We are using Tornado for our system (behind nginx) -- and everything is working fine.

Now we need to integrate a client lib (for Neo4j Graph DB) into our system. The problem is that it is blocking.

My question is:

  • is it OK to use blocking libs in non-blocking environment
  • what are the possible ramifications if we do so
  • how much work is involved if we decide to roll our own async lib

I know there is Node that does everything in non-blocking fashion. But we can't go there because our core is written in Python.

没有正确的解决方案

许可以下: CC-BY-SA归因
scroll top