Large number of calendar events just disappeared from file:// calendar
Categories
(Calendar :: Provider: ICS/WebDAV, defect, P1)
Tracking
(Not tracked)
People
(Reporter: fouuqibwtn, Unassigned)
References
(Blocks 1 open bug)
Details
(Keywords: dataloss)
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:103.0) Gecko/20100101 Firefox/103.0
Steps to reproduce:
Nothing that I know of.
Actual results:
A large number of different events just disappeared from one of my file-based calendars. No possible way I could've accidentally deleted this many different events accidentally.
Expected results:
No events should be deleted without use action.
I'm rather amazed at the number of bugs that have shown up in calendar in the updates in the last month or so. I don't know if there were any serious calendar-related security fixes, but if not, I think it would be a big improvement to completely revert the calendar code.
|
||
Comment 1•3 years ago
|
||
Very sorry to hear this fouuqibwtn.
Please help us to make your report actionable.
- where does the ics file reside?
- have you checked inside the ics file (with text editor) if the missing events are still there?
- do missing events have anything in common or anything special about them?
- are some missing events repeated events (so that accidental deletion of the sequence could result in more events deleted?)
Tentatively marking P1/S2 because reporter claims dataloss.
|
|
||
Updated•3 years ago
|
|
Reporter | |
Comment 2•3 years ago
|
||
(In reply to Thomas D. (:thomas8) from comment #1)
Very sorry to hear this fouuqibwtn.
Please help us to make your report actionable.
- where does the ics file reside?
In a folder called "calendars" on my desktop.
- have you checked inside the ics file (with text editor) if the missing events are still there?
Yes. They are not still there. The .ics file is much smaller than it used to be.
- do missing events have anything in common or anything special about them?
The only thing I've noticed is that they're all events that I created to which I invited someone else. Most of my calendars contain events that I created to which only I'm invited. The calendar that lost the events was almost all events that I created and to which I invited someone else.
- are some missing events repeated events (so that accidental deletion of the sequence could result in more events deleted?)
I'm not 100% certain, because the events are lost, but I believe that none of the lost events were repeated events. In any case, there were many many events lost that were definitely not repeated events, far too many for me to have accidentally deleted all of them. Further, I essentially never delete events from the calendar in question.
Tentatively marking P1/S2 because reporter claims dataloss.
|
||
Comment 4•2 years ago
|
||
I'm not completely certain that these two bugs are the same, but they both involve data loss in local ICS files without clear repro, so good enough.
|
|
Reporter | |
Comment 5•2 years ago
|
||
The bug that you marked this as a duplicate of involves modifying an event in the calendar. This bug just happened with no triggering event. I don't think it's reasonable to mark it as a duplicate without further investigation.
|
|
||
Comment 6•2 years ago
|
||
Please do not revert bug status without discussion. There is a clear path to data loss on write in local ICS calendars, and the original report states that reproduction steps are uncertain. Without some indication that this is a separate bug, it makes sense to me to operate on the assumption that they are the same. There simply isn't enough information in this bug to investigate separately. If the problem recurs after the known data loss issues in the ICS calendar are fixed, a separate bug can still be filed.
|
|
Reporter | |
Comment 7•2 years ago
|
||
I don't know if you have investigated the code and found that there's some way that this data loss could happen with or without a triggering event. If so, I don't see it discussed above. Personally, I wouldn't mark a bug as a duplicate without such a code investigation if there were any difference at all in the observed behavior, which, in this case, there is. You may follow different procedures. I'm not going to fight about it.
|
||
Comment 8•2 years ago
|
||
The process we follow is when a bug X does not fix bug Y (this bug for example) then Y can be unduped. And in the interim, if additional information arises that make it clear there are two seperate causes, then certainly this bug would be undped and persued.
Description
•