1438452 - Thunderbird hangs after accepting invitation - Daily 60.0a1 (2018-02-13) (64-bit)

Status: RESOLVED FIXED

(Calendar :: E-mail based Scheduling (iTIP/iMIP), defect)

Description

[Jorg K (CEST = GMT+2)]

Thunderbird hangs on the "Would you like to send a confirmation e-mail" after accepting invitation regardless of whether you click Yes or No.

I get an "unresponsive script" warning after a while, but cancelling the script doesn't help. CPU is at 25%, looks like one of the four cores runs at 100%, so some endless loop.

Most likely due to recent refactoring in Calendar.

Comment 1

[Jorg K (CEST = GMT+2)]

OK, I tried the build just before the big refactoring in bug 1436557, bug 1436491, bug 1436490, bug 1436488 and bug 1436488 and that already hangs, C-C rev bdfa5f49559e, so the regression is earlier. Alice, can you find it for us, please.

Comment 2

[Jorg K (CEST = GMT+2)]

I tried this in a debug build and I get
JavaScript error: chrome://calendar/content/calendar-unifinder.js, line 449: TypeError: this.tree is null

and later:
JavaScript warning: chrome://lightning/content/imip-bar.js, line 132: Error: Script terminated by timeout at:
getButtons@chrome://lightning/content/imip-bar.js:132:16
resetButtons@chrome://lightning/content/imip-bar.js:121:23
setupOptions@chrome://lightning/content/imip-bar.js:238:9
processFoundItems@resource://calendar/modules/calItipUtils.jsm:1676:9
onModifyItem@resource://calendar/modules/calItipUtils.jsm:1342:13
onAddItem@resource://calendar/modules/calItipUtils.jsm:1348:9
notify@resource://calendar/modules/calDataUtils.jsm:55:32
notify@resource://calendar/modules/calDataUtils.jsm:79:16
adoptItem@file:///C:/mozilla-source/comm-central/obj-i686-pc-mingw32/dist/bin/extensions/%7Be2fda1a4-762b-4020-b5ad-a41df1933103%7D/components/calStorageCalendar.js:464:9
addItem@file:///C:/mozilla-source/comm-central/obj-i686-pc-mingw32/dist/bin/extensions/%7Be2fda1a4-762b-4020-b5ad-a41df1933103%7D/components/calStorageCalendar.js:412:16
action@resource://calendar/modules/calItipUtils.jsm:1642:36
actionFunc@resource://calendar/modules/calItipUtils.jsm:1667:25
_execAction@chrome://lightning/content/imip-bar.js:350:21
executeAction@chrome://lightning/content/imip-bar.js:463:20
oncommand@chrome://messenger/content/messenger.xul:1:33

Comment 3

[Alice0775 White]

(In reply to Jorg K (GMT+1) from comment #1)
> OK, I tried the build just before the big refactoring in bug 1436557, bug
> 1436491, bug 1436490, bug 1436488 and bug 1436488 and that already hangs,
> C-C rev bdfa5f49559e, so the regression is earlier. Alice, can you find it
> for us, please.

please give me STR

Comment 4

[Jorg K (CEST = GMT+2)]

Attachment: Event Invitation New Event.eml

I've sent you and invitation, or you can import this message. Accept the invitation and a panel "Do you want to send a confirmation" comes up. Click Yes or No. Then it hangs.

To repeat, delete the accepted invitation from the calendar, then accept again.

Comment 5

[Alice0775 White]

Regression window:

https://hg.mozilla.org/comm-central/pushloghtml?fromchange=b5e155228d89c235f2bfb013b32672ab35b0d519&tochange=8ebb03cbea92cf0f8d60745ec5b3b99140eacf07
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=cac3a4e6000d7079f8cf0118a5c509b2527b3289&tochange=ef1fefe4c6d1f95e2bdf872094e02e85c962aa86

Comment 6

[Jorg K (CEST = GMT+2)]

Thanks, Alice!!

I think bug 1434737 was mostly a mechanical exercise, so it's most likely bug 1433802 from the same push. What I don't understand is that there is no test for accepting an event, or if there is, it didn't catch the problem.

Comment 7

[[:MakeMyDay]]

Quite likely bug 1433802 as already mentioned. Philipp, if you cannot have a look at it short-termly, we should consider to back out the patch from that bug.

Comment 8

[Jorg K (CEST = GMT+2)]

In that case please prepare a backout patch. Just backing it out gives a warning on calDataUtils.jsm (deleted by other) a few merge conflicts. No surprise given all the refactoring that happened later, see comment #1.

Comment 9

[Philipp Kewisch [:Fallen][📱📆]]

I'm hoping to fix this today, thanks for the investigation.

Comment 10

[Philipp Kewisch [:Fallen][📱📆]]

This might go further back. I can still reproduce this with the treeherder build of bdfa5f49559e which was before the mentioned push. Trying to find out where this starts happening.

Comment 11

[Philipp Kewisch [:Fallen][📱📆]]

Ignore the previous comment, I got the order messed up. Trying to figure out what is going on here.

Comment 12

[Philipp Kewisch [:Fallen][📱📆]]

I'm fixing this in bug 1433802. In hindsight it would have made more sense to use this bug, but I had started with the unit test and a minor fix, and had thought it might be nicer to keep the unit test with the fix.

Comment 13

[Jorg K (CEST = GMT+2)]

No worries at all. When the other bug lands, we say "Fixed by bug 1433802" and set the target milestone accordingly :-)

Comment 14

[Martin Schröder [:mschroeder]]

Philipp, can this bug report be marked fixed after the patch in bug 1433802 landed?

Comment 15

[Jorg K (CEST = GMT+2)]

Fixed by bug 1433802. Sorry I forgot.