make sure you have sufficient resources available to run ESXi, i am suspacting VMs cpu and memory is overcommited
yes, I have enough spec on my server, before the first crash in yesterday my VMS been on without any problem
new information:
my server has 2 datastore, all problem in datastore 1
when I copy VMs from datastore 1 to datastore 2, VMS can run and turn on without any problem, I think datastore 1 going to read-only
no, please check space it sufficient for VM in data store.
What is the exact VM settings? CPU, Memory ... Does it happen for the new VMs or existing ones?
Can you check if you start a VM without a vDisk and then add a VMDK to it in power-on mode, PSOD of ESXi will happen again or not?!
Thanks for the reply to my problem
it happens for existing VMs and new VMS.
4core CPU , 8GB Ram, Ubuntu 16.04LTS, 60GB HDD
I found all VMs on Esxis have a connection to datastore(that Connected with Fc San Switch Datastore) has Problem
1- Turn On VM
2- Vm Going to Boot
3- Vm boot (all process in ram)
4- try to access data(write) or log in or OS (autosave from Ram to Hard) Vm and ESXI Crash and PSOD Happen
when did this problem start?
I add a new ESXi server to shared lun in FC San. and add to vcenter, join to cluster. this is the start point of all problems, now in all ESXi(new or old) which datastore that Connected to That FC San, we have problem. I can't turn on vcenter(PSOD Happen), I can't do anything only make a copy from important VMs to Other Datastore and turn them on.
Is it all hosts or just the new host that PSOD? You will need to check the core dump, see the blog below that might help
How to Analyze Purple Screen of Death (PSOD) - VMWARE ADDICTED