issue after moving app from aws to digitalocean

Question

i recently moved my app from aws to digitalocean to reduce cost and make things simple, but after shifting i am facing some weird issues. app was working fine before but now some api calls are timing out randomly and webhooks from my payment provider (https://dodopayments.com/) sometimes not coming or delayed.

also sometimes outbound requests failing without proper error and ssl also feels bit slow or handshake delay kind of thing. using nginx + node + pm2 same as before, nothing much changed in code.

i already checked firewall, ports, basic configs everything looks okay but still these issues happening. not sure if its network issue or region or something missing from aws side setup.

anyone faced similar thing or know what to check?

alexdo · Answer

Heya, @3775ce71211c40508f8e766a810eab First thing I’d check is clock sync - SSL handshake weirdness and random timeouts are often just NTP being off on a fresh Droplet. Quick timedatectl will tell you. DNS is another one - AWS has really snappy internal resolvers and a bare Droplet doesn’t, so if your app is doing a lot of outbound DNS lookups it can feel slow in weird places. dig dodopayments.com a few times and see if resolution time looks off. For the webhooks specifically - dodopayments might be getting blocked somewhere. Check your DO cloud firewall and any ufw rules, sometimes after a migration the inbound rules are tighter than you realize and webhook requests just get silently dropped. If they publish source IPs for their webhooks I’d make sure those are explicitly allowed. Also worth checking what region you’re on vs where you were on AWS. If you moved regions your latency to dodopayments’ infrastructure might just be higher now which can cause timeouts and delays without anything actually being “broken”. Nginx proxy timeouts are worth a look too if none of the above pans out - proxy_read_timeout and proxy_connect_timeoutsometimes need adjusting when you move infra. Also which region did you land on? Regards

Bobby · Answer

Hi there, Good coverage from alexdo already. A few more things worth checking: On AWS you might have been running behind a load balancer or NAT gateway that handled connection pooling differently. On a bare Droplet, outbound connections go straight out. If your Node app is not reusing HTTP connections (missing keep-alive agent config), you will see random timeouts and slow SSL handshakes just from connection overhead. Check if dodopayments retries on timeout. If your Droplet is slow to respond within their timeout window, they may be retrying with a delay, which looks like “missing” webhooks. Add some request logging in nginx to see if the requests are actually arriving but just slow to process. Also quick things to run: curl -w "%{time_namelookup} %{time_connect} %{time_starttransfer}
" -o /dev/null -s https://dodopayments.com This will tell you fast if the issue is DNS, TCP connect, or SSL specifically. What region are you on?

Become a contributor for community

DigitalOcean Documentation

Resources for startups and AI-native businesses

The developer cloud

Start building today

Report this

issue after moving app from aws to digitalocean

Related questions

Partner billing setup for client accounts — how to ensure clean detachment later?

New account with GitHub Student Pack credits cannot create Droplets or use Launchpad