1554264 - Perma shippable opt browser_all_files_referenced.js | Uncaught exception - [Exception... "Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIChromeRegistry.convertChromeURL]" when Gecko 68 merges to release on 2019-07-01

Status: RESOLVED FIXED

(Web Compatibility :: Tooling & Investigations, defect, P1)

Description

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Beta-as-release simulation: https://treeherder.mozilla.org/#/jobs?repo=try&resultStatus=testfailed%2Cbusted%2Cexception%2Cretry%2Cusercancel%2Crunnable&revision=7f25bf6ea553b7dca734a80947fbfe8f0053f265&selectedJob=247814203

Regressed by bug 1500474? The webcompat-reporter in the chrome url is from https://searchfox.org/mozilla-beta/rev/5b3f234f798d9ef8878fd33123036be6ad468ac3/browser/extensions/report-site-issue/locales/jar.mn#7

Failure log: https://treeherder.mozilla.org/logviewer.html#?job_id=247814203&repo=try

[task 2019-05-22T17:11:09.445Z] 17:11:09 INFO - TEST-START | browser/base/content/test/static/browser_all_files_referenced.js
[task 2019-05-22T17:11:11.101Z] 17:11:11 INFO - TEST-INFO | started process screentopng
[task 2019-05-22T17:11:11.744Z] 17:11:11 INFO - TEST-INFO | screentopng: exit 0
[task 2019-05-22T17:11:11.746Z] 17:11:11 INFO - Buffered messages logged at 17:11:09
[task 2019-05-22T17:11:11.746Z] 17:11:11 INFO - Entering test bound checkAllTheFiles
[task 2019-05-22T17:11:11.746Z] 17:11:11 INFO - Buffered messages finished
[task 2019-05-22T17:11:11.747Z] 17:11:11 INFO - TEST-UNEXPECTED-FAIL | browser/base/content/test/static/browser_all_files_referenced.js | Uncaught exception - [Exception... "Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIChromeRegistry.convertChromeURL]" nsresult: "0x80520012 (NS_ERROR_FILE_NOT_FOUND)" location: "JS frame :: chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js :: getBaseUriForChromeUri :: line 260" data: no]
[task 2019-05-22T17:11:11.748Z] 17:11:11 INFO - Stack trace:
[task 2019-05-22T17:11:11.748Z] 17:11:11 INFO - getBaseUriForChromeUri@chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js:260:28
[task 2019-05-22T17:11:11.748Z] 17:11:11 INFO - trackChromeUri@chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js:265:18
[task 2019-05-22T17:11:11.748Z] 17:11:11 INFO - parseManifest/<@chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js:289:11
[task 2019-05-22T17:11:11.749Z] 17:11:11 INFO - Leaving test bound checkAllTheFiles
[task 2019-05-22T17:11:11.750Z] 17:11:11 INFO - Console message: No chrome package registered for chrome://webcompat-reporter/locale/gobbledygooknonexistentfile.reallynothere
[task 2019-05-22T17:11:11.750Z] 17:11:11 INFO - GECKO(2841) | JavaScript error: chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js, line 260: NS_ERROR_FILE_NOT_FOUND: Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIChromeRegistry.convertChromeURL]
[task 2019-05-22T17:11:11.751Z] 17:11:11 INFO - Console message: [JavaScript Error: "NS_ERROR_FILE_NOT_FOUND: Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIChromeRegistry.convertChromeURL]" {file: "chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js" line: 260}]
[task 2019-05-22T17:11:11.753Z] 17:11:11 INFO - Not taking screenshot here: see the one that was previously logged
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - TEST-UNEXPECTED-FAIL | browser/base/content/test/static/browser_all_files_referenced.js | A promise chain failed to handle a rejection: Component returned failure code: 0x80520012 (NS_ERROR_FILE_NOT_FOUND) [nsIChromeRegistry.convertChromeURL] - stack: getBaseUriForChromeUri@chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js:260:28
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - trackChromeUri@chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js:265:18
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - parseManifest/<@chrome://mochitests/content/browser/browser/base/content/test/static/browser_all_files_referenced.js:289:11
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - Rejection date: Wed May 22 2019 17:11:11 GMT+0000 (Coordinated Universal Time) - false == true - JS frame :: resource://testing-common/PromiseTestUtils.jsm :: assertNoUncaughtRejections :: line 257
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - Stack trace:
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - resource://testing-common/PromiseTestUtils.jsm:assertNoUncaughtRejections:257
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - chrome://mochikit/content/browser-test.js:nextTest:755
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - chrome://mochikit/content/browser-test.js:testScope/test_finish/<:1417
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - chrome://mochikit/content/browser-test.js:run:1354
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - GECKO(2841) | MEMORY STAT vsizeMaxContiguous not supported in this build configuration.
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - GECKO(2841) | MEMORY STAT | vsize 2883MB | residentFast 398MB | heapAllocated 220MB
[task 2019-05-22T17:11:11.758Z] 17:11:11 INFO - TEST-OK | browser/base/content/test/static/browser_all_files_referenced.js | took 1675ms

Similar for devtools: https://treeherder.mozilla.org/logviewer.html#?job_id=247814141&repo=try

Comment 1

[Mike Taylor [:miketaylr]]

Tom, could you take care of this early next week?

Comment 2

[Thomas Wisniewski [:twisniewski]]

Pike, if I change the second webcompat-reporter string in that jar file to report-site-issue, will that mean that the l10n team will have to bulk-move localization files around from browser/extensions/webcompat-reporter/locales/** to browser/extensions/report-site-issue/locales/**? If so, let me know how we should coordinate that.

Also, should I also remove the first section below from browser/locales/l10n.toml after this change, or should it be kept for now for migration purposes (or anything like that)?

[[paths]]   # should I remove this section now?
    reference = "browser/extensions/webcompat-reporter/locales/en-US/**"
    l10n = "{l}browser/extensions/webcompat-reporter/**"
    
[[paths]]
    reference = "browser/extensions/report-site-issue/locales/en-US/**"
    l10n = "{l}browser/extensions/report-site-issue/**"

Comment 3

[Francesco Lodolo [:flod]]

Didn't we already move from webcompat-reporter from report-site-issue in bug 1500474? That jar.mn looks like something that it was missed in the original bug.

Bug 1524523 is there for the l10n.toml.

Comment 4

[Thomas Wisniewski [:twisniewski]]

Flod, yes it likely was just missed. If it's safe to make that change without l10n having to worry about it, then I'll make the change right away (without bothering with the l10n.toml; thanks for mentioning the related bug!)

Comment 5

[Francesco Lodolo [:flod]]

I would assume so, but that's a question for Axel (and also why I didn't clear the NI)

Comment 6

[Axel Hecht [:Pike]]

I think this fragment is the problem: https://hg.mozilla.org/releases/mozilla-beta/diff/ba4736a11504d19edc53805e38079349f4fb9744/browser/base/content/test/static/browser_all_files_referenced.js

There are two ways to fix this, either undo that fragment, and keep all chrome:// url references as webcompat-reporter, or keep that fragment, and change all references to the chrome url to be report-site-issue.

Neither should affect l10n, I think, as we only care about source-code locations, and not about chrome:// urls.

Comment 7

[Thomas Wisniewski [:twisniewski]]

I'll go with changing the references to report-site-issue, if it's all the same (presumably). But I'm having issues with my try-runs on Treeherder (my simple patch has a flurry of orange on it that seems unrelated, and subsequent try-runs aren't going anywhere), so I'll hold off on it until that clears up.

Comment 8

[Francesco Lodolo [:flod]]

I think that's due to Fission. You might want to check with sheriffs on how to filter/ignore that out.

Comment 10

[Thomas Wisniewski [:twisniewski]]

Yes, it was a recent Fission-related issue with tests. I just ran another try-run today, and the orange is more clearly limited to unrelated intermittents: https://treeherder.mozilla.org/#/jobs?repo=try&revision=0a6381e9ce28a24f2eea941ff0727a5986e304d5

Comment 11

[Thomas Wisniewski [:twisniewski]]

Attachment: Bug 1554264 - use the correct chrome URL for report-site-issue desktop l10n files; r?Pike

use the correct chrome URL for report-site-issue desktop l10n files

Comment 12

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/38dd768397b5
use the correct chrome URL for report-site-issue desktop l10n files; r=Pike

Comment 13

[Cristian Brindusan [:cbrindusan]]

https://hg.mozilla.org/mozilla-central/rev/38dd768397b5

Comment 14

[Thomas Wisniewski [:twisniewski]]

Aryx, I'm guessing we still need to test that this patch fixes the problem in another beta-sim build, and then uplift it to beta (assuming that's true)? If so, let me know and I'll reopen this bug.

Comment 15

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

Passes a beta-as-release Try push with the central patch applied: https://treeherder.mozilla.org/#/jobs?repo=try&revision=374d70661fff4b242bd260232ecf30c94260c7c4
Please request uplift.

Comment 16

[Thomas Wisniewski [:twisniewski]]

Comment on attachment 9069262 [details]
Bug 1554264 - use the correct chrome URL for report-site-issue desktop l10n files; r?Pike

Beta/Release Uplift Approval Request

• User impact if declined: None noted so far, but automated tests will otherwise permafail.
• Is this code covered by automated tests?: Yes
• Has the fix been verified in Nightly?: Yes
• Needs manual test from QE?: No
• If yes, steps to reproduce:
• List of other uplifts needed: None
• Risk to taking this patch: Low
• Why is the change risky/not risky? (and alternatives if risky): Low-risk since the patch only updates two path references in the codebase for a chrome URL used for l10n in the report-site-issue feature, which have only been seen to affect automated tests thus far.
• String changes made/needed:

Comment 17

[Julien Cristau [:jcristau]]

Comment on attachment 9069262 [details]
Bug 1554264 - use the correct chrome URL for report-site-issue desktop l10n files; r?Pike

followup for webcompat-reporter → report-site-issue rename, approved for 68.0b8

Comment 18

[Sebastian Hengst [:aryx] (needinfo me if it's about an intermittent or backout)]

https://hg.mozilla.org/releases/mozilla-beta/rev/d30ab2027694