bug#47244: 28.0.50; SIGSEGV in long-runnning Emacs

[Michael Welsh Duggan]

Gregory Heytings <gregory@heytings.org> writes:

>>> I have no suggestion, but a question: why don't you try to bisect this 
>>> bug?
>>
>> Probably because this bug sits just around the corner - we are only not 
>> able to see it.  Bisecting is problematic for two reasons: It takes 
>> Michael some time to reproduce the bug and it's not yet clear whether 
>> the functionality of an older Emacs would allow him to reproduce that 
>> bug in the first place.
>>
>
> I see, but still, given the time you've already spent on this, would it 
> not be easier to check whether say HEAD~2000 already has the bug?  IIUC 
> the reproduction is always under the same condition: resetting Gnus after 
> switching the VPN.

Unfortunately, although it always triggers under the same condition, it
doesn't trigger every time I'm in that condition.  In fact, I don't know
that I've every been able to trigger it more than twice in a day, even
when manually attempting to create the circumstances that might do so.
And some days I haven't been able to trigger it at all.

-- 
Michael Welsh Duggan
(mwd@cert.org)