we're experiencing the message “ssp_x86Linux_boot: null_cfsp.c: losing data.” in var/log/messages with HMP 4.1 SU213 installed on Linux RHEL 7.2. The system is installed on a ESXi virtual machine.
Did someone encounter the same message?
Make sure HPET is disabled at the OS level in this case. See installation Guide section 2.8 for more info on that piece:
thanks for your answer, anyway this not fix the issue.
I disabled HPET at the OS level following the documentation, and disabled HPET on the Virtual Machine also (hpet0.present=FALSE on the ESXi VM configuration), but nothing changed. In my previous post I forgot to specify that I installed HMP on a Virtual Machine on ESXi 5.5, do you think the issue could be related to VMware environment?
You might want to try the newer SU236 in this case. I just noticed you had RH 7.2 and SU213 at the time was only tested up to RH 7.1 since that was only available at the time. Jeff
Just installed SU236 with RHEL 7.2 on a VMWare VM, with HPET disabled at OS level and VM level (hpet0.present=FALSE). Same issue.
Could I have missed something on VM config? Did someone experiment the same behaviour with VMWare?
Let me add that I don't have the issue, nor on physical server neither on AWS machine (same OS and same installation procedure).
You might want to try reviewing this technote about tuning on VMs:
I really don't understand what is going wrong.
I set CPU affinity on ESXi host, installed VMware tools on the guest VM, configured ntpd on the host and on the guest, but nothing changed.
The only strange thing is that time synchronization between guest and host using VMware Tools capability seem to not work, and for this reason I disabled it and installed ntpd.
How much memory and CPUs do you have assigned to the VM in this case?