Question

Running out of disk space due to large files associated with postgress


Badge

We have a FME server installation which has been active for a couple of years now without issues. Today I got a message from our sysadmin that the machine where FME server is installed is running out of disk space.

It turns out that in the Postgress database that is installed with FME server large files were created. I can't find any logging that would explain how and why these files were created. Everything appears to be running smoothly.

 

imageDoes anyone know a possible cause for this and how to remove these files in a responsible manner?

 

Edit: So, I have been able to connect to the Postgress database using PgAdmin4, installed with fmeserver. It appears that since some time every second a couple of error messages are written in the table fme_sysevent_history. That table is now 11 Gb. So that mystery is solved 🙂

(apparently it has something to do with https://community.safe.com/s/article/FME-Server-Engine-ERROR-Missing-or-expired-workspace-chaining-context )

 

So I'd like to truncate this table. Or are there any expected adverse effects why that is not advisable?


2 replies

Userlevel 5
Badge +29

I think you should be fine clearing out the system event history.

Badge

Adding to @hkingsbury​'s response, it should be okay to truncate the table. As an alternative, in System Configuration > System Cleanup > Scheduled Cleanups > System Event History, you can set an expiry time for these records so they're deleted. Note that this cleanup will only occur on the interval set in System Cleanup > Configuration. Also, if you haven't had a chance to resolve the Engine chaining context issue, I'd recommend submitting a case with us so we can investigate further (there is a related bug ticket for this FMESERVER-16178).

Reply