Tore Lervik

@benaadams Why have you not responded yet? It's been 2 hours already!

I have a hypothesis. Don't have much proof though.. We have quite a few apps (22) on the service plan. Two of them are heavy that each use 30% of...

I was thinking it had something to do with the ports too. But there are only a few of the 22 apps that use core. And I didn't find any...

@davidebbo When an app get stuck it never recovers until you stop and start the app service. The app service seems to think everything is running just fine.

@davidebbo @pan-wang I managed to reproduce it by making a core project that uses about 500MB of memory at startup, and then adding clones of the same project to the...

1. Added 5 app servivces with the same project that uses some memory at startup. 2. Started each app individually. All loaded up fine. 3. Marked all 5 app services,...

Further notes from testing. Each app holds on to the large chuck of memory for about 40 seconds, but GC should reduce it after that. The sum of the apps...

@davidebbo Hmm, maybe it's still port related. https://gist.github.com/hallatore/f3b021727e4e29ca05f67f1fe42b4fc7 Notice these lines. Seems it doesn't understand it failed to start properly. ``` Unable to bind to http://localhost:23649 on the IPv6 loopback...

I tried making the apps not use to much memory and I'm having a hard time making them freeze. On first try one froze, but the last 4 tries none...

@pan-wang I see a lot of `Failed to bind to address http://127.0.0.1:*: address already in use.` by using stdoutLogFile. But the apps seem to recover from this after a while....