d0fae7d
Collect different downgrade time on timer for HTTPS-First telemetry r=freddyb

Comment 7

[Cristian Tuns]

https://hg.mozilla.org/mozilla-central/rev/2deb5d0fae7d

Comment 8

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

The patch landed in nightly and beta is affected.
:mjurgens, is this bug important enough to require an uplift?

• If yes, please nominate the patch for beta approval.
• If no, please set status-firefox126 to wontfix.

For more information, please visit BugBot documentation.