524 out of nowhere

We are starting to see an increase in 524 errors from Cloudflare, saying that the origin is taking too long. Has anything changed on Railway’s side?

We’re checking on our side as well to see if it’s an issue on our end. We just found it odd that the request isn’t being registered by our observability tools, so it might be failing before that.

I also turned off Cloudflare to see if the issue is related to them.

Solved

15 Replies

found out that one of my redis databases are unreachable


(no markdown available for this content)


and a re-deploy doesn't take place, seems similar to: https://discord.com/channels/713503345364697088/1518697501547298997


also seeing that the metrics tab is totally empty but might be related to the current incident.

image.png

Attachments


was able to do a re-deploy (took awhile) but still same issue, I was able to connect to it via an external client but it takes forever to do any queries


15 days ago

curious which regions are timing out/524ing


us-east


15 days ago

(client regions, like where they are connecting from, oops)


redis clients are all in us-east too


let me see if I can restore a backup into another Redis service


was able to do that and yeah our outage went away


no idea on what happened to that Redis but it got doomed


ok only now I saw the email lol

image.png

Attachments


anyway we will implement a Redis Cluster, tbh should have done that a long time ago but busy with other stuff

https://discord.com/channels/713503345364697088/943649944072499210/1518716668140982292


15 days ago

Sorry for the interruptions!


Status changed to Solved passos 15 days ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...