Re: [one-users] Private cloud with Front-end on Virtual MAchine HA
Hi Daniel,
Thanks for the insight.
So basically, to sum it up, there is currently no way of running the
OpenNebula management node (with all functionality inside one VM) on it's
own virtualisation cluster (and thus managing itself along with the rest of
the cluster).
This would mean that you need two physical servers to create a (proper)
redundant (active / passive) setup?!
Thats a shame since we'd like to keep hardware to the minimum for this. But
if the best practice is to have two physical servers then I guess we'll
have to live with that and just maken it happen.
Although I'm still hoping for a better solution. So I'd be very interested
in hearing how others created their redundant OpenNebula management node,
is 2x server hardware the best practice or are there other solutions
(with less hardware) to achieve this?
2014-11-13 12:33 GMT+01:00 Daniel Dehennin daniel.dehen...@baby-gnu.org:
Giancarlo De Filippis gdefilip...@ltbl.it writes:
I hope (like you) that someone (users or OpenNebula Team) have best
practices on how to run OpenNebula in a VM.
Hello,
The ONE frontend VM can not manage itself, you must use something else.
I made a test with pacemaker/corosync and it can be quite easy[1]:
#+begin_src conf
primitive Stonith-ONE-Frontend stonith:external/libvirt \
params hostlist=one-frontend hypervisor_uri=qemu:///system \
pcmk_host_list=one-frontend pcmk_host_check=static-list \
op monitor interval=30m
primitive ONE-Frontend-VM ocf:heartbeat:VirtualDomain \
params config=/var/lib/one/datastores/one/one.xml \
op start interval=0 timeout=90 \
op stop interval=0 timeout=100 \
utilization cpu=1 hv_memory=1024
group ONE-Frontend Stonith-ONE-Frontend ONE-Frontend-VM
location ONE-Frontend-run-on-hypervisor ONE-Frontend \
rule $id=ONE-Frontend-run-on-hypervisor-rule 40: #uname eq
nebula1 \
rule $id=ONE-Frontend-run-on-hypervisor-rule-0 30: #uname eq
nebula3 \
rule $id=ONE-Frontend-run-on-hypervisor-rule-1 20: #uname eq
nebula2
#+end_src
I have troubles with my cluster because my nodes _and_ the ONE frontend
needs to access the same SAN.
My nodes have two LUNs over multipath FC (/dev/mapper/SAN-FS{1,2}), they
are both PV of a cluster volume group (cLVM) with a GFS2 on top.
So I need:
- corosync for messaging
- dlm for cLVM and GFS2
- cLVM
- GFS2
I add the LUNs as raw block disks to my frontend VM and install the
whole stack in it, but I'm facing some communication issues, and manage
to solve somes[3].
According to the pacemaker mailing list, having the nodes _and_ a VM in
the same pacemaker/corosync cluster ”sounds like a recipe for
disaster”[2].
Hope this will help you have a picture of the topic.
Regards.
Footnotes:
[1]
http://clusterlabs.org/doc/en-US/Pacemaker/1.1-crmsh/html-single/Clusters_from_Scratch/index.html
[2]
http://oss.clusterlabs.org/pipermail/pacemaker/2014-November/023000.html
[3]
http://oss.clusterlabs.org/pipermail/pacemaker/2014-November/022964.html
--
Daniel Dehennin
Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF
Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
--
Bart G.
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Private cloud with Front-end on Virtual MAchine HA
Bart b...@pleh.info writes: Hi Daniel, Hello, So basically, to sum it up, there is currently no way of running the OpenNebula management node (with all functionality inside one VM) on it's own virtualisation cluster (and thus managing itself along with the rest of the cluster). You can run the VM on the same hardware, but it will not be managed by OpenNebula, since you need OpenNebula to start OpenNebula VMs. That's why the documentation[1] explains the setup of an H-A system. If you have two physical servers, you may create two OpenNebula VMs, one master and one slave with replicated database between them. Regards. Footnotes: [1] http://docs.opennebula.org/4.10/advanced_administration/high_availability/oneha.html -- Daniel Dehennin Récupérer ma clef GPG: gpg --recv-keys 0xCC1E9E5B7A6FE2DF Fingerprint: 3E69 014E 5C23 50E8 9ED6 2AAD CC1E 9E5B 7A6F E2DF signature.asc Description: PGP signature ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Private cloud with Front-end on Virtual MAchine HA
Hi all, someone have some documentation for a sample of private cloud structure with: - Two nodes in HA - Front-end on virtual machine HA With storage file system on glusterfs Thanks all.. GD ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Private cloud with Front-end on Virtual MAchine HA
Hi, Do you mean to have the frontend (OpenNebula management) running on the actual OpenNebula cluster? If that's the case then I would also be very interested in this scenario :) As for GlusterFS, we've followed these instructions with success: - http://docs.opennebula.org/4.10/ - http://docs.opennebula.org/4.10/administration/storage/gluster_ds.html Also, all other instructions that we found were also on the documentation site. With it, we've created the following: - *GlusterFS:* Two storage hosts (running CentOS 7) in the back-end which provide the whole image/datastore for OpenNebula. HA is setup via round robin DNS. After testing we've found this to work quite well in terms of performance and redundancy (better then expected). All network ports are setup as a bond using balance-alb. - *Hypervisors: *a.t.m. 4 hypervisors (will eventually grow to 8) running CentOS 7. They have 2 network ports setup using bonding with balance-alb, they are connected to the virt/storage VLAN. And two network ports setup in a bond using active-backup, this is used for the bridge port (with VLAN tagging enabled). Active-backup seems to be the proper setting for a bridge port since a bridge can't work that well with balance-alb. - *OpenNebula management node:* This includes the OpenNebula daemon, sunstone and all other functions. Currently, since we're building up this environment. Hypervisors and storage have been arranged and we're happy with how those parts are setup, however the management node is currently still something we're not very sure about. We're considering buying two physical servers to do this job and create an active backup solution as described in the documentation: http://docs.opennebula.org/4.10/advanced_administration/high_availability/oneha.html. However, buying hardware just for this seems a waste since these servers won't be doing that much (resource wise). - We do have plans on testing this node, dual, on the actual OpenNebula cluster. But something is holding us back. I can't find any proof of someone achieving this. Also, we're seeing some issues that might occur when OpenNebula is managing itself (e.g. live migration, reboot, shutdown, etc.). - *Virtual machine HA setup: *We haven't actually started with this part yet, but this document describes on how you can create a setup where virtual machines become HA: http://docs.opennebula.org/4.10/advanced_administration/high_availability/ftguide.html For us this is something we'll probably start looking into once we've found a proper setup for the OpenNebula management node. Hopefully this info helps you a little and I also hope someone else can elaborate on running OpenNebula as a VM on it's own cluster. Or, at least other best practices on how to run OpenNebula in a VM without buying two physical servers for this job, (any insight on this subject would be helpful :) -- Bart 2014-11-13 9:50 GMT+01:00 Giancarlo De Filippis gdefilip...@ltbl.it: Hi all, someone have some documentation for a sample of private cloud structure with: - Two nodes in HA - Front-end on virtual machine HA With storage file system on glusterfs Thanks all.. GD ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- Bart G. ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] Private cloud with Front-end on Virtual MAchine HA
Thanks so much Bart, That's the case. because vm running on cluster is more similar to VMware HA solution (only two hosts without front-end) For the glusterfs i've already used this solution on my public cloud with success. You helped me for port-setup of hypervisors :) I hope (like you) that someone (users or OpenNebula Team) have best practices on how to run OpenNebula in a VM. Thks. Giancarlo Il 13/11/2014 11:38 Bart ha scritto: Hi, Do you mean to have the frontend (OpenNebula management) running on the actual OpenNebula cluster? If that's the case then I would also be very interested in this scenario :) As for GlusterFS, we've followed these instructions with success: * http://docs.opennebula.org/4.10/ [2] * http://docs.opennebula.org/4.10/administration/storage/gluster_ds.html [3] Also, all other instructions that we found were also on the documentation site. With it, we've created the following: * GLUSTERFS: Two storage hosts (running CentOS 7) in the back-end which provide the whole image/datastore for OpenNebula. HA is setup via round robin DNS. After testing we've found this to work quite well in terms of performance and redundancy (better then expected). All network ports are setup as a bond using balance-alb. * HYPERVISORS: a.t.m. 4 hypervisors (will eventually grow to 8) running CentOS 7. They have 2 network ports setup using bonding with balance-alb, they are connected to the virt/storage VLAN. And two network ports setup in a bond using active-backup, this is used for the bridge port (with VLAN tagging enabled). Active-backup seems to be the proper setting for a bridge port since a bridge can't work that well with balance-alb. * OPENNEBULA MANAGEMENT NODE: This includes the OpenNebula daemon, sunstone and all other functions. Currently, since we're building up this environment. Hypervisors and storage have been arranged and we're happy with how those parts are setup, however the management node is currently still something we're not very sure about. We're considering buying two physical servers to do this job and create an active backup solution as described in the documentation: http://docs.opennebula.org/4.10/advanced_administration/high_availability/oneha.html [4]. However, buying hardware just for this seems a waste since these servers won't be doing that much (resource wise). * We do have plans on testing this node, dual, on the actual OpenNebula cluster. But something is holding us back. I can't find any proof of someone achieving this. Also, we're seeing some issues that might occur when OpenNebula is managing itself (e.g. live migration, reboot, shutdown, etc.). * VIRTUAL MACHINE HA SETUP: We haven't actually started with this part yet, but this document describes on how you can create a setup where virtual machines become HA: http://docs.opennebula.org/4.10/advanced_administration/high_availability/ftguide.html [5] For us this is something we'll probably start looking into once we've found a proper setup for the OpenNebula management node. Hopefully this info helps you a little and I also hope someone else can elaborate on running OpenNebula as a VM on it's own cluster. Or, at least other best practices on how to run OpenNebula in a VM without buying two physical servers for this job, (any insight on this subject would be helpful :) -- Bart 2014-11-13 9:50 GMT+01:00 Giancarlo De Filippis gdefilip...@ltbl.it: Hi all, someone have some documentation for a sample of private cloud structure with: - Two nodes in HA - Front-end on virtual machine HA With storage file system on glusterfs Thanks all.. GD ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org [1] -- Bart G. Links: -- [1] http://lists.opennebula.org/listinfo.cgi/users-opennebula.org [2] http://docs.opennebula.org/4.10/ [3] http://docs.opennebula.org/4.10/administration/storage/gluster_ds.html [4] http://docs.opennebula.org/4.10/advanced_administration/high_availability/oneha.html [5] http://docs.opennebula.org/4.10/advanced_administration/high_availability/ftguide.html___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
