My rails app is deployed to SIN, it connects to an RDS, also in Singapore. I have recently experienced intermittent network issues between the app and the database. I checked all metrics, everything looks fine, database memory and CPU usage is consistently low, app memory and CPU are also low. I ssh-ed to the app and try performing database query, everything is fine as well. There was an instance where most of the requests timed out, and I had to redeploy the app itself to recycle the instance, then it resolved but still the intermittent timeout error to database.
Hi @sle , did you find the cause or ways to figure the cause?
I am experiencing the same issue with the same setup (app in SIN region and RDS in Singapore). So far. For unknown reason the connection from the app to RDS timeout intermittently.
I’ve never been able to figure it out. I remember changing the region to hkg and back, and that solved the problem. I guess it was because of the underlying instance having network issues, and moving the app to another region and back allocated the app to a completely different machine. A deployment didn’t help no matter how many times I deployed, and that was the only way. I remember I faced this problem not only once but twice or more and that was my workaround.
It’s a pity I never get a proper solution from fly tho.
so would scaling the app to 0 then scaling back up to 2 in a different region then scaling back to 0 then scaling back up to 2 in original region (sin) solve the problem?
We’re aware of the ongoing network congestion issues in sin, and are still looking into it with the upstream providers. Thanks for the reports - they help!