jcarroll-mediafly

Quick update. Testing with device over wifi doesn't seem to have this issue. Appears to be simulator + wifi combination. hm...

No, that seems the be when it dies off. That callback is never called. I believe the last set of calls I see in the console is about configuring /...

I haven't completely tested my theory, but if I choose to not use the LOGGER_DEFAULT_OPTIONS but rather use all those options except `kLoggerOption_BrowsePeerToPeer`, then simulator + nslogger + wifi only...

Or add a lock around the original lines in `cancel()` and the `finish(aggregatedErrors)` in the queue callback.

Yeah, I struggled with this a bit myself. Especially when using group operations and dealing with race conditions. If you haven't seen my reported issue I would like to hear...

Change was small enough I went ahead with it. I started a discussion below where we talk about merits of the proper solution. https://github.com/pluralsight/PSOperations/issues/69

Would like to get other peoples thoughts on this. Is no one else calling cancel on a group operation only to find out in the completion the cancel status is...

Correct it all comes down to the state not being thread safe. In cancel we cancel all the operations, mark the queue suspended as false and finally call super. The...

Yes, by calling super.cancel() the group operation may finish before all the individual operations finish correct. But before my changes this was always a risk item, either all the ops...

I did notice the change on the statelock, without knowing the history I'm unsure of the specific reason it was expanded upon. Might want to start a new discussion in...