Michael Delgado

This has been happening a lot to me lately when writing to zarr. Thanks to @bolliger32 for the tip - this usually works like a charm: ```python for v in...

ha - yeah that's a good flag. I definitely didn't intend for that to be a universally applied patch! so probably should have included a buyer beware. but we did...

I am getting the same error when downloading a large file through the node.js interface. ``` 761431 ms: Mark-sweep 1394.8 (1445.7) -> 1394.7 (1445.7) MB, 442.9 / 0 ms [allocation...

Thanks for your question and interest! Unfortunately, this looks like it's still tricky. zarr support for R is being worked on, but it looks like it's in it's early stages....

note - this is cross-posted on stack overflow (though I think it's a better fit for here or a discourse question): https://stackoverflow.com/questions/73254926/how-to-copy-whole-directory-folder-and-paste-in-a-different-location-in-jupyter?noredirect=1#comment129415457_73254926

@martindurant per-worker browser-based authentication is a tough route, especially when used in connection with dask-kubernetes or a pangeo-style setup. I've been experimenting with json-based authentication using some combination of `client.upload`...

if distributed + gcsfs were able to recognize the setup of the fuse system as a required worker dependency that would do the trick. is there some hook we could...

ok thanks