Thanks! Looking again at the pernosco session, the case there is similar but not the same. We arrive way before any shutdown phase started inside [`PermissionManager::Init`](https://searchfox.org/mozilla-central/rev/1749ba14c2ebe200855802c478780122135727f2/extensions/permissions/PermissionManager.cpp#638) but when we try to add the shutdown blocker, the `nsIAsyncShutdownClient` interface is not available and thus we do not add any resulting in a similar situation as you described. My take would be: If we cannot guarantee a proper shutdown, probably we should not even arrive at [`InitDB`](https://searchfox.org/mozilla-central/rev/1749ba14c2ebe200855802c478780122135727f2/extensions/permissions/PermissionManager.cpp#704) ? Or there might be process types / test harnesses that break ?
Bug 1707963 Comment 23 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
Thanks! Looking again at the pernosco session, the case there is similar but not the same. We arrive way before any shutdown phase started inside [`PermissionManager::Init`](https://searchfox.org/mozilla-central/rev/1749ba14c2ebe200855802c478780122135727f2/extensions/permissions/PermissionManager.cpp#638) but when we try to add the shutdown blocker, the `nsIAsyncShutdownClient` interface is not available and thus we do not add any blocker, resulting in a similar situation as you described. My take would be: If we cannot guarantee a proper shutdown, probably we should not even arrive at [`InitDB`](https://searchfox.org/mozilla-central/rev/1749ba14c2ebe200855802c478780122135727f2/extensions/permissions/PermissionManager.cpp#704) ? Or there might be process types / test harnesses that break ?