Live migration of VM(0 downtime) while Hypervisor goes down in ovirt
------=_Part_1968242_1358094662.1518204324810 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Hi,Anyone can suggest how to setup VM Live migration (without restart vm) while Hypervisor goes down in ovirt?Using glusterfs is it possible? Then how? Thanks & RegardsRanjith Sent from Yahoo Mail on Android ------=_Part_1968242_1358094662.1518204324810 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 7bit Hi,<div id="yMail_cursorElementTracker_1518203906835">Anyone can suggest how to setup VM Live migration (without restart vm) while Hypervisor goes down in ovirt?</div><div id="yMail_cursorElementTracker_1518204210420">Using glusterfs is it possible? Then how?</div><div id="yMail_cursorElementTracker_1518204099547"><br></div><div id="yMail_cursorElementTracker_1518204100461">Thanks & Regards</div><div id="yMail_cursorElementTracker_1518204113848">Ranjith<br><br><div id="ymail_android_signature"><a href="https://overview.mail.yahoo.com/mobile/?.src=Android">Sent from Yahoo Mail on Android</a></div></div> ------=_Part_1968242_1358094662.1518204324810--
On Fri, Feb 9, 2018 at 9:25 PM, ranjithspr13@yahoo.com < ranjithspr13@yahoo.com> wrote:
Hi, Anyone can suggest how to setup VM Live migration (without restart vm) while Hypervisor goes down in ovirt?
I think there are two parts to achieving this: 1. Have a script that migrates VMs off a specific host. This should be easy to write using the Python/Ruby/Java SDK, Ansible or using REST directly. 2. Having this script run as a service when a host shuts down, in the right order - well before libvirt and VDSM shut down, and would be fast enough not to be terminated by systemd. This is a bit more challenging. Who's shutting down the hypervisor? (Or perhaps it is shutdown externally, due to overheating or otherwise?) Y.
Using glusterfs is it possible? Then how?
Thanks & Regards Ranjith
Sent from Yahoo Mail on Android <https://overview.mail.yahoo.com/mobile/?.src=Android>
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
------=_Part_2261449_1917009092.1518247846803 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi,
Who's shutting down the hypervisor? (Or perhaps it is shutdown externally= , due to overheating or otherwise?) We need a continuous availability of VM's in our production setup. If the h= ypervisor goes down due to any hardware failure or work load then VM's abov= e hypervisor will reboot and started on available hypervisors. This is norm= ally happening but it disrupting VM's. Can you suggest a solution in this c= ase? Can we achieve this challenge using glusterfs? Thanks & RegardsRanjith
Sent from Yahoo Mail on Android=20 =20 On Sat, Feb 10, 2018 at 2:07 AM, Yaniv Kaul<ykaul@redhat.com> wrote: =20 On Fri, Feb 9, 2018 at 9:25 PM, ranjithspr13@yahoo.com <ranjithspr13@yahoo.= com> wrote: Hi,Anyone can suggest how to setup VM Live migration (without restart vm) w= hile Hypervisor goes down in ovirt? I think there are two parts to achieving this:1. Have a script that migrate= s VMs off a specific host. This should be easy to write using the Python/Ru= by/Java SDK, Ansible or using REST directly.2. Having this script run as a = service when a host shuts down, in the right order - well before libvirt an= d VDSM shut down, and would be fast enough not to be terminated by systemd.= This is a bit more challenging. Who's shutting down the hypervisor? (Or perhaps it is shutdown externally, = due to overheating or otherwise?)Y.=C2=A0 Using glusterfs is it possible? Then how? Thanks & RegardsRanjith Sent from Yahoo Mail on Android ______________________________ _________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/ mailman/listinfo/users =20 ------=_Part_2261449_1917009092.1518247846803 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi,<div id=3D"yMail_cursorElementTracker_1518247387660"><br></div><div id= =3D"yMail_cursorElementTracker_1518247388075">>><span style=3D"font-f= amily: sans-serif;" id=3D"yMail_cursorElementTracker_1518247420756">Who's s= hutting down the hypervisor? (Or perhaps it is shutdown externally, due to = overheating or otherwise?)</span></div><div id=3D"yMail_cursorElementTracke= r_1518247388075"><font face=3D"sans-serif"><br></font></div><div id=3D"yMai= l_cursorElementTracker_1518247388075"><font face=3D"sans-serif" id=3D"yMail= _cursorElementTracker_1518247423263">We need a continuous availability of V= M's in our production setup. If the hypervisor goes down due to any hardwar= e failure or work load then VM's above hypervisor will reboot and started o= n available hypervisors. This is normally happening but it disrupting VM's.= Can you suggest a solution in this case? Can we achieve this challenge usi= ng glusterfs?</font></div><div id=3D"yMail_cursorElementTracker_15182473880= 75"><font face=3D"sans-serif"><br></font></div><div id=3D"yMail_cursorEleme= ntTracker_1518247388075"><font face=3D"sans-serif" id=3D"yMail_cursorElemen= tTracker_1518247756720">Thanks & Regards</font></div><div id=3D"yMail_c= ursorElementTracker_1518247388075"><font face=3D"sans-serif" id=3D"yMail_cu= rsorElementTracker_1518247774898">Ranjith<br></font><div id=3D"yMail_cursor= ElementTracker_1518247372890"><br><div id=3D"ymail_android_signature"><a hr= ef=3D"https://overview.mail.yahoo.com/mobile/?.src=3DAndroid">Sent from Yah= oo Mail on Android</a></div> <br> <blockquote style=3D"margin: 0 0 20px 0;"=
<div style=3D"font-family:Roboto, sans-serif; color:#6D00F6;"> <div>On Sa= t, Feb 10, 2018 at 2:07 AM, Yaniv Kaul</div><div><ykaul@redhat.com> w= rote:</div> </div> <div style=3D"padding: 10px 0 0 20px; margin: 10px 0 0 0= ; border-left: 1px solid #6D00F6;"> <div id=3D"yiv6654448903"><div><div dir= =3D"ltr"><br clear=3D"none"><div class=3D"yiv6654448903gmail_extra"><br cle= ar=3D"none"><div class=3D"yiv6654448903gmail_quote">On Fri, Feb 9, 2018 at = 9:25 PM, <a rel=3D"nofollow" shape=3D"rect" ymailto=3D"mailto:ranjithspr13@= yahoo.com" target=3D"_blank" href=3D"mailto:ranjithspr13@yahoo.com">ranjith= spr13@yahoo.com</a> <span dir=3D"ltr"><<a rel=3D"nofollow" shape=3D"rect= " ymailto=3D"mailto:ranjithspr13@yahoo.com" target=3D"_blank" href=3D"mailt= o:ranjithspr13@yahoo.com">ranjithspr13@yahoo.com</a>></span> wrote:<br c= lear=3D"none"><blockquote class=3D"yiv6654448903gmail_quote" style=3D"margi= n:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Hi,<div id=3D"yi= v6654448903m_-2539711703381469396yMail_cursorElementTracker_1518203906835">= Anyone can suggest how to setup VM Live migration (without restart vm) whil= e Hypervisor goes down in ovirt?</div></blockquote><div><br clear=3D"none">= </div><div>I think there are two parts to achieving this:</div><div>1. Have= a script that migrates VMs off a specific host. This should be easy to wri= te using the Python/Ruby/Java SDK, Ansible or using REST directly.</div><di= v>2. Having this script run as a service when a host shuts down, in the rig= ht order - well before libvirt and VDSM shut down, and would be fast enough= not to be terminated by systemd.</div><div>This is a bit more challenging.= </div><div><br clear=3D"none"></div><div id=3D"yMail_cursorElementTracker_1= 518247230119">Who's shutting down the hypervisor? (Or perhaps it is shutdow= n externally, due to overheating or otherwise?)</div><div>Y.</div><div clas= s=3D"yiv6654448903yqt0402040647 yQTDBase" id=3D"yiv6654448903yqtfd95121"><d= iv> </div></div><blockquote class=3D"yiv6654448903gmail_quote" style= =3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div cl= ass=3D"yiv6654448903yqt0402040647 yQTDBase" id=3D"yiv6654448903yqtfd74292">= <div id=3D"yiv6654448903m_-2539711703381469396yMail_cursorElementTracker_15= 18204210420">Using glusterfs is it possible? Then how?</div><div id=3D"yiv6= 654448903m_-2539711703381469396yMail_cursorElementTracker_1518204099547"><b= r clear=3D"none"></div><div id=3D"yiv6654448903m_-2539711703381469396yMail_= cursorElementTracker_1518204100461">Thanks & Regards</div><div id=3D"yi= v6654448903m_-2539711703381469396yMail_cursorElementTracker_1518204113848">= Ranjith<br clear=3D"none"><br clear=3D"none"><div id=3D"yiv6654448903m_-253= 9711703381469396ymail_android_signature"><a rel=3D"nofollow" shape=3D"rect"= target=3D"_blank" href=3D"https://overview.mail.yahoo.com/mobile/?.src=3DA= ndroid">Sent from Yahoo Mail on Android</a></div></div></div><br clear=3D"n= one">______________________________ _________________<br clear=3D"none"> Users mailing list<br clear=3D"none"> <a rel=3D"nofollow" shape=3D"rect" ymailto=3D"mailto:Users@ovirt.org" targe= t=3D"_blank" href=3D"mailto:Users@ovirt.org" id=3D"yMail_cursorElementTrack= er_1518247296623">Users@ovirt.org</a><br clear=3D"none"> <a rel=3D"nofollow" shape=3D"rect" target=3D"_blank" href=3D"http://lists.o= virt.org/mailman/listinfo/users">http://lists.ovirt.org/ mailman/listinfo/u= sers</a><div class=3D"yiv6654448903yqt0402040647 yQTDBase" id=3D"yiv6654448= 903yqtfd77509"><br clear=3D"none"> <br clear=3D"none"></div></blockquote></div><div class=3D"yiv6654448903yqt0= 402040647 yQTDBase" id=3D"yiv6654448903yqtfd89824"><br clear=3D"none"></div= </div></div></div></div> </div> </blockquote></div></div> ------=_Part_2261449_1917009092.1518247846803--
On 02/09/2018 01:25 PM, ranjithspr13@yahoo.com wrote:
Hi, Anyone can suggest how to setup VM Live migration (without restart vm) while Hypervisor goes down in ovirt? Using glusterfs is it possible? Then how?
Can't speak for glusterfs specifically, but this is how things work. All nodes know about the storage domains... this is so a different node can "take over" in a live migration. Not necessarily designed for a node to go down hard (but let's say the node has problems communicating, it might get fenced and live migration begins, and of course, the usual maintenance mode would force a live migration).... for hard down scenarios you might want to look a HA for the VMs.
participants (4)
-
Christopher Cox -
Ranjith P -
ranjithspr13＠yahoo.com
-
Yaniv Kaul