Linux/Python: Uncomplete invocation due to container restart from accidental recycling state instead of drain mode

Per Microsoft support's suggestion, I'm added an issue on the repo so my team can better track when this issue, confirmed by the azure product group, is resolved. TrackingID#2110070040006605. There is already an internal work item called "Increase max recycling container count default" on Linux consumption instances.

I'll repeat what I've learned from support as best I can. The backend process wherein recycled containers are put into drain mode which should ensure that the worker doesn’t handle any new invocations is being recycled instead of put into drain mode. This causes our azure function invocation to suddenly stop without error or issue - it simply restarts due to the underlying container.

I believe the last feedback I've heard on this is that the plan is to increase the number of recycling containers to process concurrently.

Nell'issue non sono identificati file del repository, test o punti di ingresso. Inizia esaminando il riciclo dei container Linux consumption e il comportamento del drain-mode di Azure Functions, quindi usa il TrackingID riportato e il contesto del work item interno per determinare se il riavvio è stato risolto e se il riciclo concorrente si comporta come previsto.

Scritto dal modello di indicizzazione a partire dal testo della issue.