vm_fault: pager read error, pid 98103 (rrdtool)
-
Hey I've seen this post regarding the vm_fault with page error on this forums before however the language was in Ukraine.
I'm running pfSense 2.8.0 CE virtualized on xcp-ng. The VM is allocated 8Gb RAM with 200 GB disk space. I've rebooted the VM however I'm still getting:
Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool) Aug 24 08:28:01 kernel vm_fault: pager read error, pid 28551 (rrdtool)From what I've read this seems like a hardware error, however this is in a virtual environment. I don't see any error within the xcp-ng hypervisor host or other VM's running on the hypervisor. Am I best running a mem86 test on the host, see if there is failing RAM -- replace if needed -- and if not recreate pfsense VM?
-
So I ended up running mem86+ on the xcp-ng host. No RAM errors reported -- All 14 tests x 4. I'll probably just end up reinstalling the VM but it's really strange. There isn't too much out on the web on this type of error, and what's out there would suggest RAM module failure, however I'm not certain this is accurate within a VM, as clearly the host RAM is not the cause. I'm not getting a lot of feedback on this issue unfortunately.
-
There was another post in the forum that suggested clearing all the rrdtool files with the following shell command....
rm -rf /var/db/rrd/*
This seemed to work for me. I had the same issue after a power loss on the system.
No longer see these errors repeated in syslogs. My guess is one of the files in that directory may become corrupted.
-
@coletrain23 shoot I wish I would have known this as I just reinstalled pfsense in a new vm and uploaded the configuration file. I suppose if I have time I could boot the old VM and run these commands to see if it solves the problem. Thanks for heads up