similar to: Transparent HugePages question

So basically you're saying that this THP feature is broken for CentOS 7 ? The feature is enabled by default. Is there a thread opened on this issue ? ?n mie., 24 aug. 2016 la 18:13, <m.roth at 5-cent.us> a scris: > Laurentiu Soica wrote: > > Hello, > > > > I have a CentOS 7 installation on baremetal with 2 CPUs, 10 cores each > and > > HT enabled, 128 GB

It sounds weird, but attempt to disable KSM and see would it help or no ? ________________________________ From: centos-virt-bounces at centos.org <centos-virt-bounces at centos.org> on behalf of Laurentiu Soica <laurentiu at soica.ro> Sent: Wednesday, August 17, 2016 4:56 AM To: Discussion about the virtualization on CentOS Subject: Re: [CentOS-virt] Nested KVM issue Enabled the

For myself KSM is unpredictable feature. The problem is Compute, just this node does "copy on write" , so only Compute. My concern exactly is where would it lead to worse or better Guest behavior ? I am not expecting complete fix. I would track via top/htop and dmesg via Cron on 1-2 hr period. ________________________________ From: centos-virt-bounces at centos.org

I would enable ksmtuned logging ,if it has been done verify logs ________________________________ From: centos-virt-bounces at centos.org <centos-virt-bounces at centos.org> on behalf of Laurentiu Soica <laurentiu at soica.ro> Sent: Tuesday, August 16, 2016 7:16 AM To: Discussion about the virtualization on CentOS Subject: Re: [CentOS-virt] Nested KVM issue Yes. It is on both

Is KSM enabled on your Compute Nodes ( presuming CentOS 7.2 on bare metal ) ? ________________________________ From: centos-virt-bounces at centos.org <centos-virt-bounces at centos.org> on behalf of Laurentiu Soica <laurentiu at soica.ro> Sent: Tuesday, August 16, 2016 5:25 AM To: Discussion about the virtualization on CentOS Subject: Re: [CentOS-virt] Nested KVM issue Running the

No luck with qemu-kvm-ev, the behavior is the same. Running perf record -a -g on the baremetal shows that most of the CPU time is in _raw_spin_lock Children Self Command Shared Object Symbol - 93.62% 93.62% qemu-kvm [kernel.kallsyms] [k] _raw_spin_lock - _raw_spin_lock + 45.30%

Sorry, How you trigger the problem ? B. ________________________________ From: centos-virt-bounces at centos.org <centos-virt-bounces at centos.org> on behalf of Laurentiu Soica <laurentiu at soica.ro> Sent: Tuesday, August 16, 2016 3:28 AM To: Discussion about the virtualization on CentOS Subject: Re: [CentOS-virt] Nested KVM issue Hello, The issue reproduced again and it

I would attempt to decrease number of VCPUS allocated to cloud VMs. Say try 4 => 2 . My guess there is not enough VCPUs to run OS itself. I also guess CPU model << Haswell. Please , confirm ( or not) if possible. Since Haswell was launched via my experience Intel Xeons based on this kernel (or latter kernels ) behaves much better then SandyBridge or IvyBridge based. Boris.

________________________________ From: centos-virt-bounces at centos.org <centos-virt-bounces at centos.org> on behalf of Laurentiu Soica <laurentiu at soica.ro> Sent: Sunday, August 14, 2016 10:17 AM To: Discussion about the virtualization on CentOS Subject: Re: [CentOS-virt] Nested KVM issue More details on the subject: I suppose it is a nested KVM issue because it raised after I

Reports posted look good for me. Config should provide the best available performance for cloud VM (L2) on Compute Node. 1. Please, remind me what goes wrong from your standpoint ? 2. Which CPU is installed on Compute Node && how much RAM ? Actually , my concern is :- Number_of_ Cloud_VMs versus Number_CPU_Cores ( not threads) Please, check `top` report in

Hello Xlord, The CPU does have VT support and I already have the nested KVM enabled. Checking the Linux Kernel release notes I saw that the nested kvm feature was implemented in 3.10 but he EPT support is in 3.12. So Centos 7's kernel which is 3.10 does have the EPT support? Is there a way to check that the EPT support is available in the kernel? Thank you, Laurentiu On Thu, Aug 4, 2016,

Laurentiu Soica wrote: > Hello, > > I have a CentOS 7 installation on baremetal with 2 CPUs, 10 cores each and > HT enabled, 128 GB RAM. > > The system has transparent hugetables enabled. <snip> > So it turns out that even is I have the THP enabled the pages still have a > size of 4K . > > Is there any extra setup in order to get 2MB pages ? > Be aware that

Hello, I have an OpenStack setup in virtual environment on CentOS 7. The baremetal has *nested KVM* enabled and 1 compute node as a VM. Inside the compute node I have multiple VMs running. After about every 3 days the VMs get inaccessible and the compute node reports high CPU usage. The qemu-kvm process for each VM inside the compute node reports full CPU usage. Please help me with some hints

Hi Xlord, Yes, the CPU has support for EPT. I wrongly thought that the nested EPT was first introduced in 3.12. Following your instructions I see that I have it enabled on my system as well. However, checking the kernel commits from 3.12 on search string 'nested ept' I found about 10 code changes/fixes for nested EPT.

Hello, I am looking for a Linux kernel for Centos 7 that implements a feature introduced in kernel version 3.12: "nested EPT support to KVM's nested VMX." If anyone has used this feature with Centos 7, please let me know. Laurentiu -------------- next part -------------- An HTML attachment was scrubbed... URL:

Dear Laurentiu, Agree with Johnny for kernel upgrade if upgrade helps, alternative if 3.10 or above already supported it, you can check from kernel parameter or options if they are disabled on purpose. You can do that via the "make menuconfig" via "https://wiki.centos.org/HowTos/I_need_the_Kernel_Source" or "https://wiki.centos.org/HowTos/BuildingKernelModules". Hope

I've tried with KSM disabled and nothing changed. I've upgraded KVM to qemu-kvm-ev. I'm waiting to see if there are any improvements and report back. ?n mie., 17 aug. 2016 la 15:10, Boris Derzhavets <bderzhavets at hotmail.com> a scris: > For myself KSM is unpredictable feature. The problem is Compute, just this > node > > does "copy on write" , so only

Both baremetal and compute ? Are there any other metrics do you consider useful to collect for troubleshooting purposes ? ?n mie., 17 aug. 2016 la 13:04, Boris Derzhavets <bderzhavets at hotmail.com> a scris: > It sounds weird, but attempt to disable KSM and see would it help or no ? > > > ------------------------------ > *From:* centos-virt-bounces at centos.org

Enabled the logging on both compute and baremetal. Nothing strange in logs: on baremetal : Wed Aug 17 11:51:01 EEST 2016: committed 62310764 free 58501808 Wed Aug 17 11:51:01 EEST 2016: 87025667 < 123574516 and free > 24714903, stop ksm on compute: Wed Aug 17 08:52:52 UTC 2016: committed 24547132 free 76730936 Wed Aug 17 08:52:52 UTC 2016: 45139624 < 102962460 and free > 20592492,

Yes. It is on both baremetal and compute node. ?n mar., 16 aug. 2016 la 13:37, Boris Derzhavets <bderzhavets at hotmail.com> a scris: > Is KSM enabled on your Compute Nodes ( presuming CentOS 7.2 on bare metal > ) ? > > > ------------------------------ > *From:* centos-virt-bounces at centos.org <centos-virt-bounces at centos.org> > on behalf of Laurentiu Soica