Re: [ServiceWorker] Race when registering multiple service workers for the same scope (#783)

> The current behavior in the spec is the second job will replace the first job. That is in the example above, when the second update job gets in the synchronized steps, it'll replace the current installing worker in the first place by terminating the executing installing worker. Then, this update job will continue and advance to the Install phase if the conditions are all met. The terminated installing worker will get redundant in the Install step 16. I think I missed to pop the top element from the queue in this error case. Will double check it.

In my example above there is no installing worker yet.  The first install job has not been executed yet.  Its a race between running the install job or the second update job.

---
Reply to this email directly or view it on GitHub:
https://github.com/slightlyoff/ServiceWorker/issues/783#issuecomment-165129974