CHR on ESXi - VM got stopped

RouterOS Virtualization
1

Folks,

I got an issue with CHR 6.37.5 (bugfix) with P1-Trial license running on VMware ESXi 5.1.0. After ~3 days CHR VM got stopped. I checked logs on ESXi node:

~
2017-06-11T04:58:37.208Z| vcpu-1| I120: CPU reset: soft (mode 2)
2017-06-11T04:58:37.220Z| vcpu-2| I120: CPU reset: soft (mode 2)
2017-06-11T04:58:37.231Z| vcpu-3| I120: CPU reset: soft (mode 2)
2017-06-11T04:58:37.824Z| vcpu-0| I120: SVGA: Unregistering IOSpace at 0x10d0
2017-06-11T04:58:37.824Z| vcpu-0| I120: SVGA: Unregistering MemSpace at 0xd8000000(0xd8000000) and 0xd0800000(0xd0800000)
2017-06-11T04:58:37.825Z| vcpu-0| I120: SVGA: Registering IOSpace at 0x10d0
2017-06-11T04:58:37.825Z| vcpu-0| I120: SVGA: Registering MemSpace at 0xd8000000(0xd8000000) and 0xd0800000(0xd0800000)
2017-06-11T04:58:37.826Z| vcpu-0| I120: SVGA: Unregistering IOSpace at 0x10d0
~

Determining why a virtual machine was powered off or restarted (1019064) says:

A Guest OS initiated reset generates this event in the virtual machine logs:

Mar 04 17:04:29.638: vcpu-0| CPU reset: soft

>
> To continue troubleshooting, contact the operating system vendor to determine > **why the guest operating system initiated the reset**> . If the operating system vendor suspects that the outage was caused by VMware Tools or the virtual hardware, contact VMware support.

I powered up VM and it started like nothing happened. Log says nothing:

```text
10:51:03 system,error,critical router was rebooted without proper shutdown
10:51:05 interface,info ether1 link up
13:23:25 system,info,account user alex logged in from 1.1.1.1 via ssh

Setup parameters:

  • CHR 6.37.5 (bugfix)
  • License: P1-Trial
  • disk0: Extended to 1GB.
  • disk1: 200GB for The Dude 6 server install.
  • VMware ESXi 5.1.0
  • One vNIC E1000 with static MAC.
  • 4vCPUs
  • 2GB RAM.

Questions:

  1. Is it known issue? If so, how to fix it?
  2. Based on ESXi logs CHR initiated the reset. Does it mean it just crashed? How to see/find/analyze crashdump, if it exists?
  3. Is there any way to enable a fuature to generate a crashdump file in case if any failure in the future?

Thank you.

2

Consider at least upgrading to ESXi 5.5, preferably 6.5. I had MANY issues with VM and even full host crashes with 5.1, most have been resolved after moving to 6.5

3

Hi David,

Thank you for the response. Sure thing, we will upgrade ESXi soon.

Would you please help me to see how can I get crashdump on RouterOS (CHR, in particular) and other MikroTik boxes in general?

4

alex1,

Thank you for your question in the forum thread The Dude v6 server - CCR vs CHR as a platform.
I will try to ask my colleagues more details about our virtual environment.

Now I see uptime 8 days.
It was rebooted but I don’t know why, maybe this was done by my colleague. Will monitor what is going on.

5

alex1,
Our CHR is running in ESXi 6.0 cluster.

Thank you!