I have a server with ESXI 5.5 and two virtual machines running pfSense, the problem is that almost every morning the server crashes, I can not even access the virtual machines or the own server by the vSphere Client. What could it be? Hardware problem? Or otherwise virtual machines can cause the host crash? Thanks
Any suggestions?
Thanks
Hi there, can you please try installing these two vibs found on this site: http://www.vladan.fr/realtek-8169-nics-not-detected-under-esxi-5-5/ - A walkthrough on how to do it is there.
What came to my mind is that you might want to remove one NIC card and leave the second there to determine the faulting component.
Good luck!
Fttz,Can you try to shutdown the Pfsense VM and see how your ESXi is stable or try to change the network driver(vmxnet3 or e1000) of your Pfsense VM?
It is almost 1 week issue,better try to submit the SR ticket with VMware.
Today I put another VM to run on the host and stopped the two that were with pfSense to better assess and see if will crash with the other vm. If you lock will do what you guys recommend me. I'm sorry for the delay in return, had other more pressing problems.
Thanks
Today I put another VM to run on the host and stopped the two that were with pfSense to better assess and see if will crash with the other vm. If you lock will do what you guys recommend me. I'm sorry for the delay in return, had other more pressing problems.
Thanks
Yesterday the server crashed again (but with no purple screen error). The logs are attached.
I installed new drivers as recommended by Alistar. And I changed the type of E1000 adapter to vmxnet 3 as recommended by Dee006. Let's wait.
But I believe to be a hardware problem with the server.
I am at a loss, the logs do not show anything remotely useful in your case
The only line that *could* say something that it could indeed be the Realtek NIC:
2015-01-22T17:01:46.814Z cpu2:33343)WARNING: LinDMA: Linux_DMACheckConstraints:138: Cannot map machine address = 0x1153bb5f0, length = 134 for device 0000:05:02.0; reason = address exceeds dma_mask (0xffffffff))
which is the realtek NIC:
2015-01-21T17:42:10.906Z cpu3:33307)PCI: driver r8169 claimed device 0000:05:02.0
I'll keep my fingers crossed that your new drivers remedied your issue. If anyone else has some useful pointers I'll be really glad to see them!
uptime:
15:48:40 up 3 days, 19:57:40, load average: 0.42, 0.41, 0.41
After the last driver update. Let's wait...
Thanks
fingers crossed!
Crashed again. With the windows 7 VM ran smoothly for 3 days (using only the Realtek NIC). This was after updating the drivers. So I went back VMs with pfSense and the server came back to catch. Even changed the settings, I stopped using the Realtek NIC and used only the 3 D-Link 528T NICs and still crashed.
But the images of pfSense running on another server worked for days without crashing.
Attached below the logs, none of the times presented purple screen.
Stopped again !!! I migrated the service to another server and turned off the problematic, I will give another destination for this hardware and use the other servers to VMs.
; (((
You are running microcode patch level 0x12. The current microcode patch level for your CPU is 0x1b. I would recommend a BIOS upgrade.
I updated the BIOS and removed a RAM memory stick (leaving the system with 2 X Kingston DDR3 2GB).
now:
00:21:59 up 3 days, 5:34:43, load average: 0.05, 0:07, 0:07
Running the two VM with pfSense (using the 4 NICs) and also put a openSUSE for testing ...
Let's wait ...
Thanks
After: 18:35:36 up 7 days, 23:48:20, load average: 0.23 0:14, 0:10
of uptime
I turned off the server and put back the cover of the cabinet and turned on again.
But not before completing 24 hours the server crashed again (the same way, without displaying the purple screen).
Could be an overheating problem?
I have not seen any indication of a high-temperature.
Attached logs