Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Martin Perina
Hi,

unfortunately we have a new blocker [1] related to upgrade from 4.1.z, but
we will try to do our best to post a fix before 10:00 today.

Martin

[1] https://bugzilla.redhat.com/show_bug.cgi?id=1519301




On Thu, Nov 30, 2017 at 8:49 AM, Sandro Bonazzola 
wrote:

> Hi,
> we still have 4 open acknowledged blockers according to
> https://bugzilla.redhat.com/buglist.cgi?quicksearch=
> flag%3Ablocker%2B%20target_milestone%3Aovirt-4.2.0%
> 20status%3Anew%2Cassigned%2Cpost
>
> Bug ID Product Assignee Status Summary
> 1518887 
> ovirt-hosted-engine-ha b...@ovirt.org NEW ovirt-ha-agent fails parsing
> the OVF_STORE due to a change in OVF namespace URI
> 1516113 
> cockpit-ovirt phbai...@redhat.com POST Deploy the HostedEngine failed
> with the default CPU type
> 1518693  ovirt-engine
> akrej...@redhat.com POST Quota is needed to copy template disk
> 1507277  ovirt-engine
> era...@redhat.com POST [RFE][DR] - Vnic Profiles mapping in VMs register
> from data storage domain should be supported also for templates
>
> There are also 3 proposed blockers that need either to be acknowledged or
> rejected: https://bugzilla.redhat.com/buglist.cgi?
> quicksearch=flag%3Ablocker%3F%20target_milestone%3Aovirt-4.
> 2.0%20status%3Anew%2Cassigned%2Cpost
>
> Bug ID Product Assignee Summary
> 1450061  Red Hat
> Enterprise Virtualization Manager rh-spice-b...@redhat.com Copy-paste:
> filename encoding in Win guest
> 1517810  ovirt-engine
> stira...@redhat.com Adding additional ha-host fails.
> 1502920  Red Hat
> Enterprise Virtualization Manager rba...@redhat.com File missing after
> upgrade of RHVH node from version RHVH-4.1-20170925.0 to latest.
>
>
> Given current status I would reschedule oVirt 4.2.0 RC to next week,
> tentatively on Monday.
>
> --
>
> SANDRO BONAZZOLA
>
> ASSOCIATE MANAGER, SOFTWARE ENGINEERING, EMEA ENG VIRTUALIZATION R
>
> Red Hat EMEA 
> 
> TRIED. TESTED. TRUSTED. 
>
>
> ___
> Devel mailing list
> Devel@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/devel
>



-- 
Martin Perina
Associate Manager, Software Engineering
Red Hat Czech s.r.o.
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Martin Sivak
>
>> 1517810 
>>> ovirt-engine stira...@redhat.com Adding additional ha-host fails.
>>>
>>
> This one has a verified engine fix now, we just need to merge it and
> update Node 0 setup (also verified).
>

Merged.
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Martin Sivak
On Thu, Nov 30, 2017 at 11:28 AM, Martin Sivak  wrote:

> I think a reasonable alternative is to revert the change, adjust HE to use
>> both formats, and then introduce the new format. The first step (revert)
>> should be done for 4.2.0 (today?) and the rest for 4.2.1.
>> Y.
>>
>>
> The format is still the same, we just needed to autodetect the used NS and
> then query for attributes using the detected namespace. Patch posted, tests
> are passing, but since we want to be 100% sure we will test with OST and
> Node 0 before pushing it. Should not take long.
>


Verified and Merged.


> Martin
>
>
>
>>
>>>


>
> 1516113 
>> cockpit-ovirt phbai...@redhat.com POST Deploy the HostedEngine
>> failed with the default CPU type
>>
>
> Would be happy if the remaining patch could get reviewed quickly.
>




>
> 1518693 
>> ovirt-engine akrej...@redhat.com POST Quota is needed to copy
>> template disk
>>
>
> This is only via REST and the default quota can be used as a
> workaround - why is this a blocker?
>

 Automation added it because Raz marked it as Regression. But the change
 was intentional.



> 1517810 
>> ovirt-engine stira...@redhat.com Adding additional ha-host fails.
>>
>
 This one has a verified engine fix now, we just need to merge it and
 update Node 0 setup (also verified).


 Martin


 ___
 Devel mailing list
 Devel@ovirt.org
 http://lists.ovirt.org/mailman/listinfo/devel

>>>
>>>
>>
>
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Martin Sivak
>
> I think a reasonable alternative is to revert the change, adjust HE to use
> both formats, and then introduce the new format. The first step (revert)
> should be done for 4.2.0 (today?) and the rest for 4.2.1.
> Y.
>
>
The format is still the same, we just needed to autodetect the used NS and
then query for attributes using the detected namespace. Patch posted, tests
are passing, but since we want to be 100% sure we will test with OST and
Node 0 before pushing it. Should not take long.

Martin



>
>>
>>>
>>>

 1516113 
> cockpit-ovirt phbai...@redhat.com POST Deploy the HostedEngine failed
> with the default CPU type
>

 Would be happy if the remaining patch could get reviewed quickly.

>>>
>>>
>>>
>>>

 1518693 
> ovirt-engine akrej...@redhat.com POST Quota is needed to copy
> template disk
>

 This is only via REST and the default quota can be used as a workaround
 - why is this a blocker?

>>>
>>> Automation added it because Raz marked it as Regression. But the change
>>> was intentional.
>>>
>>>
>>>
 1517810 
> ovirt-engine stira...@redhat.com Adding additional ha-host fails.
>

>>> This one has a verified engine fix now, we just need to merge it and
>>> update Node 0 setup (also verified).
>>>
>>>
>>> Martin
>>>
>>>
>>> ___
>>> Devel mailing list
>>> Devel@ovirt.org
>>> http://lists.ovirt.org/mailman/listinfo/devel
>>>
>>
>>
>
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Michal Skrivanek

> On 30 Nov 2017, at 05:13, Yaniv Kaul  wrote:
> 
> 
> 
> On Thu, Nov 30, 2017 at 11:38 AM, Arik Hadas  > wrote:
> 
> 
> On Thu, Nov 30, 2017 at 11:10 AM, Martin Sivak  > wrote:
> 
> 
> Bug IDProduct AssigneeStatus  Summary
> 1518887  
> ovirt-hosted-engine-ha  b...@ovirt.org   NEW 
> ovirt-ha-agent fails parsing the OVF_STORE due to a change in OVF namespace 
> URI
> 
> I'm in favor of reverting the virt change personally.
> 
> Unless something else depends on it, the commit message said vdsm needs this.
> 
> vdsm needs it because we would like vdsm to be able to parse OVA files that 
> are generated by us and OVA files that  are generated by others (and are 
> VMware-compatbile) with the existing code in vdsm. The existing code is 
> tailored to VMware-compatible OVA files that are generated by others, in 
> which the uri doesn't include that slash at the end.
> 
> It would be best to adjust ovirt-ha-agent to parse the right uri.
> However, it that's too complicated, an alternative solution is to keep 
> writing the previous uri to ovirt's OVFs, i.e., those in OVF_STORE and in 
> snapshot's configuration. That would be a pitty since we want to minimize the 
> differences between the OVFs we generate, but it would be better than 
> reverting the change..
> 
> I think a reasonable alternative is to revert the change, adjust HE to use 
> both formats, and then introduce the new format. The first step (revert) 
> should be done for 4.2.0 (today?) and the rest for 4.2.1.

It shouldn’t take too long to polish [1] and rebuild ovirt-hosted-engine-ha
Thanks Martin for posting it!

Thanks,
michal

[1] https://gerrit.ovirt.org/#/c/84933/

> Y.
> 
>  
>  
> 
> 1516113  cockpit-ovirt   
> phbai...@redhat.com POSTDeploy the 
> HostedEngine failed with the default CPU type
> 
> Would be happy if the remaining patch could get reviewed quickly.
> 
> 
>  
> 
> 1518693  ovirt-engine
> akrej...@redhat.com POSTQuota is 
> needed to copy template disk
> 
> This is only via REST and the default quota can be used as a workaround - why 
> is this a blocker?
> 
> Automation added it because Raz marked it as Regression. But the change was 
> intentional.
>  
>  
> 1517810  ovirt-engine
> stira...@redhat.com Adding additional 
> ha-host fails.
> 
> This one has a verified engine fix now, we just need to merge it and update 
> Node 0 setup (also verified).
> 
> 
> Martin
> 
> 
> ___
> Devel mailing list
> Devel@ovirt.org 
> http://lists.ovirt.org/mailman/listinfo/devel 
> 
> 
> 
> ___
> Devel mailing list
> Devel@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/devel

___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Yaniv Kaul
On Thu, Nov 30, 2017 at 11:38 AM, Arik Hadas  wrote:

>
>
> On Thu, Nov 30, 2017 at 11:10 AM, Martin Sivak  wrote:
>
>>

 Bug ID Product Assignee Status Summary
 1518887 
 ovirt-hosted-engine-ha b...@ovirt.org NEW ovirt-ha-agent fails parsing
 the OVF_STORE due to a change in OVF namespace URI

>>>
>>> I'm in favor of reverting the virt change personally.
>>>
>>
>> Unless something else depends on it, the commit message said vdsm needs
>> this.
>>
>
> vdsm needs it because we would like vdsm to be able to parse OVA files
> that are generated by us and OVA files that  are generated by others (and
> are VMware-compatbile) with the existing code in vdsm. The existing code is
> tailored to VMware-compatible OVA files that are generated by others, in
> which the uri doesn't include that slash at the end.
>
> It would be best to adjust ovirt-ha-agent to parse the right uri.
> However, it that's too complicated, an alternative solution is to keep
> writing the previous uri to ovirt's OVFs, i.e., those in OVF_STORE and in
> snapshot's configuration. That would be a pitty since we want to minimize
> the differences between the OVFs we generate, but it would be better than
> reverting the change..
>

I think a reasonable alternative is to revert the change, adjust HE to use
both formats, and then introduce the new format. The first step (revert)
should be done for 4.2.0 (today?) and the rest for 4.2.1.
Y.


>
>>
>>
>>>
>>> 1516113 
 cockpit-ovirt phbai...@redhat.com POST Deploy the HostedEngine failed
 with the default CPU type

>>>
>>> Would be happy if the remaining patch could get reviewed quickly.
>>>
>>
>>
>>
>>
>>>
>>> 1518693 
 ovirt-engine akrej...@redhat.com POST Quota is needed to copy template
 disk

>>>
>>> This is only via REST and the default quota can be used as a workaround
>>> - why is this a blocker?
>>>
>>
>> Automation added it because Raz marked it as Regression. But the change
>> was intentional.
>>
>>
>>
>>> 1517810 
 ovirt-engine stira...@redhat.com Adding additional ha-host fails.

>>>
>> This one has a verified engine fix now, we just need to merge it and
>> update Node 0 setup (also verified).
>>
>>
>> Martin
>>
>>
>> ___
>> Devel mailing list
>> Devel@ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/devel
>>
>
>
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Arik Hadas
On Thu, Nov 30, 2017 at 11:10 AM, Martin Sivak  wrote:

>
>>>
>>> Bug ID Product Assignee Status Summary
>>> 1518887 
>>> ovirt-hosted-engine-ha b...@ovirt.org NEW ovirt-ha-agent fails parsing
>>> the OVF_STORE due to a change in OVF namespace URI
>>>
>>
>> I'm in favor of reverting the virt change personally.
>>
>
> Unless something else depends on it, the commit message said vdsm needs
> this.
>

vdsm needs it because we would like vdsm to be able to parse OVA files that
are generated by us and OVA files that  are generated by others (and are
VMware-compatbile) with the existing code in vdsm. The existing code is
tailored to VMware-compatible OVA files that are generated by others, in
which the uri doesn't include that slash at the end.

It would be best to adjust ovirt-ha-agent to parse the right uri.
However, it that's too complicated, an alternative solution is to keep
writing the previous uri to ovirt's OVFs, i.e., those in OVF_STORE and in
snapshot's configuration. That would be a pitty since we want to minimize
the differences between the OVFs we generate, but it would be better than
reverting the change..


>
>
>>
>> 1516113 
>>> cockpit-ovirt phbai...@redhat.com POST Deploy the HostedEngine failed
>>> with the default CPU type
>>>
>>
>> Would be happy if the remaining patch could get reviewed quickly.
>>
>
>
>
>
>>
>> 1518693 
>>> ovirt-engine akrej...@redhat.com POST Quota is needed to copy template
>>> disk
>>>
>>
>> This is only via REST and the default quota can be used as a workaround -
>> why is this a blocker?
>>
>
> Automation added it because Raz marked it as Regression. But the change
> was intentional.
>
>
>
>> 1517810 
>>> ovirt-engine stira...@redhat.com Adding additional ha-host fails.
>>>
>>
> This one has a verified engine fix now, we just need to merge it and
> update Node 0 setup (also verified).
>
>
> Martin
>
>
> ___
> Devel mailing list
> Devel@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/devel
>
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Martin Sivak
>
>
>>
>> Bug ID Product Assignee Status Summary
>> 1518887 
>> ovirt-hosted-engine-ha b...@ovirt.org NEW ovirt-ha-agent fails parsing
>> the OVF_STORE due to a change in OVF namespace URI
>>
>
> I'm in favor of reverting the virt change personally.
>

Unless something else depends on it, the commit message said vdsm needs
this.


>
> 1516113 
>> cockpit-ovirt phbai...@redhat.com POST Deploy the HostedEngine failed
>> with the default CPU type
>>
>
> Would be happy if the remaining patch could get reviewed quickly.
>




>
> 1518693  ovirt-engine
>> akrej...@redhat.com POST Quota is needed to copy template disk
>>
>
> This is only via REST and the default quota can be used as a workaround -
> why is this a blocker?
>

Automation added it because Raz marked it as Regression. But the change was
intentional.



> 1517810  ovirt-engine
>> stira...@redhat.com Adding additional ha-host fails.
>>
>
This one has a verified engine fix now, we just need to merge it and update
Node 0 setup (also verified).


Martin
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel

Re: [ovirt-devel] oVirt 4.2.0 blockers review - Day 3

2017-11-30 Thread Yaniv Kaul
On Thu, Nov 30, 2017 at 9:49 AM, Sandro Bonazzola 
wrote:

> Hi,
> we still have 4 open acknowledged blockers according to
> https://bugzilla.redhat.com/buglist.cgi?quicksearch=
> flag%3Ablocker%2B%20target_milestone%3Aovirt-4.2.0%
> 20status%3Anew%2Cassigned%2Cpost
>
> Bug ID Product Assignee Status Summary
> 1518887 
> ovirt-hosted-engine-ha b...@ovirt.org NEW ovirt-ha-agent fails parsing
> the OVF_STORE due to a change in OVF namespace URI
>

I'm in favor of reverting the virt change personally.

1516113  cockpit-ovirt
> phbai...@redhat.com POST Deploy the HostedEngine failed with the default
> CPU type
>

Would be happy if the remaining patch could get reviewed quickly.

1518693  ovirt-engine
> akrej...@redhat.com POST Quota is needed to copy template disk
>

This is only via REST and the default quota can be used as a workaround -
why is this a blocker?

1507277  ovirt-engine
> era...@redhat.com POST [RFE][DR] - Vnic Profiles mapping in VMs register
> from data storage domain should be supported also for templates
>

I'm not sure why it's a blocker for 4.2.0.


> There are also 3 proposed blockers that need either to be acknowledged or
> rejected: https://bugzilla.redhat.com/buglist.cgi?
> quicksearch=flag%3Ablocker%3F%20target_milestone%3Aovirt-4.
> 2.0%20status%3Anew%2Cassigned%2Cpost
>
> Bug ID Product Assignee Summary
> 1450061  Red Hat
> Enterprise Virtualization Manager rh-spice-b...@redhat.com Copy-paste:
> filename encoding in Win guest
>

Moved to 4.2.1. The Spice team did not touch it for quite some time since
it was opened.

1517810  ovirt-engine
> stira...@redhat.com Adding additional ha-host fails.
> 1502920  Red Hat
> Enterprise Virtualization Manager rba...@redhat.com File missing after
> upgrade of RHVH node from version RHVH-4.1-20170925.0 to latest.
>

Stuck on POST since2017-10-20 00:47:07 IDT ?


>
> Given current status I would reschedule oVirt 4.2.0 RC to next week,
> tentatively on Monday.
>

I'd review the status tomorrow as well - perhaps we can build the RC
tomorrow. I don't see too many blockers that cannot be completed today
above.
Y.

>
> --
>
> SANDRO BONAZZOLA
>
> ASSOCIATE MANAGER, SOFTWARE ENGINEERING, EMEA ENG VIRTUALIZATION R
>
> Red Hat EMEA 
> 
> TRIED. TESTED. TRUSTED. 
>
>
> ___
> Devel mailing list
> Devel@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/devel
>
___
Devel mailing list
Devel@ovirt.org
http://lists.ovirt.org/mailman/listinfo/devel