Functions Stops and needs a restart: Host thresholds exceeded [Connections]

Hi,

I am facing an issue where randomly a function stops working. This happens frequently when the load increases. **The problem is that no errors occur** and after a restart/stopstart the function starts working normally.

The Azure function is:

1) Triggered using EventHub Trigger and reading in batches
2) Reads from Storage using a SINGLE CloudTableClient
3) Writes to SQL - A single connection is being opened when needed and disposed of immediately

Any idea what is causing this issue? I had a similar issue on Functions V1 in the past but decided to put it on hold till I migrate to V2. Another thing that confuses me is that even after all the load goes down or stops, the function does not recover and stays in an idle(running) state without processing the EventHub queue.

The report names an Azure Function using an EventHub trigger, batched reads, a single CloudTableClient, and SQL writes, but no source files or tests. Start by isolating the trigger and connection behavior under increased load and after the load stops; a useful outcome would be a reproducible cause and a documented fix or diagnostic result.

Written by the indexing model from the issue text.