User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:100.0) Gecko/20100101 Firefox/100.0

Steps to reproduce:

I followed the steps for "Profiling Thunderbird Performance" from here:
https://support.mozilla.org/en-US/kb/profiling-thunderbird-performance

Actual results:

When trying to finalize/stop the recording, I am not given a Stop Recording Performance button, but rather a Capture recording button which results in the following error in the console:

Failed to get the active browserId while starting the profiler. 3 recording-utils.js:53:11
    getActiveBrowserID resource://devtools/shared/performance-new/recording-utils.js:53
    startProfiler resource://devtools/shared/performance-new/gecko-profiler-interface.js:70
    actorBridgeWithSpec resource://devtools/server/actors/common.js:107
    handler resource://devtools/shared/protocol/Actor.js:175
    onPacket resource://devtools/server/devtools-server-connection.js:379
    _onJSONObjectReady resource://devtools/shared/transport/transport.js:464
    makeInfallible resource://devtools/shared/ThreadSafeDevToolsUtils.js:103
    makeInfallible resource://devtools/shared/ThreadSafeDevToolsUtils.js:103

Expected results:

The recording should be saved

On Mac using beta 102.0b3 I see it when attempting to start a profile

Failed to get the active browserId while starting the profiler. recording-utils.js:53:11
getActiveBrowserID resource://devtools/shared/performance-new/recording-utils.js:53
startProfiler resource://devtools/shared/performance-new/gecko-profiler-interface.js:70
actorBridgeWithSpec resource://devtools/server/actors/common.js:107
handler resource://devtools/shared/protocol/Actor.js:175
onPacket resource://devtools/server/devtools-server-connection.js:379
_onJSONObjectReady resource://devtools/shared/transport/transport.js:464
makeInfallible resource://devtools/shared/ThreadSafeDevToolsUtils.js:103
makeInfallible resource://devtools/shared/ThreadSafeDevToolsUtils.js:103

A regression introduced from m-c ?
Fixed profiler bugs in the past 3 months https://mzl.la/3NR4SEB

This needs immediate attention

Meichthys, can you help us by getting the regression range? https://mozilla.github.io/mozregression/

Gently redirecting to Geoff

I'll try to get the regression range, however i'm having trouble installing mozregression via pip pip install mozregression is requiring C++ build tools. I've tried installing a few different versions but still get an error when pip installing - it fails when trying to install the aiohttp dependency..

I'll keep trying tho!

Let me just say.. mozregression is VERY cool.
I did end up getting it installed. The issue was with my package manager (pipenv).
Here are the results:

 4:52.16 INFO: Last good revision: 542ef278cf7fe6e6cf52ec8cb4626a282a7741b5
 4:52.16 INFO: First bad revision: 7d1d815986ff74f82e0535cca9d4b95b0bbe66dd
 4:52.16 INFO: Pushlog:
https://hg.mozilla.org/comm-central/pushloghtml?fromchange=542ef278cf7fe6e6cf52ec8cb4626a282a7741b5&tochange=7d1d815986ff74f82e0535cca9d4b95b0bbe66dd

(In reply to Meichthys from comment #6)

Let me just say.. mozregression is VERY cool.
I did end up getting it installed. The issue was with my package manager (pipenv).
Here are the results:

 4:52.16 INFO: Last good revision: 542ef278cf7fe6e6cf52ec8cb4626a282a7741b5
 4:52.16 INFO: First bad revision: 7d1d815986ff74f82e0535cca9d4b95b0bbe66dd
 4:52.16 INFO: Pushlog:
https://hg.mozilla.org/comm-central/pushloghtml?fromchange=542ef278cf7fe6e6cf52ec8cb4626a282a7741b5&tochange=7d1d815986ff74f82e0535cca9d4b95b0bbe66dd

Very cool indeed.
Alex, could it be caused by Bug 1765650 - Thunderbird build busted after extProtService->LoadURI method changed

I have no idea as I simply did a very simple m-c port on that bug.
I know nothing of how the profiler and this whole section works.

More likely bug 1764505, which is in the matching mozilla-central range. I haven't had a chance to look closely at it yet.

Interesting. I don't understand why bug 1764505 is not marked as a regression.

Will need someone to look at this soonish

I'm already looking at it. It's complex.

Okay, I've got it going. Fortunately everything except the error message about browserId can be fixed without changing mozilla-central/esr102 code. And the error message doesn't need fixing, other than to make it go away.

There's a few things going on here:

• Added a Fluent file in a specific place so the profiler page loads properly.
• Made about:profiling work, which is needed for the profiler settings.
• Moved the link opening functions to utilityOverlay.js so they get loaded in the toolbox process, and change them so a new window can be opened if necessary.
• Removed the old profiler's preferences.

Pushed by geoff@darktrojan.net:
https://hg.mozilla.org/comm-central/rev/67274225257d
Fix the new developer tools profiler. r=john.bieling

Comment on attachment 9284308 [details]
Bug 1771304 - Fix the new developer tools profiler. r=john.bieling

[Approval Request Comment]
Regression caused by (bug #):
User impact if declined: Profile doesn't work
Testing completed (on c-c, etc.): 3 days
Risk to taking this patch (and alternatives if risky): Potential for something unexpected, unlikely

Comment on attachment 9284308 [details]
Bug 1771304 - Fix the new developer tools profiler. r=john.bieling

[Triage Comment]
Approved for beta

Thunderbird 103.0b6:
https://hg.mozilla.org/releases/comm-beta/rev/1c88c5cb7b8c

Comment on attachment 9284308 [details]
Bug 1771304 - Fix the new developer tools profiler. r=john.bieling

[Triage Comment]
Approved for esr102
Risk, see comment 16

Thunderbird 102.1.0:
https://hg.mozilla.org/releases/comm-esr102/rev/e8ca45744512

FYI - I'm still seeing "Failed to get the active browserId while starting the profiler." in the console logs when running a profile, but the profile seems to work fine.