I can confirm that in the past (probably acs versions 4.2 or 4.1) I had several cases where ACS would restart vms which were running on a host which got disconnected fro ACS. As a result, I had two instances of the same vm running, creating a number of issues with disk consistency. I had to switch HA for that reason and rely on other means.
Andrei ----- Original Message ----- > From: "Simon Weller" <swel...@ena.com> > To: "users" <users@cloudstack.apache.org> > Cc: "Mohd Zainal Abidin Rabani" <zai...@nocser.net> > Sent: Wednesday, 3 February, 2016 19:05:20 > Subject: Re: Re[2]: High Availability Cloudstack > i'll test it out in our lab 4.7.1 lab and see what happens. > > ________________________________________ > From: Nux! <n...@li.nux.ro> > Sent: Wednesday, February 3, 2016 12:34 PM > To: users@cloudstack.apache.org > Cc: Mohd Zainal Abidin Rabani > Subject: Re: Re[2]: High Availability Cloudstack > > Simon, > > I think this is no longer correct. After a certain timeout the management > server > will bring up the VMs on another hypervisor. We discussed this in the somewhat > recent past, but can't find the thread right now. > There was some nice insight from Wido or Marcus. > > Lucian > > -- > Sent from the Delta quadrant using Borg technology! > > Nux! > www.nux.ro > > ----- Original Message ----- >> From: "Simon Weller" <swel...@ena.com> >> To: "users" <users@cloudstack.apache.org> >> Cc: "Mohd Zainal Abidin Rabani" <zai...@nocser.net> >> Sent: Wednesday, 3 February, 2016 16:23:55 >> Subject: Re: Re[2]: High Availability Cloudstack > >> My understanding is that HA will never actually function on KVM there way >> most >> expect, unless the original host that failed comes back up and the agents >> report the VMs not running. This is due to Cloudstack not wanting to risk >> starting a vm that already is locked by another process, on a different >> server. >> >> The only way I know of to force HA to function is to remove the host out of >> CloudStack, to force CloudStack to reallocate the VMs. >> >> There has been a lot of talk about introducing a reliable fencing mechanism >> into >> management, so that CloudStack management is 100% confident the vm is safe to >> restart on a new host. >> >> The current proposal for that is here: >> https://cwiki.apache.org/confluence/display/CLOUDSTACK/KVM+HA+with+IPMI+Fencing >> >> - Si >> >> >> >> ________________________________________ >> From: Andrei Mikhailovsky <and...@arhont.com> >> Sent: Wednesday, February 3, 2016 1:49 AM >> To: users >> Cc: Mohd Zainal Abidin Rabani >> Subject: Re: Re[2]: High Availability Cloudstack >> >> Hi, >> >> from what i can can see, this issue has been resolved in version 4.4 and >> Mohd is >> running 4.6. >> >> From my testing with KVM, the management server performs a poll once in a >> while >> and the agent replies if a vm is up. It might take some time before the state >> changes, like 30 seconds or so, so give it a bit more time. My testing show >> that the management server does update the state if the vm is manually >> powered >> down. >> >> I've not tested tha HA feature for sometime, but last time i checked it was >> working. You should also set the HA option in the vm's system offering. >> >> Andrei >> ----- Original Message ----- >>> From: "Suresh Sadhu" <suresh.sa...@citrix.com> >>> To: "users" <users@cloudstack.apache.org>, "Mohd Zainal Abidin Rabani" >>> <zai...@nocser.net> >>> Sent: Wednesday, 3 February, 2016 06:24:49 >>> Subject: RE: Re[2]: High Availability Cloudstack >> >>> Please see below issue: >>> https://issues.apache.org/jira/browse/CLOUDSTACK-5582 >>> >>> regards >>> sadhu >>> >>> -----Original Message----- >>> From: Mohd Zainal Abidin Rabani [mailto:zai...@nocser.net] >>> Sent: Wednesday, February 3, 2016 10:12 AM >>> To: users@cloudstack.apache.org >>> Subject: Re[2]: High Availability Cloudstack >>> >>> I'm using KVM that come with centos 7. That management also using centos 7. >>> Cloudstack version is 4.6. >>> >>> --------- >>> Regards, >>> Mohd Zainal Abidin Rabani >>> Technical Support >>> >>> ModernOne Data Solutions Sdn. Bhd.(1119382-D) No. 83-2, Jalan TKS 1, Kajang >>> Sentral, 43000 Kajang, Selangor. >>> T : 03-8737 0030 | F : 03-8737 0070 >>> E : zai...@nocser.net >>> W : www.modern.com.my | www.nocser.net >>> >>> >>> >>> ------ Original Message ------ >>> From: "Abhinandan Prateek" <abhinandan.prat...@shapeblue.com> >>> To: "users@cloudstack.apache.org" <users@cloudstack.apache.org>; "Mohd >>> Zainal >>> Abidin Rabani" <zai...@nocser.net> >>> Sent: 03/02/2016 12:38:38 >>> Subject: Re: High Availability Cloudstack >>> >>>>MS version, Hypervisor ? Are there more hosts in the cluster ? >>>> >>>> >>>> >>>> >>>>Abhinandan PrateekSoftware Architect, ShapeBlued: | s: +44 203 603 >>>>0540 | m: +91 970 11 99011e: abhinandan.prat...@shapeblue.com | t: | >>>>w: www.shapeblue.coma: 53 Chandos Place, Covent Garden London WC2N 4HS >>>>UKShape Blue Ltd is a company incorporated in England & Wales. >>>>ShapeBlue Services India LLP is a company incorporated in India and is >>>>operated under license from Shape Blue Ltd. Shape Blue Brasil >>>>Consultoria Ltda is a company incorporated in Brasil and is operated >>>>under license from Shape Blue Ltd. ShapeBlue SA Pty Ltd is a company >>>>registered by The Republic of South Africa and is traded under license >>>>from Shape Blue Ltd. ShapeBlue is a registered trademark. >>>>This email and any attachments to it may be confidential and are >>>>intended solely for the use of the individual to whom it is addressed. >>>>Any views or opinions expressed are solely those of the author and do >>>>not necessarily represent those of Shape Blue Ltd or related companies. >>>>If you are not the intended recipient of this email, you must neither >>>>take any action based upon its contents, nor copy or show it to anyone. >>>>Please contact the sender if you believe you have received this email >>>>in error. >>>> >>>> >>>>On 03/02/16, 9:24 AM, "Mohd Zainal Abidin Rabani" <zai...@nocser.net> >>>>wrote: >>>> >>>> >Hi, >>>> > >>>> >I try to do HA test for cloudstack. We are using DL380 G7. Same spec. >>>> >VM's have put option with HA. >>>> > >>>> >When i shutdown host A, from instance i can see all state is running. >>>> >But from insfrastructure host A already disconnected. Other instance >>>> >that not set using HA also state running. Suppose on instance that >>>> >use option HA is running and other than that is stopped. Is this is a >>>> >kind of bug? >>>> > >>>> >HA using what? Ping? Heartbeat? Why still show running? >>>> > >>>> >--------- >>>> >Regards, >>>> >Mohd Zainal Abidin Rabani >>>> >Technical Support >>>> > >>>> >ModernOne Data Solutions Sdn. Bhd.(1119382-D) No. 83-2, Jalan TKS 1, >>>> >Kajang Sentral, 43000 Kajang, Selangor. >>>> >T : 03-8737 0030 | F : 03-8737 0070 >>>> >E : zai...@nocser.net >>>> >W : www.modern.com.my | www.nocser.net >>>>Find out more about ShapeBlue and our range of CloudStack related >>>>services: >>>>IaaS Cloud Design & Build | CSForge – rapid IaaS deployment framework >>>>CloudStack Consulting | CloudStack Software Engineering CloudStack > > > >Infrastructure Support | CloudStack Bootcamp Training Courses
