releasing hammer v0.94.7 ?

releasing hammer v0.94.7 ?

[Loic Dachary]

Hi Sage,

Here are the issues that would be in hammer v0.94.7 if we decided to release it now: http://tracker.ceph.com/versions/511
And here is the backport queue : http://tracker.ceph.com/projects/ceph/issues?query_id=78

Do you think it is time to cut the release and ask QE to start testing ?

Cheers

-- 
Loïc Dachary, Artisan Logiciel Libre
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: releasing hammer v0.94.7 ?

[Wido den Hollander]

> Op 7 april 2016 om 16:36 schreef Loic Dachary <loic@dachary.org>:
> 
> 
> Hi Sage,
> 
> Here are the issues that would be in hammer v0.94.7 if we decided to release
> it now: http://tracker.ceph.com/versions/511
> And here is the backport queue :
> http://tracker.ceph.com/projects/ceph/issues?query_id=78
> 
> Do you think it is time to cut the release and ask QE to start testing ?
> 

As this might be the last Hammer release I wonder if we could fix this bug:
http://tracker.ceph.com/issues/15348

I already submitted a PR for it: https://github.com/ceph/ceph/pull/8441

I backported it on a production cluster to Hammer and it works fine. Might be
something we can squeeze in?

Wido

> Cheers
> 
> -- 
> Loïc Dachary, Artisan Logiciel Libre
> --
> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: releasing hammer v0.94.7 ?

[Loic Dachary]

Hi Wido,

On 08/04/2016 08:35, Wido den Hollander wrote:
> 
>> Op 7 april 2016 om 16:36 schreef Loic Dachary <loic@dachary.org>:
>>
>>
>> Hi Sage,
>>
>> Here are the issues that would be in hammer v0.94.7 if we decided to release
>> it now: http://tracker.ceph.com/versions/511
>> And here is the backport queue :
>> http://tracker.ceph.com/projects/ceph/issues?query_id=78
>>
>> Do you think it is time to cut the release and ask QE to start testing ?
>>
> 
> As this might be the last Hammer release I wonder if we could fix this bug:
> http://tracker.ceph.com/issues/15348

This will not be the last hammer release. 

Once jewel is released, people will want to upgrade from hammer to jewel. This will uncover some problems (hopefully minor) that will need fixing and require a new hammer release. When jewel matures, more and more people will migrate to jewel: this will take a few months. Eventually, as it happened with firefly, the number of hammer clusters in production will be reduced and the bug reports regarding upgrade and operations will be reduced in proportion. At some point there won't be much incentive to publish a new hammer release and they will be less and less frequent. Until it's decided to just stop. In http://docs.ceph.com/docs/master/releases/ I estimated it will happen around november this year, give or take a few months.

> I already submitted a PR for it: https://github.com/ceph/ceph/pull/8441
> 
> I backported it on a production cluster to Hammer and it works fine. Might be
> something we can squeeze in?

This is something you can certainly discuss with Yehuda :-)

Cheers

-- 
Loïc Dachary, Artisan Logiciel Libre
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: releasing hammer v0.94.7 ?

[Sage Weil]

On Fri, 8 Apr 2016, Wido den Hollander wrote:
> > Op 7 april 2016 om 16:36 schreef Loic Dachary <loic@dachary.org>:
> > 
> > Hi Sage,
> > 
> > Here are the issues that would be in hammer v0.94.7 if we decided to release
> > it now: http://tracker.ceph.com/versions/511
> > And here is the backport queue :
> > http://tracker.ceph.com/projects/ceph/issues?query_id=78
> > 
> > Do you think it is time to cut the release and ask QE to start testing ?
> > 
> 
> As this might be the last Hammer release I wonder if we could fix this bug:
> http://tracker.ceph.com/issues/15348

Don't worry, it won't be the last one--we have at least another half 
year left.

> I already submitted a PR for it: https://github.com/ceph/ceph/pull/8441
> 
> I backported it on a production cluster to Hammer and it works fine. Might be
> something we can squeeze in?

Can you add a test to s3-tests for this?

sage

Re: releasing hammer v0.94.7 ?

[Sage Weil]

On Thu, 7 Apr 2016, Loic Dachary wrote:
> Hi Sage,
> 
> Here are the issues that would be in hammer v0.94.7 if we decided to release it now: http://tracker.ceph.com/versions/511
> And here is the backport queue : http://tracker.ceph.com/projects/ceph/issues?query_id=78
> 
> Do you think it is time to cut the release and ask QE to start testing ?

Let's do it!

s

Re: releasing hammer v0.94.7 ?

[Wido den Hollander]

> Op 8 april 2016 om 14:43 schreef Sage Weil <sweil@redhat.com>:
> 
> 
> On Fri, 8 Apr 2016, Wido den Hollander wrote:
> > > Op 7 april 2016 om 16:36 schreef Loic Dachary <loic@dachary.org>:
> > > 
> > > Hi Sage,
> > > 
> > > Here are the issues that would be in hammer v0.94.7 if we decided to
> > > release
> > > it now: http://tracker.ceph.com/versions/511
> > > And here is the backport queue :
> > > http://tracker.ceph.com/projects/ceph/issues?query_id=78
> > > 
> > > Do you think it is time to cut the release and ask QE to start testing ?
> > > 
> > 
> > As this might be the last Hammer release I wonder if we could fix this bug:
> > http://tracker.ceph.com/issues/15348
> 
> Don't worry, it won't be the last one--we have at least another half 
> year left.
> 

Ok, great! In the meantime I created a PR for Hammer:
https://github.com/ceph/ceph/pull/8528

> > I already submitted a PR for it: https://github.com/ceph/ceph/pull/8441
> > 
> > I backported it on a production cluster to Hammer and it works fine. Might
> > be
> > something we can squeeze in?
> 
> Can you add a test to s3-tests for this?
> 

Good one. Never touched those before. I'll look into that.

Wido

> sage