OctoPrint has become extremely slow to respond

What is the problem?

OctoPrint web interface has become incredibly slow to load. Using Raspberry Pi 3b. Problem only started recently after a few months without using the printer.

Only command line on rpi, and unsure how best to monitor. After restart, cpu utilisation is high for about 30 seconds and UI responds. If I attempt to refresh the page (as often prompted to reload), the UI disappears for minutes and top shows minimum utilization (<3% for any process). By example, restarted OctoPrint at UI, UI responded within 30 seconds, then on browser refresh UI went white for over 4 minutes until OctoPrint UI returned.

What did you already try to solve it?

Restarted in safemode, same problem

I kept a lot of gcode files on the rpi, deleted them in case it was causing the slowdown. No obvious effect.

Have you tried running in safe mode?

yes

Did running in safe mode solve the problem?

no

Systeminfo Bundle

You can download this in OctoPrint's System Information dialog ... no bundle, no support!)
octoprint-systeminfo-20250814235938.zip (397.0 KB)

Additional information about your setup

OctoPrint version, OctoPi version, printer, firmware, browser, operating system, ... as much data as possible
OctoPrint 1.11.2
Raspbery Pi 3b 1.2
Prusa i3 MK3S+
Google Chrome 139, Windows 11

Have you already cleared the browser cache (Crtl-F5) from time to time?

Yes, that's my standard go-to when I have a browser glitch (along with restarting PC)

Fault-finding has indicated a network problem (wow), why did it only effect OctoPrint?).

I'm still investigating, as OctoPrint's general response is sluggish, but the UI is now consistently loading in under a minute (with or without safemode).

Will post back here if I need more help.

1 Like

Can confirm fault was entirely with network. Symptoms were very slow UI load. In fact, I believe connectivity between the browser and OctoPrint server were being disrupted, and only OctoPrint's robust retry methods of reconnecting from the browser were finally getting a UI load. This type of fault would be pretty uncommon amongst modern network equipment - there were high rates of packet loss and latency involved.