994092 - Breakpoints don't always trigger in AJAX requests

Status: RESOLVED DUPLICATE of bug 801304

(DevTools :: Debugger, defect, P2)

Description

[Jan Honza Odvarko [:Honza] (always need-info? me)]

Breakpoint within AJAX response handler doesn't work

Follow instructions about how to reproduce the problem:
https://getfirebug.com/tests/manual/issues/5770/issue5770.html

Originally reported as Firebug issue:
http://code.google.com/p/fbug/issues/detail?id=5770


Honza

Comment 1

[Nick Fitzgerald [:fitzgen] [⏰PST; UTC-8]]

This works for me in Nightly 31. I hit the breakpoint 6 times. What channel/release are you experiencing this on, Honza?

Comment 2

[Jan Honza Odvarko [:Honza] (always need-info? me)]

This is weird, I am also on 31 (latest Nightly build)

Honza

Comment 3

[Nick Fitzgerald [:fitzgen] [⏰PST; UTC-8]]

Interesting. Now when I do the STR it gets hit 3-4 times and isn't consistent... Perhaps this is a problem with the test case?

Comment 4

[Nick Fitzgerald [:fitzgen] [⏰PST; UTC-8]]

No I don't think that is it. When you don't set a breakpoint, you always get 5 console messages, so we should always hit the BP 5 times.

(The STR should expect 5 not 6 hits, IINM).

Comment 5

[Jan Honza Odvarko [:Honza] (always need-info? me)]

(In reply to Nick Fitzgerald [:fitzgen] from comment #4)
> No I don't think that is it. When you don't set a breakpoint, you always get
> 5 console messages, so we should always hit the BP 5 times.
I have updated the test.

Now there is a second button executing another method with debugger; keyword.
Does the debugger; keyword hits 5 times for you?

https://getfirebug.com/tests/manual/issues/5770/issue5770.html

> (The STR should expect 5 not 6 hits, IINM).
Yep, fixed.

Honza

Comment 6

[Jan Honza Odvarko [:Honza] (always need-info? me)]

(In reply to Nick Fitzgerald [:fitzgen] from comment #3)
> Interesting. Now when I do the STR it gets hit 3-4 times and isn't
> consistent... Perhaps this is a problem with the test case?
Ah, I missed this comment. So, looks like you agree that there is a bug...?

Honza

Comment 7

[Nick Fitzgerald [:fitzgen] [⏰PST; UTC-8]]

Yeah there's a bug.

Now I'm consistently hitting the BP 4 times, and the debugger statement 6 times.

Comment 8

[Jeff Griffiths (:canuckistani) (:⚡︎)]

Re-testing in nightly 40 with e10s enabled, the behaviour is now very consistent - the handler only ever breaks once, it seems to be a random request as opposed to (for example) always the first request, which is probably to be expected ( I made an altered version of the demo code that binds i into the callback )

Comment 9

[James Long (:jlongster)]

This sounds suspiciously like bug 977972, which might be fixed by fixing bug 801304. I think eddy's going to look at the latter bug and we'll see if it fixes these (unfortunately not by devedition-40 though)

Comment 10

[Patrick Brosset <:pbro>]

Still very much an issue with the new debugger. This is most likely a server issue, so we should keep this bug on bugzilla.
However, keeping it as a P1 doesn't mean anything. We use P1s for crashers, data-loss, security bugs, etc. Things we should really be fixing now and not wait.
Now, I really would like us to get to these types of server debugger bugs now, but realistically, this is a P2.
I'll keep track of it and make sure we get to it in the near future.

Comment 11

[Jason Laster [:jlast]]

I think this is a duplicate of 801304