similar to: Out of memory causes VM shutdown

On 07/31/2017 02:55 PM, Michal Hocko wrote: > On Mon 31-07-17 12:13:33, Wei Wang wrote: >> Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and >> shouldn't be given to the host ksmd to scan. > Could you point me where this MADV_DONTNEED is done, please? Sure. It's done in the hypervisor when the balloon pages are received. Please see line 40 at

On 07/31/2017 02:55 PM, Michal Hocko wrote: > On Mon 31-07-17 12:13:33, Wei Wang wrote: >> Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and >> shouldn't be given to the host ksmd to scan. > Could you point me where this MADV_DONTNEED is done, please? Sure. It's done in the hypervisor when the balloon pages are received. Please see line 40 at

This patch is a revert of 'commit bb01b64cfab7 ("mm/balloon_compaction.c: enqueue zero page to balloon device")' Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and shouldn't be given to the host ksmd to scan. Therefore, it is not necessary to zero ballooned pages, which is very time consuming when the page amount is large. The ongoing fast balloon tests

This patch is a revert of 'commit bb01b64cfab7 ("mm/balloon_compaction.c: enqueue zero page to balloon device")' Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and shouldn't be given to the host ksmd to scan. Therefore, it is not necessary to zero ballooned pages, which is very time consuming when the page amount is large. The ongoing fast balloon tests

Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and shouldn't be given to the host ksmd to scan. Therefore, it is not necessary to zero ballooned pages, which is very time consuming when the page amount is large. The ongoing fast balloon tests show that the time to balloon 7G pages is increased from ~491ms to 2.8 seconds with __GFP_ZERO added. So, this patch removes the flag.

Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and shouldn't be given to the host ksmd to scan. Therefore, it is not necessary to zero ballooned pages, which is very time consuming when the page amount is large. The ongoing fast balloon tests show that the time to balloon 7G pages is increased from ~491ms to 2.8 seconds with __GFP_ZERO added. So, this patch removes the flag.

Has anyone tried using glusterfs to handle live sysvol replication? Cheers, Justin.

The VMs just boot up with the default network interface model which is 100Mb, is there a way to set it to a 1Gb model? Cheers, Justin.

If I ssh to a vm running Centos and then execute "poweroff" from within the vm, the vm shuts down as expected but the host in the admin ui becomes unavailable. Is anyone else experiencing this? I haven't checked to see if other guest OS's do the same thing yet. The painful thing is that then I have to shutdown all the VMs and reboot both the host and the management server

Hi, I have set up a 40GB iscsi target on my san and when I add it in to the storage pool and it shows up as 40GB. When I try to create a 40GB volume in the ovirt UI it tries to but fails. The error message is that the volume will not fit. If I create one smaller than 40GB it works. Is anyone else experiencing this? Also, if I then start a virtual machine up to install on the newly

On Mon 31-07-17 16:23:26, ZhenweiPi wrote: > On 07/31/2017 03:51 PM, Michal Hocko wrote: > > >On Mon 31-07-17 15:41:49, Wei Wang wrote: > >>>On 07/31/2017 02:55 PM, Michal Hocko wrote: > >>>> >On Mon 31-07-17 12:13:33, Wei Wang wrote: > >>>>> >>Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and >

On Mon 07-08-17 10:44:50, David Hildenbrand wrote: > On 03.08.2017 13:59, Wei Wang wrote: > > This patch is a revert of 'commit bb01b64cfab7 ("mm/balloon_compaction.c: > > enqueue zero page to balloon device")' > > > > Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and > > shouldn't be given to the host ksmd to scan.

On Mon 07-08-17 10:44:50, David Hildenbrand wrote: > On 03.08.2017 13:59, Wei Wang wrote: > > This patch is a revert of 'commit bb01b64cfab7 ("mm/balloon_compaction.c: > > enqueue zero page to balloon device")' > > > > Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and > > shouldn't be given to the host ksmd to scan.

I did a yum update on my management server and now the ovirt gui is no longer accessable. Is it supposed to just work or do I need to do something after the yum update? Justin.

On Thu 03-08-17 19:59:17, Wei Wang wrote: > This patch is a revert of 'commit bb01b64cfab7 ("mm/balloon_compaction.c: > enqueue zero page to balloon device")' > > Ballooned pages will be marked as MADV_DONTNEED by the hypervisor and > shouldn't be given to the host ksmd to scan. I find MADV_DONTNEED reference still quite confusing. What do you think about the

Are the build instructions out of date? I've been following them but when I get to "make update-app" I get make: *** No rule to make target `update-app'. Stop. I took a look in the Makefile and sure enough there's no updated-app target or update-host for that matter... What's the procedure now? Cheers, Justin.

Hello A box of mine running RELENG_7_0 and ZFS over a couple of disks (6 disks, 3 mirrors) seems to have gotten stuck. From Ctrl-T: load: 0.50 cmd: zsh 40188 [zfs:&buf_hash_table.ht_locks[i].ht_lock] 0.02u 0.04s 0% 3404k load: 0.43 cmd: zsh 40188 [zfs:&buf_hash_table.ht_locks[i].ht_lock] 0.02u 0.04s 0% 3404k load: 0.10 cmd: zsh 40188 [zfs:&buf_hash_table.ht_locks[i].ht_lock]

Hi, When I click on "Edit Network" on my managed node in the hosts tab of the oVirt UI I only see one NIC. My managed node has 2 NICs, one on the guest network and one on the admin network. The NIC on the admin network is not listed in the network editor. Is this normal? I'm trying to create a VM and boot the cobbler profile I created for Fedora 10 but I can only boot the VM

Hello, I having a weird problem while writing a bytecode module to a string, and after read/parse it for unsing on a jit. I write a pass to export function to module, and put this module inside a global variable. I use WriteBitcodeToFile for this. For debuging, after this write, I try to load the exported module with parseBitcodeFile. This two step works. After, while the compiled program is

Hi, Fantastic that Samba 4.0 has been released, I've just installed and joined it to an existing AD (single Win2k8R2 server running Exchange). The installation seemed to go well but I'm seeing odd messages when I show the replication status on both the Samba server and the Win2k8R2 server. Is there something I missed in the install? I followed the instructions outlined at