[ovirt-users] Frustration defines the deployment of Hosted Engine

Sandro Bonazzola sbonazzo at redhat.com
Wed Jun 28 10:33:47 UTC 2017


On Mon, Jun 26, 2017 at 9:15 AM, Yaniv Kaul <ykaul at redhat.com> wrote:

>
>
> On Mon, Jun 26, 2017 at 1:16 AM, Vinícius Ferrão <ferrao at if.ufrj.br>
> wrote:
>
>> Hello again,
>>
>> Setting the folder permissions to vdsm:kvm (36:36) done the trick to make
>> NFS work. I wasn't expecting this to work since it does not make sense to
>> me, making a parallel with the iSCSI problem.
>>
>> I'm starting to believe that's something just bad. Perhaps the Storage
>> system is running with something broken or the host machine is just
>> unstable.
>>
>> I will consider this as solved, since further inside the oVirt Engine
>> panel I was able to mount and use the same iSCSI sharing.
>>
>> But let me ask for sincere answers here:
>>
>> 1. oVirt is feature complete comparing to RHEV?
>>
>
> Yes. In fact it may have additional features unavailable in RHV.
>
>
>>
>> 2. Should I migrate from XenServer to oVirt? This is biased, I know, but
>> I would like to hear opinions. The folks with @redhat.com email
>> addresses will know how to advocate in favor of oVirt.
>>
>
> I don't know, and I don't think (having a @redhat.com email myself) I
> should advocate.
> If someone made a comparison (even if it's just for their own specific use
> case), it could be great if it could be shared. If you do - please share
> your thoughts.
>
>
>>
>> 3. Some "journalists" says that oVirt is like Fedora in comparison to
>> RHEL, is this really true? Or it's more aligned with a CentOS-like release?
>> Because Fedora isn't really an Enterprise OS, and I was looking for an
>> Enterprise Hypervisor. I'm aware that oVirt is the upstream from RHEV.
>>
>
> I'd cautiously say 'in between'. We strive to ensure oVirt is stable, and
> I believe we make good progress in every release.
> We also make an effort to quickly release fixes (minor releases). That
> being said, RHV has a longer life cycle, and for example, when oVirt
> stopped releasing oVirt 3.6.x, Red Hat continued to release minor versions
> of it.
>
> We have hundreds of oVirt users running it in production, many in large
> scale, with mission critical workloads.
>

And probably not surprisingly, oVirt is running oVirt in production:
https://engine-phx.ovirt.org/ovirt-engine/
for all the oVirt Ci infrastructure (13 hosts, 200 VMs).


>
> Lastly, oVirt  enables features in upstream before they are delivered in
> RHV.
>
> (That being said, I've had a very good experience with Fedora 24 which was
> rock solid for me, then I've had some misfortune with Fedora 25, and now
> I'm assessing if I should upgrade to F26 beta...)
>

For what it's worth, I'm on Fedora 26 since April 14th and works like a
charm for me.



>
>
>> 4. There is any good SPICE client for macOS? Or should I just use the
>> HTML5 version instead?
>>
>
> I'm afraid not.
> Y.
>
>
>>
>> Thanks,
>> V.
>>
>> Sent from my iPhone
>>
>> On 23 Jun 2017, at 18:50, Yaniv Kaul <ykaul at redhat.com> wrote:
>>
>>
>>
>> On Sat, Jun 24, 2017 at 12:23 AM, Vinícius Ferrão <ferrao at if.ufrj.br>
>> wrote:
>>
>>> Hello Adam and Karli,
>>>
>>> I will remap uid and gid of NFS to 36 and try again with NFS sharing.
>>>
>>> But this does not make much sense, because on iSCSI this should not
>>> happen. There are no permissions involved and when oVirt runs the
>>> hosted-engine setup it creates the ext3 filesystem on the iSCSI share
>>> without any issue. Here’s a photo of the network bandwidth during the OVF
>>> deployment: http://www.if.ufrj.br/~ferrao/ovirt/bandwidth-iscsi-ovf.jpg
>>>
>>> So it’s appears to be working. Something happens after the deployment
>>> that brokes the connections and kills vdsm.
>>>
>>
>> Indeed - may be two different issues. Let us know how the NFS works
>> first, then let's try with iSCSI.
>> Y.
>>
>>
>>>
>>> Thanks,
>>> V.
>>>
>>> On 23 Jun 2017, at 17:47, Adam Litke <alitke at redhat.com> wrote:
>>>
>>>
>>>
>>> On Fri, Jun 23, 2017 at 4:40 PM, Karli Sjöberg <karli at inparadise.se>
>>> wrote:
>>>
>>>>
>>>>
>>>> Den 23 juni 2017 21:08 skrev Vinícius Ferrão <ferrao at if.ufrj.br>:
>>>>
>>>> Hello oVirt folks.
>>>>
>>>> I’m a traitor of the Xen movement and was looking for some good
>>>> alternatives for XenServer hypervisors. I was aware of KVM for a long time
>>>> but I was missing a more professional and appliance feeling of the product,
>>>> and oVirt appears to deliver exactly what I’m looking for.
>>>>
>>>> Don’t get me wrong, I’m not saying that Xen is not good, I’m looking
>>>> for equal or better alternatives, but I’m starting to get frustrated with
>>>> oVirt.
>>>>
>>>> Firstly I’ve tried to install the oVirt Node on a VM in VMware Fusion
>>>> on my notebook, it was a no go. For whatever reasons I don’t know
>>>> vdsmd.service and libvirtd failed to start. I make sure that I was running
>>>> with EPT support enabled to achieve nested virtualization, but as I said:
>>>> it was a no go.
>>>>
>>>> So I’ve decommissioned a XenServer machine that was in production just
>>>> to try oVirt. The hardware is not new, but’s it’s very capable: Dual Xeon
>>>> E5506 with 48GB of system RAM, but I can’t get the hosted engine to work,
>>>> it always insults my hardware: --- Hosted Engine deployment failed: this
>>>> system is not reliable, please check the issue,fix and redeploy.
>>>>
>>>> It’s definitely a problem on the storage subsystem, the error is just
>>>> random, at this moment I’ve got:
>>>>
>>>> [ ERROR ] Failed to execute stage 'Misc configuration': [-32605] No
>>>> response for JSON-RPC StorageDomain.detach request.
>>>>
>>>> But on other tries it came up with something like this:
>>>>
>>>> No response for JSON-RPC Volume.getSize request.
>>>>
>>>> I was thinking that the problem was on the NFSv3 server on our FreeNAS
>>>> box, so I’ve changed to an iSCSI backend, but the problem continues.
>>>>
>>>>
>>>> Can't say anything about your issues without the logs but there's
>>>> nothing wrong with FreeNAS (FreeBSD) NFS, I've been running oVirt with
>>>> FreeBSD's NFS since oVirt 3.2 so...
>>>>
>>>
>>> Could you share your relevant exports configuration to make sure he's
>>> using something that you know works?
>>>
>>>
>>>>
>>>> "You're holding it wrong":) Sorry, I know you're frustrated but that's
>>>> what I can add to the conversation.
>>>>
>>>> /K
>>>>
>>>> This happens at the very end of the ovirt-hosted-engine-setup command,
>>>> which leads me to believe that’s an oVirt issue. The OVA was already copied
>>>> and deployed to the storage:
>>>>
>>>> [ INFO ] Starting vdsmd
>>>> [ INFO ] Creating Volume Group
>>>> [ INFO ] Creating Storage Domain
>>>> [ INFO ] Creating Storage Pool
>>>> [ INFO ] Connecting Storage Pool
>>>> [ INFO ] Verifying sanlock lockspace initialization
>>>> [ INFO ] Creating Image for 'hosted-engine.lockspace' ...
>>>> [ INFO ] Image for 'hosted-engine.lockspace' created successfully
>>>> [ INFO ] Creating Image for 'hosted-engine.metadata' ...
>>>> [ INFO ] Image for 'hosted-engine.metadata' created successfully
>>>> [ INFO ] Creating VM Image
>>>> [ INFO ] Extracting disk image from OVF archive (could take a few
>>>> minutes depending on archive size)
>>>> [ INFO ] Validating pre-allocated volume size
>>>> [ INFO ] Uploading volume to data domain (could take a few minutes
>>>> depending on archive size)
>>>> [ INFO ] Image successfully imported from OVF
>>>> [ ERROR ] Failed to execute stage 'Misc configuration': [-32605] No
>>>> response for JSON-RPC StorageDomain.detach request.
>>>> [ INFO ] Yum Performing yum transaction rollback
>>>> [ INFO ] Stage: Clean up
>>>> [ INFO ] Generating answer file '/var/lib/ovirt-hosted-engine-
>>>> setup/answers/answers-20170623032541.conf'
>>>> [ INFO ] Stage: Pre-termination
>>>> [ INFO ] Stage: Termination
>>>> [ ERROR ] Hosted Engine deployment failed: this system is not reliable,
>>>> please check the issue,fix and redeploy
>>>> Log file is located at /var/log/ovirt-hosted-engine-s
>>>> etup/ovirt-hosted-engine-setup-20170623023424-o9rbt0.log
>>>>
>>>> At this point I really don’t know what I should try. And the log file
>>>> is too verborragic (hoping this word exists) to look for errors.
>>>>
>>>> Any guidance?
>>>>
>>>> Thanks,
>>>> V.
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at ovirt.org
>>>> http://lists.ovirt.org/mailman/listinfo/users
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> Users at ovirt.org
>>>> http://lists.ovirt.org/mailman/listinfo/users
>>>>
>>>>
>>>
>>>
>>> --
>>> Adam Litke
>>>
>>>
>>>
>>> _______________________________________________
>>> Users mailing list
>>> Users at ovirt.org
>>> http://lists.ovirt.org/mailman/listinfo/users
>>>
>>>
>>
>
> _______________________________________________
> Users mailing list
> Users at ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
>


-- 

SANDRO BONAZZOLA

ASSOCIATE MANAGER, SOFTWARE ENGINEERING, EMEA ENG VIRTUALIZATION R&D

Red Hat EMEA <https://www.redhat.com/>
<https://red.ht/sig>
TRIED. TESTED. TRUSTED. <https://redhat.com/trusted>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/users/attachments/20170628/a1485a11/attachment.html>


More information about the Users mailing list