------=_Part_89_1367891.1363600245934 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Ok, now I understand, so what do we need to get the power management config= ured and working?=20 Regards=20 Jose=20 ----- Mensagem original ----- De: "Shu Ming" <shuming@linux.vnet.ibm.com>=20 Para: "Tim Hildred" <thildred@redhat.com>=20 Cc: Users@ovirt.org=20 Enviadas: Segunda-feira, 18 de Mar=C3=A7o de 2013 5:18:28=20 Assunto: Re: [Users] High Availability=20 Tim Hildred:=20
If one host is down because of network interrupt or power=20 failure, the engine should know how many HA VMs are down and find=20 out the VM images on the storage domain to start the VM instances on=20 another host in the cluster. Why do we need power manager to be=20 configured?=20 Power management allows the Manager to start highly available virtual mac= hines on new hosts without worrying that virtual machine hard disk images w= ill be corrupted.=20 =20 Imagine a situation in which the Manager cannot communicate with the host= a highly available virtual machine is running on. If the host is still run= ning as expected, and the virtual machine is also still running, the virtua= l machine is writing to its hard disk image.=20 Tim,=20
=20 If the Manager starts that virtual machine on another host in the cluster= , then both virtual machine instances will try and write to the disk image,= and cause hard disk corruption.=20 =20 Power management lets the Manager be sure that only one instance of the h= ighly available virtual machine is running, because the instance on the hos= t the Manager couldn't communicate cannot survive a host reboot.=20 =20 Tim Hildred, RHCE=20 Content Author II - Engineering Content Services, Red Hat, Inc.=20 Brisbane, Australia=20 Email: thildred@redhat.com=20 Internal: 8588287=20 Mobile: +61 4 666 25242=20 IRC: thildred=20 =20 ----- Original Message -----=20
From: "Shu Ming" <shuming@linux.vnet.ibm.com>=20 To: "Tim Hildred" <thildred@redhat.com>=20 Cc: dron@redhat.com, Users@ovirt.org=20 Sent: Monday, March 18, 2013 1:31:12 PM=20 Subject: Re: [Users] High Availability=20 =20 =20 Tim,=20 =20 Thanks for your information. I am not sure why we need power=20 management to be configured for the hosts running HA virtual=20 machines. We only need a method to check the VM or host status and=20 a method to restart the VM instances with existing VM images on=20 another host. Is it required to force powering down the failing host=20 forever to make sure the failing host will not come back to live=20 again?=20 =20 =20 =20 =20 *=20 Power management must be configured for the hosts running the highly=20 available virtual machines.=20 *=20 The host running the highly available virtual machine must be part of=20 a cluster which has other available hosts.=20 *=20 The destination host must be running.=20 *=20 The source and destination host must have access to the data domain=20 on which the virtual machine resides.=20 *=20 The source and destination host must have access to the same virtual=20 networks and VLANs.=20 *=20 There must be enough CPUs on the destination host that are not in use=20 to support the virtual machine's requirements.=20 *=20 There must be enough RAM on the destination host that is not in use=20 to support the virtual machine's requirements.=20 =20 =20 Tim Hildred:=20 =20 =20 You might also find this helpful:=20 https://access.redhat.com/knowledge/docs/en-US/Red_Hat_Enterprise_Virtua=
Thanks for clarification. That is what I expected.=20 lization/3.1/html/Administration_Guide/High_availability_considerations.htm= l=20
The topics before and after it explain a bit more about high=20 availability.=20 =20 Tim Hildred, RHCE=20 Content Author II - Engineering Content Services, Red Hat, Inc.=20 Brisbane, Australia=20 Email: thildred@redhat.com Internal: 8588287=20 Mobile: +61 4 666 25242=20 IRC: thildred=20 =20 ----- Original Message -----=20 =20 From: "Dafna Ron" <dron@redhat.com> To: suporte@logicworks.pt Cc:=20 Users@ovirt.org Sent: Monday, March 18, 2013 4:46:51 AM=20 Subject: Re: [Users] High Availability=20 =20 I think that there is some confusion here so I will explain what are=20 the=20 configurations for fail-over.=20 =20 power management will reboot your host if a connectivity issue is=20 detected so all your vm's will be killed.=20 resilience policy will allow you to choose vm migration policy during=20 a=20 host failure and its configured in the cluster level (clusters ->=20 select=20 cluster -> general sub tab -> edit policy)=20 High Availability is configured only for servers type vm's and what=20 it=20 does is re-run the vm in case the pid of the vm is killed (so most=20 commonly, if you have power management configured, and the host is=20 rebooted, the vm will start automatically on a different host).=20 =20 so it really depends what you want. if you want vm migration than=20 look=20 into cluster policy, if you want a specific vm to always be up and=20 you=20 don't care about the other vm's than configure power management and a=20 HA=20 vm. it really depends on what you need.=20 =20 =20 =20 =20 On 03/17/2013 07:15 PM, suporte@logicworks.pt wrote:=20 =20 Is it Mandatory to have power manamement enabled? if yes what=20 equipment do you recommend to use with it?=20 =20 Regards=20 Jose=20 =20 =20 ----- Mensagem original -----=20 De: "Ren=C3=A9 Koch" <r.koch@ovido.at> Para: suporte@logicworks.pt ,=20 Users@ovirt.org Enviadas: Domingo, 17 Mar=C3=A7o, 2013 16:47:41=20 Assunto: RE: [Users] High Availability=20 =20 Hi,=20 =20 You have to configure power management to make high availability=20 working and mark the vms high availability checkbox...=20 =20 =20 Regards,=20 Ren=C3=A9=20 =20 =20 -----Original message-----=20 =20 From:suporte@logicworks.pt <suporte@logicworks.pt> Sent: Sunday 17th=20 March 2013 17:28=20 To: Users@ovirt.org Subject: [Users] High Availability=20 =20 What should I need to configure to put HA working? I mean, when a=20 host broke all the VM automatically move to another host.=20 Do I need to have Power management enabled?=20 =20 Thanks=20 _______________________________________________=20 Users mailing list Users@ovirt.org=20 http://lists.ovirt.org/mailman/listinfo/users=20 _______________________________________________=20 Users mailing list Users@ovirt.org=20 http://lists.ovirt.org/mailman/listinfo/users --=20 Dafna Ron=20 _______________________________________________=20 Users mailing list Users@ovirt.org=20 http://lists.ovirt.org/mailman/listinfo/users=20 _______________________________________________=20 Users mailing list Users@ovirt.org=20 http://lists.ovirt.org/mailman/listinfo/users=20 =20 --=20 ---=20 =E8=88=92=E6=98=8E Shu Ming=20 Open Virtualization Engineerning; CSTL, IBM Corp.=20 Tel: 86-10-82451626 Tieline: 9051626 E-mail: shuming@cn.ibm.com or=20 shuming@linux.vnet.ibm.com Address: 3/F Ring Building, ZhongGuanCun=20 Software Park, Haidian District, Beijing 100193, PRC=20
>> another host in the cluster. Why do we need power manager to be<b= r>>> configured?<br>> Power management allows the Manager to start= highly available virtual machines on new hosts without worrying that virtu= al machine hard disk images will be corrupted.<br>><br>> Imagine a si= tuation in which the Manager cannot communicate with the host a highly avai= lable virtual machine is running on. If the host is still running as expect= ed, and the virtual machine is also still running, the virtual machine is w= riting to its hard disk image.<br>Tim,<br><br>Thanks for clarification. Tha= t is what I expected.<br><br>><br>> If the Manager starts that virtua= l machine on another host in the cluster, then both virtual machine instanc= es will try and write to the disk image, and cause hard disk corruption.<br= ><br>> Power management lets the Manager be sure that only one insta= nce of the highly available virtual machine is running, because the instanc= e on the host the Manager couldn't communicate cannot survive a host reboot= .<br>><br>> Tim Hildred, RHCE<br>> Content Author II - Engineering= Content Services, Red Hat, Inc.<br>> Brisbane, Australia<br>> Email:=
>> Tim Hildred:<br>>><br>>><br>>> You might also f= ind this helpful:<br>>> https://access.redhat.com/knowledge/docs/en-U= S/Red_Hat_Enterprise_Virtualization/3.1/html/Administration_Guide/High_avai= lability_considerations.html<br>>> The topics before and after it exp= lain a bit more about high<br>>> availability.<br>>><br>>>= ; Tim Hildred, RHCE<br>>> Content Author II - Engineering Content Ser= vices, Red Hat, Inc.<br>>> Brisbane, Australia<br>>> Email: thi= ldred@redhat.com Internal: 8588287<br>>> Mobile: +61 4 666 25242<br>&= gt;> IRC: thildred<br>>><br>>> ----- Original Message -----<= br>>><br>>> From: "Dafna Ron" <dron@redhat.com> To: supor= te@logicworks.pt Cc:<br>>> Users@ovirt.org Sent: Monday, March 18, 20= 13 4:46:51 AM<br>>> Subject: Re: [Users] High Availability<br>>>= ;<br>>> I think that there is some confusion here so I will explain w= hat are<br>>> the<br>>> configurations for fail-over.<br>>&g= t;<br>>> power management will reboot your host if a connectivity iss= ue is<br>>> detected so all your vm's will be killed.<br>>> res= ilience policy will allow you to choose vm migration policy during<br>>&= gt; a<br>>> host failure and its configured in the cluster level (clu= sters -><br>>> select<br>>> cluster -> general sub tab -&= gt; edit policy)<br>>> High Availability is configured only for serve= rs type vm's and what<br>>> it<br>>> does is re-run the vm in c= ase the pid of the vm is killed (so most<br>>> commonly, if you have =
--=20 ---=20 =E8=88=92=E6=98=8E Shu Ming=20 Open Virtualization Engineerning; CSTL, IBM Corp.=20 Tel: 86-10-82451626 Tieline: 9051626 E-mail: shuming@cn.ibm.com or shuming@= linux.vnet.ibm.com=20 Address: 3/F Ring Building, ZhongGuanCun Software Park, Haidian District, B= eijing 100193, PRC=20 _______________________________________________=20 Users mailing list=20 Users@ovirt.org=20 http://lists.ovirt.org/mailman/listinfo/users=20 ------=_Part_89_1367891.1363600245934 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable <html><head><style type=3D'text/css'>p { margin: 0; }</style></head><body><= div style=3D'font-family: verdana,helvetica,sans-serif; font-size: 10pt; co= lor: #330066'>Ok, now I understand, so what do we need to get the power man= agement configured and working?<br><br>Regards<br>Jose<br><br><hr id=3D"zwc= hr"><div style=3D"color: rgb(0, 0, 0); font-weight: normal; font-style: nor= mal; text-decoration: none; font-family: Helvetica,Arial,sans-serif; font-s= ize: 12pt;"><b>De: </b>"Shu Ming" <shuming@linux.vnet.ibm.com><br><b>= Para: </b>"Tim Hildred" <thildred@redhat.com><br><b>Cc: </b>Users@ovi= rt.org<br><b>Enviadas: </b>Segunda-feira, 18 de Mar=C3=A7o de 2013 5:18:28<= br><b>Assunto: </b>Re: [Users] High Availability<br><br>Tim Hildred:<br>>= ;> If one host is down because of network interrupt or power<br>>>= failure, the engine should know how many HA VMs are down and find<br>>&= gt; out the VM images on the storage domain to start the VM instances on<br= thildred@redhat.com<br>> Internal: 8588287<br>> Mobile: +61 4 666 25= 242<br>> IRC: thildred<br>><br>> ----- Original Message -----<br>&= gt;> From: "Shu Ming" <shuming@linux.vnet.ibm.com><br>>> To:= "Tim Hildred" <thildred@redhat.com><br>>> Cc: dron@redhat.com,= Users@ovirt.org<br>>> Sent: Monday, March 18, 2013 1:31:12 PM<br>>= ;> Subject: Re: [Users] High Availability<br>>><br>>><br>>= ;> Tim,<br>>><br>>> Thanks for your information. I am not su= re why we need power<br>>> management to be configured for the hosts = running HA virtual<br>>> machines. We only need a method to che= ck the VM or host status and<br>>> a method to restart the VM instanc= es with existing VM images on<br>>> another host. Is it required to f= orce powering down the failing host<br>>> forever to make sure the fa= iling host will not come back to live<br>>> again?<br>>><br>>= ;><br>>><br>>><br>>> *<br>>>= Power management must be configured for the hosts running the highly<br>&g= t;> available virtual machines.<br>>> *<br>>= ;> The host running the highly available virtual machine must be part of= <br>>> a cluster which has other available hosts.<br>>> = *<br>>> The destination host must be running.<br>>>= ; *<br>>> The source and destination host must ha= ve access to the data domain<br>>> on which the virtual machine resid= es.<br>>> *<br>>> The source and destinatio= n host must have access to the same virtual<br>>> networks and VLANs.= <br>>> *<br>>> There must be enough CPUs on= the destination host that are not in use<br>>> to support the virtua= l machine's requirements.<br>>> *<br>>> The= re must be enough RAM on the destination host that is not in use<br>>>= ; to support the virtual machine's requirements.<br>>><br>>><br= power management configured, and the host is<br>>> rebooted, the vm w= ill start automatically on a different host).<br>>><br>>> so it= really depends what you want. if you want vm migration than<br>>> lo= ok<br>>> into cluster policy, if you want a specific vm to always be = up and<br>>> you<br>>> don't care about the other vm's than con= figure power management and a<br>>> HA<br>>> vm. it really depe= nds on what you need.<br>>><br>>><br>>><br>>><br>&g= t;> On 03/17/2013 07:15 PM, suporte@logicworks.pt wrote:<br>>><br>= >> Is it Mandatory to have power manamement enabled? if yes what<br>&= gt;> equipment do you recommend to use with it?<br>>><br>>> = Regards<br>>> Jose<br>>><br>>><br>>> ----- Mensagem= original -----<br>>> De: "Ren=C3=A9 Koch" <r.koch@ovido.at> Pa= ra: suporte@logicworks.pt ,<br>>> Users@ovirt.org Enviadas: Domingo, = 17 Mar=C3=A7o, 2013 16:47:41<br>>> Assunto: RE: [Users] High Availabi= lity<br>>><br>>> Hi,<br>>><br>>> You have to config= ure power management to make high availability<br>>> working and mark= the vms high availability checkbox...<br>>><br>>><br>>> = Regards,<br>>> Ren=C3=A9<br>>> <br>>> <br>&= gt;> -----Original message-----<br>>><br>>> From:suporte@log= icworks.pt <suporte@logicworks.pt> Sent: Sunday 17th<br>>> Marc= h 2013 17:28<br>>> To: Users@ovirt.org Subject: [Users] High Availabi= lity<br>>><br>>> What should I need to configure to put HA work= ing? I mean, when a<br>>> host broke all the VM automatically move to= another host.<br>>> Do I need to have Power management enabled?<br>&= gt;><br>>> Thanks<br>>> ____________________________________= ___________<br>>> Users mailing list Users@ovirt.org<br>>> http= ://lists.ovirt.org/mailman/listinfo/users<br>>> _____________________= __________________________<br>>> Users mailing list Users@ovirt.org<b= r>>> http://lists.ovirt.org/mailman/listinfo/users --<br>>> Daf= na Ron<br>>> _______________________________________________<br>>&= gt; Users mailing list Users@ovirt.org<br>>> http://lists.ovirt.org/m= ailman/listinfo/users<br>>> _________________________________________= ______<br>>> Users mailing list Users@ovirt.org<br>>> http://li= sts.ovirt.org/mailman/listinfo/users<br>>><br>>> --<br>>>= ---<br>>> =E8=88=92=E6=98=8E Shu Ming<br>>> Open Virtualizatio= n Engineerning; CSTL, IBM Corp.<br>>> Tel: 86-10-82451626 Tieli= ne: 9051626 E-mail: shuming@cn.ibm.com or<br>>> shuming@linux.vnet.ib= m.com Address: 3/F Ring Building, ZhongGuanCun<br>>> Software Park, H= aidian District, Beijing 100193, PRC<br><br><br>-- <br>---<br>=E8=88=92=E6= =98=8E Shu Ming<br>Open Virtualization Engineerning; CSTL, IBM Corp.<br>Tel= : 86-10-82451626 Tieline: 9051626 E-mail: shuming@cn.ibm.com or shumi= ng@linux.vnet.ibm.com<br>Address: 3/F Ring Building, ZhongGuanCun Software = Park, Haidian District, Beijing 100193, PRC<br><br><br>____________________= ___________________________<br>Users mailing list<br>Users@ovirt.org<br>htt= p://lists.ovirt.org/mailman/listinfo/users<br></div><br></div></body></html=
------=_Part_89_1367891.1363600245934--