Michael J Gruber <git@grubix.eu> writes:
>> 
>> It's hard to see what can go wrong (maybe perl doesn't work?), but
>> failing to generate a message should be a fatal error. Maybe try
>> something like

I was unclear. I meant "should" as in "in a perfect world", not as a
prediction about the current code. So we should probably try to add some
error checking either where the patch put it, or in generate_message
itself.

>> 
>
> Well, as you can see in both reports, the pattern is as follows:
>
> The first subttest sees 1 less message than expected.
> The second sees 1 more message than expected.
>
> So in summary, they are generated but "notmuch new" is not run or does
> not pick them up. Though it's a Heisenbug on a specific arch, I'll try
> your suggestion and hammer our infra with it ...
>

Yeah, debugging by batch submission is no fun, no matter what our
grandparents tell us.  I do have interactive access to s390x
systems running Debian and Ubuntu, but the problems you report do not
occur in Debian (probably not in Ubuntu either, or I would have heard
about it, but I didn't check if e.g. Ubuntu disables the tests).

There is not (supposed to be) anything asynchronous about the process of
generating a message and indexing it. You could also try adding
--full-scan to the relevant "notmuch new" invocation. This will disable
the heuristics based on directory mtimes that notmuch uses to speed up
indexing.

d

_______________________________________________
notmuch mailing list -- notmuch@notmuchmail.org
To unsubscribe send an email to notmuch-leave@notmuchmail.org