17 days ago
Hi Railway,
We're experiencing surprisingly slow checkpoint writes on our Postgres instance.
Here's an example log entry:
2025-07-16 21:04:51.476 UTC [27] LOG: checkpoint complete: wrote 52 buffers (0.3%); 0 WAL file(s) added, 0 removed, 0 recycled; write=5.029 s, sync=2.539 s, total=23.583 s; sync files=9, longest=2.210 s, average=0.283 s; distance=195 kB, estimate=195 kBIf I understand correctly, the times described in that line are orders of magnitude higher than what we should expect to see for this kind of checkpoint.
I know that tuning our postgres checkpoint configuration is one option here, but based on similar threads I've found on this forum I've gotten the sense that this is likely due to a Railway infrastructure issue. It also seems like moving this database from US West to US East could help since I've seen that suggestion in other threads:
- https://station.railway.com/questions/database-performance-issues-slow-stor-f6213a55
- https://station.railway.com/questions/assistance-requested-postgre-sql-sudden-04258f1a
Before I take that step and incur the downtime, I wanted to confirm whether this does look like the pattern observed in other threads and that switching to US East may help.
6 Replies
17 days ago
Hey there! We've found the following might help you get unblocked faster:
🧵 Database Performance Issues - Slow Storage I/O Causing Extended Checkpoint
🧵 n8n Workflow Stuck on “Loading” — Doesn’t Run After Deploy
If you find the answer from one of these, please let us know by solving the thread!
16 days ago
Hello!
We're acknowledging your issue and attaching a ticket to this thread.
We don't have an ETA for it, but, our engineering team will take a look and you will be updated as we update the ticket.
Please reply to this thread if you have any questions!
16 days ago
Hey there James,
This is unfortunately a known issue on some hosts. I have made it so that you are able to deploy back into some GCP hosts that we have kept online for customers with business performance needs such as Arcol's.
We will give you updates as we see long term resolution implemented.
Status changed to Awaiting User Response Railway • 16 days ago
16 days ago
Thanks for the update, we'll try going back to the GCP hosts and let you know if we still see the issue.
Status changed to Awaiting Railway Response Railway • 16 days ago
16 days ago
The initial results from switching back from Railway Metal to the legacy GCP hosts are extremely positive - thanks for letting us get back on those.
Status changed to Solved jamesbvaughan • 16 days ago
5 days ago
🛠️ The ticket Performance issue with disk operations on metal has been marked as todo.