rust/neon

mirror of https://github.com/neondatabase/neon.git synced 2025-12-22 21:59:59 +00:00

Files

Arseny Sher 0112a602e1 Add timeout on proxy -> compute connection establishment.

Otherwise we sit up to default tcp_syn_retries (about 2+ min) before gettings os
error 110 if compute has been migrated to another pod.

2023-04-27 09:50:52 +04:00

src

Add timeout on proxy -> compute connection establishment.

2023-04-27 09:50:52 +04:00

Cargo.toml

Make proxy shutdown when all connections are closed (#3764 )

2023-04-13 19:31:30 +03:00

README.md

Follow-up for neondatabase/neon#2448 (#2452 )

2022-09-15 15:21:22 +03:00

README.md

Proxy

Proxy binary accepts --auth-backend CLI option, which determines auth scheme and cluster routing method. Following backends are currently implemented:

console new SCRAM-based console API; uses SNI info to select the destination project (endpoint soon)
postgres uses postgres to select auth secrets of existing roles. Useful for local testing
link sends login link for all usernames

Using SNI-based routing on localhost

Now proxy determines project name from the subdomain, request to the round-rice-566201.somedomain.tld will be routed to the project named round-rice-566201. Unfortunately, /etc/hosts does not support domain wildcards, so I usually use *.localtest.me which resolves to 127.0.0.1. Now we can create self-signed certificate and play with proxy:

openssl req -new -x509 -days 365 -nodes -text -out server.crt -keyout server.key -subj "/CN=*.localtest.me"

start proxy

./target/debug/proxy -c server.crt -k server.key

and connect to it

PGSSLROOTCERT=./server.crt psql 'postgres://my-cluster-42.localtest.me:1234?sslmode=verify-full'