[openstack-dev] [Heat] Convergence status

2016-09-06 Thread Thomas Herve
Hi all,

We're approaching release time, and there are still some concerns
whether or not we can make convergence the default. In particular we
have a somewhat longstanding issue with canceling creation [1]. The
good news is that this is almost fixed [2], and update cancel is
almost there too [3]. We expect to see them merged by RC1 next week,
so please review those if you can.

I also have some confidence overall as projects like Magnum and Sahara
have been running tests with convergence enabled for about 3 months
now, and no particular issues have been reported about it.

That said, if we can't manage to get those in by RC1, we'll have to
revert back to legacy. Those behaviors are too important to release
without them. Hopefully, we'll make it work, and we can focus on
making TripleO using convergence during the next cycle.

Thanks,

-- 
Thomas


[1] https://bugs.launchpad.net/heat/+bug/1592374
[2] https://review.openstack.org/#/c/354000/
[3] https://review.openstack.org/#/c/362346/

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev


[openstack-dev] [heat] Convergence status

2016-01-18 Thread Anant Patil

Hi,

This is to discuss the status of convergence patches and plans for
making it default.

The convergence gate jobs have been running successfully since more than
a month now. There were three integration tests skipped:
1. StackValidationTest
2. UpdateStackTest.test_stack_update_alias_type
3. UpdateStackTest.test_stack_update_alias_changes

2 and 3 of above are addressed by patches and have been running
successfully:
https://review.openstack.org/#/c/248676/
https://review.openstack.org/#/c/259865/

The StackValidationTest fails because the test uses an image without
cfn-tools and convergence heat will wait for the signal to arrive.
https://bugs.launchpad.net/heat/+bug/1486281/comments/3 . This should be
fixed when we address https://bugs.launchpad.net/heat/+bug/1533176 . The
delete request can cancel the currently running check-resource requests
and it can then proceed without having to wait for resources to
complete. Also, if we use a proper image, this is is not seen.

There are few important patches in review which should close most of the
bugs:
https://review.openstack.org/#/c/264748/
https://review.openstack.org/#/c/262374/
https://review.openstack.org/#/c/261208/
https://review.openstack.org/#/c/264675/
https://review.openstack.org/#/c/208790/

There are other bugs for which the patches should land soon. We should
plan to make convergence default earlier in m-3 phase so that we can
thoroughly test it before release. Let me know your opinion on this.

- Anant

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev


Re: [openstack-dev] [heat] Convergence status

2016-01-18 Thread Sergey Kraynev
Anant, thank you for the summary.

I planned to announce it (enabling convergence by default) on the next week
(after m-2).
I think, that we need to land two first mentioned patches before it ^.


On 19 January 2016 at 09:44, Anant Patil  wrote:

>
> Hi,
>
> This is to discuss the status of convergence patches and plans for
> making it default.
>
> The convergence gate jobs have been running successfully since more than
> a month now. There were three integration tests skipped:
> 1. StackValidationTest
> 2. UpdateStackTest.test_stack_update_alias_type
> 3. UpdateStackTest.test_stack_update_alias_changes
>
> 2 and 3 of above are addressed by patches and have been running
> successfully:
> https://review.openstack.org/#/c/248676/
> https://review.openstack.org/#/c/259865/
>
> The StackValidationTest fails because the test uses an image without
> cfn-tools and convergence heat will wait for the signal to arrive.
> https://bugs.launchpad.net/heat/+bug/1486281/comments/3 . This should be
> fixed when we address https://bugs.launchpad.net/heat/+bug/1533176 . The
> delete request can cancel the currently running check-resource requests
> and it can then proceed without having to wait for resources to
> complete. Also, if we use a proper image, this is is not seen.
>
> There are few important patches in review which should close most of the
> bugs:
> https://review.openstack.org/#/c/264748/
> https://review.openstack.org/#/c/262374/
> https://review.openstack.org/#/c/261208/
> https://review.openstack.org/#/c/264675/
> https://review.openstack.org/#/c/208790/
>
> There are other bugs for which the patches should land soon. We should
> plan to make convergence default earlier in m-3 phase so that we can
> thoroughly test it before release. Let me know your opinion on this.
>
> - Anant
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>



-- 
Regards,
Sergey.
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev