Automated (million monkey?) tests
Michael and Robinson discussed this quite a bit at the Boston Hackfest and both seemed to like the idea generally. They'll have to work out the details but it's basically just a machine randomly doing everything and collecting crasher reports I believe. I believe this will be quite time consuming
#1 Updated by Bjoern Michaelsen over 3 years ago
IMHO (and with the experience of stacktraces from https://errors.ubuntu.com) those have a very long tail. That is, while this might find issues, it likely will still find a lot of issues that are not happening in real life -- OTOH, we currently have > 800 possible crashers that not only happened to real people, but likely also happen repeatedly because most don report a bug on first encounter.
#2 Updated by Bjoern Michaelsen over 3 years ago
- Project changed from 49 to Infrastructure
moving to infrastructure project as per https://redmine.documentfoundation.org/issues/599#note-2
#4 Updated by Florian Effenberger over 3 years ago
IMHO that is fixed by the crash testing hardware that was ordered - can you confirm?
(Hardware not delivered yet.)
#5 Updated by Florian Effenberger about 3 years ago
Feedback welcome, so we can put it on the list for FOSDEM if needed :-)