Hi everyone,

I'm testing the stress test on my C236-WSI4-85L these days.
The test is simple, I created 4 VMs which are 2 cores / 14GB memory each.
And running Unigine Valley Benchmark each VM.

And another machine which is E3V4, created same 4 VMs, only difference between E3V5 is that the VM spec is 2 cores / 6GB memory each.

(P.S Every VM are running Windows 7.)

After few days test, I found that the E3V5 host crashed after 60~70 hours, which crashing is "freeze".
Then I checked htop and dmesg watch monitor and found two things:

1. There is no error log when crash, only kernel trace about i915 driver when starting VM.
2. When testing, the memory usage is increasing.

For example, I got 64189MB total memory when boot into Ubuntu, and used about 593XX MB memory after 4 VMs started.
After about 36 hours, the memory was already used 613XX MB memory.
Then After 48 hours, the memory was already used 624XX MB memory, and started using swap.

When host crashed, the htop shows that already used 634XX MB memory, and used about 2GB swap size.

I guess that memory overloaded cause the "freeze" crash.
But it seems like memory usage is not increase when testing on E3V4 platform.
And I'm sure that I didn't run any other program. Then the Ubuntu is running with server edition, which means didn't have any GUI.

Besides, I'm using kernel command line below:

BOOT_IMAGE=/vmlinuz-4.3.0 root=/dev/mapper/os-root ro console=tty0 net.ifnames=0 biosdevname=0 rootdelay=90 root=UUID=5d69db35-c6cb-4990-947f-9b562ea8908b nomdmonddf nomdmonisw intel_iommu=igfx_off i915.hvm_boot_foreground=0 i915.propagate_monitor_to_guest=0 i915.preemption_policy=3 i915.enable_cmd_parser=0 loglvl=all guest_loglvl=all conring_size=4M noreboot


Any suggestions?

Thanks,
Eddie.