You recently removed servers in WAW and moved them to ARN. It’s fine but now, when I try to do manual pg_dump it takes minutes for 300 MB DB which previously took seconds.
What I’m doing:
fly proxy 15432:5432 -a raczekteam-db
docker run -it --rm -e PGPASSWORD=<PASSWORD> postgres:latest pg_dump -h host.docker.internal -p 15432 -U <USER_NAME> -d <DATABASE_NAME> > raczekteam_$(date '+%Y-%m-%d_%H-%M-%S').sql
I’ve also tried different way but it’s always super slow. Can you help with that?
Hi!
Is it possible for you to try using Wireguard instead of fly proxy?
Let me know if running this through Wireguard has better performance.
Another thing you could try (and please report if it helps!) is creating a new wireguard peer to be used by fly proxy: fly wireguard reset should do it, then try fly proxy again and let me know if this helps in any way.
It didn’t help. Still it’s super slow with both WireGuard and reseting wg. It takes around 48 min to do pg_dump for 300 MB db.
It didn’t help. Still it’s super slow with both WireGuard and reseting wg. It takes around 48 min to do pg_dump for 300 MB db.
I was experiencing the same issues with flyproxy - i was using a gitlab runner to run backups hourly so the slow down was eating into my compute minute costs on gitlab. After talking with support it looks like there definitely something up with flyproxy right now because with a direct wireguard connection my pgdump takes seconds (vs 10-15 minutes on flyproxy). I had to jump through some hoops but got wireguard setup on an ec2 instance to run my gitlab runner jobs and now the entire backup takes seconds and is working. I would prefer if flyproxy worked as it is much less setup for me but just thought id chime in here that a direct wireguard connection did fix the slow downs for me.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.