Hi everyone, we’ve been running an interactive tutorial on CryoSPARC for ~15 users, and noticed that the CryoSPARC consistently disconnects from the web GUI: we see 502 “bad gateway” via reverse proxy, and “webpage unaccessible” when connecting via ssh-tunnel by 39000 port.
How can we troubleshoot that, and avoid that in the future? I’m not even sure what’s causing it – whether some of the attendees have done something nasty, or it’s just too high load for the CryoSPARC database?..
oh, forgot to mention – we’re running v4.1.2, with cryosparc database folder on fast SSD. And during crashes we didn’t see any io_wait warnings from the netdata application.
What were the specs (RAM, CPU cores) of the CryoSPARC master host?
16 cpu cores (no multithreading), 256 Gb RAM.
No RAM/CPU overflows were detected as well.
Maybe the logs contain some clues. Please can you email us the file produced by the command
Hi @marinegor ,
Thank you for providing us your logs. Upon further investigation it’s unclear why the application seems to be crashing (no errors of note are logged). One action you can perform is modifying the CryoSPARC supervisor configuration file (
- Turn off CryoSPARC:
- Open the config file in an editor:
- Navigate to the following block:
- Save the file and start CryoSPARC:
Please let us know if this helps!