Min RK

You can use the result database on the Hub for this (if you don't disable it and it doesn't fill up). It's not the most robust API, though. If you...

[docs are here][db-docs]. It's in-memory by default, and the default limit to start culling records is 1024 tasks or 1GB of results, whichever comes first. These can be configured with...

I would like to have this, but it's not planned, per se. A Client can retrieve the state of the queue with `Client.queue_status()`. A web view could be made from...

These decorators shouldn't be necessary if you use cloudpickle, because they should detect imported names correctly, unlike the default serialization. That doesn't address the concern of writing code that works...

I imagine this is mostly attributable to imports. I can run some import profiling to see if there are any that can be delayed to avoid unnecessary usage. I suspect...

I think the Hub should probably recognize the name and drop the message without a warning.

The connection/process diagram hasn't been updated to include the broadcast scheduler, which is itself multi-process for the tree, and accounts for $2^{depth+1}-1$ processes. The default is a depth of 1,...

```python # ipcluster_config.py (or cluster.config) c.EngineLauncher.engine_args=["--IPEngine.enable_nanny=False"] c.ControllerLauncher.controller_args = ["--usethreads"] ```

I don't have time to work on this right now, but if you wanted to have a stab, I'm happy to review.

I think it's hard to do this _in general_ such that it fits in the base class, but Launchers have two relevant methods: 1. `_log_output` which is called on stop....