Simon Skog

> > Can that single goroutine that drains the buffer be a bottleneck? Just setting a low interval e.g 1s or even sub-second i dont feel is enough to drain...

i compiled it to test the metric part only while i am waiting to get access to our staging cortex to test num-writers. Metric i get is ``` gnmic_prometheus_write_output_number_of_prometheus_write_msgs_sent_fail_total{reason="status_code=400"} 1234

been trying this pull with multiple remote writers now towards our staging env. I still see out of order samples but not as many as before. Will change this cortex...

I would be interested to know also and to share our setup and learn how you might have done. Our most busiest collector today has 900 devices loaded and 17k...

We would love this feature also.

I would also like to have this implemented.

RPC command: get-fpc-information XML output: ``` 0 Empty 1 Empty 2 Empty 3 Online 41 15 0 20 18 17 3584 8 25 4 Online 44 15 0 14 15...

Would really like to have netconf support added into the exporter.

Anyone?

What i thought then. I have during the day tried to update the code already on a local version to get signing working. Already looked into that [Commit.create_with_signature](https://www.rubydoc.info/gems/rugged/1.1.0/Rugged%2FCommit.create_with_signature) Thanks for...