[release][stable][trove] Please add Lingxian Kong to trove-stable-maint group
Hi Stable Maintenance Core team, I'm current Trove PTL, could you please add myself to trove-stable-maint so that I could merge those patches[1][2][3] for stable/stein? [1] https://review.openstack.org/#/c/645647/ [2] https://review.openstack.org/#/c/645646/ [3] https://review.openstack.org/#/c/645648/ --- Cheers, Lingxian Kong Catalyst Cloud
On 4/4/2019 5:28 AM, Lingxian Kong wrote:
I'm current Trove PTL, could you please add myself to trove-stable-maint so that I could merge those patches[1][2][3] for stable/stein?
[1] https://review.openstack.org/#/c/645647/ [2] https://review.openstack.org/#/c/645646/ [3] https://review.openstack.org/#/c/645648/
I've approved the first two, the last one is on master. -- Thanks, Matt
---- On Thu, 04 Apr 2019 13:55:54 -0500 Matt Riedemann <mriedemos@gmail.com> wrote ----
On 4/4/2019 5:28 AM, Lingxian Kong wrote:
I'm current Trove PTL, could you please add myself to trove-stable-maint so that I could merge those patches[1][2][3] for stable/stein?
[1] https://review.openstack.org/#/c/645647/ [2] https://review.openstack.org/#/c/645646/ [3] https://review.openstack.org/#/c/645648/
I've approved the first two, the last one is on master.
Thanks, matt, there are more stable patches stuck on trove side [1] and lxkong current PTL does not have +2 authority. Can we add him as trove-stable-maint maintainer? [1] https://review.openstack.org/#/c/649744/ https://review.openstack.org/#/c/649745/ https://review.openstack.org/#/c/648884/ -gmann
--
Thanks,
Matt
On 4/4/2019 3:14 PM, Ghanshyam Mann wrote:
https://review.openstack.org/#/c/649744/ https://review.openstack.org/#/c/649745/
These are approved.
I rebased this into a series so ping me to approve when it's passing CI. -- Thanks, Matt
Thanks Matt, and two more please, they are aiming at fixing the CI issue for job migration to Ubuntu Bionic. After merged, I will rebase this one https://review.openstack.org/#/c/648884/ from Ghanshyam https://review.openstack.org/#/c/649744/ <https://review.openstack.org/#/c/649745/> https://review.openstack.org/#/c/649745/ --- Cheers, Lingxian Kong Catalyst Cloud On Fri, Apr 5, 2019 at 7:59 AM Matt Riedemann <mriedemos@gmail.com> wrote:
On 4/4/2019 5:28 AM, Lingxian Kong wrote:
I'm current Trove PTL, could you please add myself to trove-stable-maint so that I could merge those patches[1][2][3] for stable/stein?
[1] https://review.openstack.org/#/c/645647/ [2] https://review.openstack.org/#/c/645646/ [3] https://review.openstack.org/#/c/645648/
I've approved the first two, the last one is on master.
--
Thanks,
Matt
On Thu, Apr 04, 2019 at 11:28:48PM +1300, Lingxian Kong wrote:
Hi Stable Maintenance Core team,
I'm current Trove PTL, could you please add myself to trove-stable-maint so that I could merge those patches[1][2][3] for stable/stein?
Being PTL doesn't give you automatic rights to the stable branches as the requirements there are different. I'm really happy to start mentoring you on the stable policy and have you review stables patches then after 3months(ish) we can add you to the stable group assuming everything looks good. As an added bonus I expect we're mostly in the same TZ ;P Yours Tony.
Hi Tony, Thanks for the explanation of the stable branch policy, and very good to know we are in the similar TZ (guess you locate in Aus). It's not a hard requirement for me to become one of the stable branch core reviewers, the original demand was to merge Trove stable/stein patches in order to finish the community runtime requirement for Stein dev cycle. The patches were all sorted out thanks to Matt, so all good for now. --- Cheers, Lingxian Kong Catalyst Cloud On Wed, Apr 10, 2019 at 12:37 PM Tony Breeds <tony@bakeyournoodle.com> wrote:
On Thu, Apr 04, 2019 at 11:28:48PM +1300, Lingxian Kong wrote:
Hi Stable Maintenance Core team,
I'm current Trove PTL, could you please add myself to trove-stable-maint so that I could merge those patches[1][2][3] for stable/stein?
Being PTL doesn't give you automatic rights to the stable branches as the requirements there are different. I'm really happy to start mentoring you on the stable policy and have you review stables patches then after 3months(ish) we can add you to the stable group assuming everything looks good.
As an added bonus I expect we're mostly in the same TZ ;P
Yours Tony.
On Wed, Apr 10, 2019 at 12:54:03PM +1200, Lingxian Kong wrote:
Hi Tony,
Thanks for the explanation of the stable branch policy, and very good to know we are in the similar TZ (guess you locate in Aus).
Yup so not total overlap but some :)
It's not a hard requirement for me to become one of the stable branch core reviewers, the original demand was to merge Trove stable/stein patches in order to finish the community runtime requirement for Stein dev cycle.
The patches were all sorted out thanks to Matt, so all good for now.
Cool. As I've said before I'm happy to mentor you, which means you need to be doing stable reviews. It's a great idea when doign those reviews to articulate the aspects of the stable policy you evaluated. It doens't need to be verbose, somethign like: 'Clean backport, closes a user visible bug, low risk' ; or 'Clean backport except where described in the conflict section. Related to a bug No API impact' that sort of thing Yours Tony.
On 4/9/2019 7:37 PM, Tony Breeds wrote:
Being PTL doesn't give you automatic rights to the stable branches as the requirements there are different. I'm really happy to start mentoring you on the stable policy and have you review stables patches then after 3months(ish) we can add you to the stable group assuming everything looks good.
Maybe we should start a FAQ for this in the stable docs. This question comes up every release. BTW, this is a good example of why PTLs aren't automatically on the stable team: https://review.openstack.org/#/c/651672/ That change is extremely questionable as a backport. -- Thanks, Matt
Matt, I've already replied in the patch, since you also mentioned the patch here, i think it's necessary also reply you here. That patch failed to get attention from other team members and the community who are not interested in such change, as the PTL, i have to merge it myself to not block other patches. At the same time, some users were complaining the current trove image build approach, which could be improved with that patch. If self-approved is something you disagree, I'd appreciate if you could offer some help for the review. --- Cheers, Lingxian Kong Catalyst Cloud On Fri, Apr 12, 2019 at 5:11 AM Matt Riedemann <mriedemos@gmail.com> wrote:
On 4/9/2019 7:37 PM, Tony Breeds wrote:
Being PTL doesn't give you automatic rights to the stable branches as the requirements there are different. I'm really happy to start mentoring you on the stable policy and have you review stables patches then after 3months(ish) we can add you to the stable group assuming everything looks good.
Maybe we should start a FAQ for this in the stable docs. This question comes up every release.
BTW, this is a good example of why PTLs aren't automatically on the stable team:
https://review.openstack.org/#/c/651672/
That change is extremely questionable as a backport.
--
Thanks,
Matt
Sent from my iPhone
On Apr 11, 2019, at 4:47 PM, Lingxian Kong <anlin.kong@gmail.com> wrote:
Matt,
I've already replied in the patch, since you also mentioned the patch here, i think it's necessary also reply you here.
That patch failed to get attention from other team members and the community who are not interested in such change, as the PTL, i have to merge it myself to not block other patches. At the same time, some users were complaining the current trove image build approach, which could be improved with that patch.
If self-approved is something you disagree, I'd appreciate if you could offer some help for the review.
Should we perhaps look into the possibility of removing Trove from following stable policy by any chance? It’s not ideal but I feel if the projects needs to start getting overhauled and brought to a good place, we may need to do this at cost of its stability?
--- Cheers, Lingxian Kong Catalyst Cloud
On Fri, Apr 12, 2019 at 5:11 AM Matt Riedemann <mriedemos@gmail.com> wrote: On 4/9/2019 7:37 PM, Tony Breeds wrote:
Being PTL doesn't give you automatic rights to the stable branches as the requirements there are different. I'm really happy to start mentoring you on the stable policy and have you review stables patches then after 3months(ish) we can add you to the stable group assuming everything looks good.
Maybe we should start a FAQ for this in the stable docs. This question comes up every release.
BTW, this is a good example of why PTLs aren't automatically on the stable team:
https://review.openstack.org/#/c/651672/
That change is extremely questionable as a backport.
--
Thanks,
Matt
On Thu, Apr 11, 2019 at 08:00:16PM -0400, Mohammed Naser wrote:
Should we perhaps look into the possibility of removing Trove from following stable policy by any chance?
The change in question was published and merged by a single person in 3 days on master. It isn't a stable policy/tag thing. Yours Tony.
On Fri, Apr 12, 2019 at 1:55 PM Tony Breeds <tony@bakeyournoodle.com> wrote:
On Thu, Apr 11, 2019 at 08:00:16PM -0400, Mohammed Naser wrote:
Should we perhaps look into the possibility of removing Trove from following stable policy by any chance?
The change in question was published and merged by a single person in 3 days on master. It isn't a stable policy/tag thing.
So, how long do you think it's appropriate for the person who is mostly the only one cares about the improvement before merging the patch by himself?
---- On Thu, 11 Apr 2019 21:14:40 -0500 Lingxian Kong <anlin.kong@gmail.com> wrote ----
On Fri, Apr 12, 2019 at 1:55 PM Tony Breeds <tony@bakeyournoodle.com> wrote: On Thu, Apr 11, 2019 at 08:00:16PM -0400, Mohammed Naser wrote:
Should we perhaps look into the possibility of removing Trove from following stable policy by any chance?
The change in question was published and merged by a single person in 3 days on master. It isn't a stable policy/tag thing.
So, how long do you think it's appropriate for the person who is mostly the only one cares about the improvement before merging the patch by himself?
I think this is ok on master changes, there are lot of projects which has self approve culter due to no other core Active. Trove is lacking in term of active or Core members and Lingxian is trying his best to fill that gap. So IMO, doing self approve on the master is all ok as per current situation. -gmann
On 4/11/2019 3:47 PM, Lingxian Kong wrote:
I've already replied in the patch, since you also mentioned the patch here, i think it's necessary also reply you here.
That patch failed to get attention from other team members and the community who are not interested in such change, as the PTL, i have to merge it myself to not block other patches. At the same time, some users were complaining the current trove image build approach, which could be improved with that patch.
If self-approved is something you disagree, I'd appreciate if you could offer some help for the review.
I realize Trove is in a tough spot so I'm not trying to criticize and I understand the frustration of waiting to get something merged that you or your customers need. I don't have great answers for these problems. I was simply trying to highlight the fact that while that patch may be appropriate to merge on master it may not be appropriate to merge on stable, so there are different rules if you're following the stable policy and that's why cores on master aren't necessarily cores on stable branches. At the same time I'm coming from a place where we try to enforce some kind of standards across projects under openstack TC governance, and trove is in a tough spot there. But as Mohammed pointed out, if Trove elects to *not* follow the stable policy then we simply have to remove the "stable:follows-policy" tag from governance and trove is free to backport as it pleases (I think? Tony can correct me). -- Thanks, Matt
On Fri, Apr 12, 2019 at 2:27 PM Matt Riedemann <mriedemos@gmail.com> wrote:
On 4/11/2019 3:47 PM, Lingxian Kong wrote:
I've already replied in the patch, since you also mentioned the patch here, i think it's necessary also reply you here.
That patch failed to get attention from other team members and the community who are not interested in such change, as the PTL, i have to merge it myself to not block other patches. At the same time, some users were complaining the current trove image build approach, which could be improved with that patch.
If self-approved is something you disagree, I'd appreciate if you could offer some help for the review.
I realize Trove is in a tough spot so I'm not trying to criticize and I understand the frustration of waiting to get something merged that you or your customers need. I don't have great answers for these problems. I was simply trying to highlight the fact that while that patch may be appropriate to merge on master it may not be appropriate to merge on stable, so there are different rules if you're following the stable policy and that's why cores on master aren't necessarily cores on stable branches.
Totally understand, thanks for the explanation. That's why I explained in the above email that it's not my original requirement to become a stale core, the first email was just a heading up to ask help for the stable branch review.
At the same time I'm coming from a place where we try to enforce some kind of standards across projects under openstack TC governance, and trove is in a tough spot there.
But as Mohammed pointed out, if Trove elects to *not* follow the stable policy then we simply have to remove the "stable:follows-policy" tag from governance and trove is free to backport as it pleases (I think? Tony can correct me).
In the current situation, i prefer to follow the ways that people who care about the project(I mean, Trove) and are willing to offer help to fix things should have the permission to merge patches to stable. So, according to https://governance.openstack.org/tc/reference/tags/stable_follows-policy.htm..., as Trove PTL of Train dev cycle, can I ask to remove stable:follows-policy tag for Trove? --- Cheers, Lingxian Kong Catalyst Cloud Team
On 4/11/2019 9:39 PM, Lingxian Kong wrote:
In the current situation, i prefer to follow the ways that people who care about the project(I mean, Trove) and are willing to offer help to fix things should have the permission to merge patches to stable. So, according to https://governance.openstack.org/tc/reference/tags/stable_follows-policy.htm..., as Trove PTL of Train dev cycle, can I ask to remove stable:follows-policy tag for Trove?
It would probably be best if you propose a change to the governance repo to remove the tag from the trove project. -- Thanks, Matt
---- On Thu, 11 Apr 2019 21:39:45 -0500 Lingxian Kong <anlin.kong@gmail.com> wrote ----
On Fri, Apr 12, 2019 at 2:27 PM Matt Riedemann <mriedemos@gmail.com> wrote: On 4/11/2019 3:47 PM, Lingxian Kong wrote:
I've already replied in the patch, since you also mentioned the patch here, i think it's necessary also reply you here.
That patch failed to get attention from other team members and the community who are not interested in such change, as the PTL, i have to merge it myself to not block other patches. At the same time, some users were complaining the current trove image build approach, which could be improved with that patch.
If self-approved is something you disagree, I'd appreciate if you could offer some help for the review.
I realize Trove is in a tough spot so I'm not trying to criticize and I understand the frustration of waiting to get something merged that you or your customers need. I don't have great answers for these problems. I was simply trying to highlight the fact that while that patch may be appropriate to merge on master it may not be appropriate to merge on stable, so there are different rules if you're following the stable policy and that's why cores on master aren't necessarily cores on stable branches.
Totally understand, thanks for the explanation. That's why I explained in the above email that it's not my original requirement to become a stale core, the first email was just a heading up to ask help for the stable branch review.
At the same time I'm coming from a place where we try to enforce some kind of standards across projects under openstack TC governance, and trove is in a tough spot there.
But as Mohammed pointed out, if Trove elects to *not* follow the stable policy then we simply have to remove the "stable:follows-policy" tag from governance and trove is free to backport as it pleases (I think? Tony can correct me).
In the current situation, i prefer to follow the ways that people who care about the project(I mean, Trove) and are willing to offer help to fix things should have the permission to merge patches to stable. So, according to https://governance.openstack.org/tc/reference/tags/stable_follows-policy.htm..., as Trove PTL of Train dev cycle, can I ask to remove stable:follows-policy tag for Trove?---Cheers,Lingxian KongCatalyst Cloud Team
As Matt, Naser mentioned, you can do that anytime. But I am concerned that, adding the tag is always good but removing is definitely not. I want to understand, is this the only case based on which you are taking the decision to remove the "stable:follows-policy" tag ? Or you think trove will have/need more cases like that? -gmann
On Thu, Apr 11, 2019 at 09:24:33PM -0500, Matt Riedemann wrote:
But as Mohammed pointed out, if Trove elects to *not* follow the stable policy then we simply have to remove the "stable:follows-policy" tag from governance and trove is free to backport as it pleases (I think? Tony can correct me).
That is correct. Yours Tony.
On Thu, Apr 11, 2019 at 12:05:14PM -0500, Matt Riedemann wrote:
On 4/9/2019 7:37 PM, Tony Breeds wrote:
Being PTL doesn't give you automatic rights to the stable branches as the requirements there are different. I'm really happy to start mentoring you on the stable policy and have you review stables patches then after 3months(ish) we can add you to the stable group assuming everything looks good.
Maybe we should start a FAQ for this in the stable docs. This question comes up every release.
I'm working on it. I think this can be added to the project-teams-guide as it'd generally applicable.
BTW, this is a good example of why PTLs aren't automatically on the stable team:
https://review.openstack.org/#/c/651672/
That change is extremely questionable as a backport.
Agreeed. Yours Tony.
participants (5)
-
Ghanshyam Mann
-
Lingxian Kong
-
Matt Riedemann
-
Mohammed Naser
-
Tony Breeds