Turbo write has nothing to do with RAM

13 minutes ago, johnnie.black said:

Turbo write has nothing to do with RAM

Ok- just thought that might have something to do with it since that is apparently why the machines are stopping.  It looked similar.

I did post in KVM also but not getting any feedback.  Any other ideas?

Not really, but VMs pause when they get out of space, not out of RAM.

Yeah, just an avenue I went down because after the machines paused- I tried to boot another VM and it was out of RAM and refused.  Required a reboot to free up enough to start any machines after that.  This is very frustrating..... really wish LT would chime in.  Guess I will email them.....

 

It still seems similar to the problem with caching that you were working on on that other thread...

Edited January 30, 20197 yr by jordanmw

Small hint. Check your power plans inside the VMs and set them to "high performance" and make sure they don't go to sleep. If that happens after some time idling they show as paused.

4 hours ago, bastl said:

Small hint. Check your power plans inside the VMs and set them to "high performance" and make sure they don't go to sleep. If that happens after some time idling they show as paused.

I wish it was that easy... the vms run fine until I copy files to the drives.

It is definitely an issue with my file copy from a network share eating up all the free RAM in unraid and pausing the vm.....

 

Somebody has to have seen this before?!? 

I had something like this originally on my system, I have somesort of issue with my motherboard that doesn't like something in the linux kernel...  A device kept throwing "Corrected" errors with the "Advanced Error Reporting" or AER...  This would after a while max out my log files before they could be rotated or cleared, and kill my VM's...  I noticed that your syslog includes:

Jan 27 11:32:59 Tower kernel: pcieport 0000:40:01.3: AER: Corrected error received: 0000:00:00.0
Jan 27 11:32:59 Tower kernel: pcieport 0000:40:01.3: PCIe Bus Error: severity=Corrected, type=Data Link Layer, (Receiver ID)
Jan 27 11:32:59 Tower kernel: pcieport 0000:40:01.3:   device [1022:1453] error status/mask=00000040/00006000
Jan 27 11:32:59 Tower kernel: pcieport 0000:40:01.3:    [ 6] Bad TLP               

Which on your system is the "PCIe GPP Bridge" devices throwing errors:

00:01.1 PCI bridge [0604]: Advanced Micro Devices, Inc. [AMD] Family 17h (Models 00h-0fh) PCIe GPP Bridge [1022:1453]
	Kernel driver in use: pcieport
00:01.2 PCI bridge [0604]: Advanced Micro Devices, Inc. [AMD] Family 17h (Models 00h-0fh) PCIe GPP Bridge [1022:1453]
	Kernel driver in use: pcieport
00:01.3 PCI bridge [0604]: Advanced Micro Devices, Inc. [AMD] Family 17h (Models 00h-0fh) PCIe GPP Bridge [1022:1453]
	Kernel driver in use: pcieport
.....

Which is just like the errors I was getting...

 

I had to change my Syslinux configuration to include: 

append pci=noaer initrd=/bzroot

This disables AER reporting, and killed the error...  I have been running that way for several months now with no issues...

 

Now for what I have been able to find on the explanation for this issue... Killing all AER is like fixing the engine light in your car by cutting the wires to the light...  Not the ideal solution...  I have attempted several times to figure this out to get it fixed, but since it doesn't seem to affect anything other than the log file, correcting it has not been a high priority for me...

 

This link is the best I have seen for an explanation to get this fixed...

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1521173

 

 

Edited January 31, 20197 yr by Warrentheo

Thanks Warrentheo, I give that a shot but I am not seeing logs fill anything up.  Worth a shot though.

I suspect that if it is not the issue, it is at least a symptom of the issue...

 

I noticed that on your system you have NVMe drives, which I also have... My current best guess for these errors are something in the linux kernel, and the BTFS drivers when introduced to NVMe drives... But that is guesswork from someone who barely knows what I am doing...