Nightly RPM and tests have been broken for around 2 weeks now. This morning
a bit of a regression was merged to foreman core to fix the breaking RPM
aspect and I can report that nightly RPMs are now building. However, this
leads to a breakage in plugin asset usage with the newer React components.
To potentially address this I have opened [1] for testing and input. As
part of the original breakage, I added to test_develop running npm install
and webpack compile the same way our RPMs do in order to catch these sort
of issues earlier.
The second half is that after RPMs were built, repoclosure on the test
pipeline is currently failing [2]. The highlight being:
nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
need to add both of these as top level packages carried in
foreman-packaging. Can anyone confirm or deny that is how this should be
working? Or should another package we build be providing these?
>Nightly RPM and tests have been broken for around 2 weeks now. This morning
>a bit of a regression was merged to foreman core to fix the breaking RPM
>aspect and I can report that nightly RPMs are now building. However, this
>leads to a breakage in plugin asset usage with the newer React components.
>To potentially address this I have opened [1] for testing and input. As
>part of the original breakage, I added to test_develop running npm install
>and webpack compile the same way our RPMs do in order to catch these sort
>of issues earlier.
Thanks for this!
>The second half is that after RPMs were built, repoclosure on the test
>pipeline is currently failing [2]. The highlight being:
>
>20:10:35 package: nodejs-react-dom-15.6.2-1.el7.noarch from
>undertest-yum_el7-4203183943-6820:10:35 unresolved deps:
>20:10:35 npm(object-assign) >= 0:4.1.020:10:35
>npm(loose-envify) < 0:220:10:35 npm(loose-envify) >= 0:1.1.0
>
>
>
>nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
>need to add both of these as top level packages carried in
>foreman-packaging. Can anyone confirm or deny that is how this should be
>working? Or should another package we build be providing these?
I do believe we need to package this ourselves since we need a newer
version than in EPEL.
···
On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
Nightly RPM and tests have been broken for around 2 weeks now. This
morning
a bit of a regression was merged to foreman core to fix the breaking RPM
aspect and I can report that nightly RPMs are now building. However, this
leads to a breakage in plugin asset usage with the newer React components.
To potentially address this I have opened [1] for testing and input. As
part of the original breakage, I added to test_develop running npm install
and webpack compile the same way our RPMs do in order to catch these sort
of issues earlier.
Thanks for this!
The second half is that after RPMs were built, repoclosure on the test
pipeline is currently failing [2]. The highlight being:
nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
need to add both of these as top level packages carried in
foreman-packaging. Can anyone confirm or deny that is how this should be
working? Or should another package we build be providing these?
I do believe we need to package this ourselves since we need a newer
version than in EPEL.
Why dont we ask the maintainer to pkg a new version or someone offer to
become a co-maintainer and get a new version into EPEL ?
···
On Mon, Oct 16, 2017 at 9:31 PM, Ewoud Kohl van Wijngaarden < ewoud@kohlvanwijngaarden.nl> wrote:
On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
Nightly RPM and tests have been broken for around 2 weeks now. This
morning
a bit of a regression was merged to foreman core to fix the breaking RPM
aspect and I can report that nightly RPMs are now building. However, this
leads to a breakage in plugin asset usage with the newer React components.
To potentially address this I have opened [1] for testing and input. As
part of the original breakage, I added to test_develop running npm install
and webpack compile the same way our RPMs do in order to catch these sort
of issues earlier.
Thanks for this!
The second half is that after RPMs were built, repoclosure on the test
pipeline is currently failing [2]. The highlight being:
nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
need to add both of these as top level packages carried in
foreman-packaging. Can anyone confirm or deny that is how this should be
working? Or should another package we build be providing these?
I do believe we need to package this ourselves since we need a newer
version than in EPEL.
EPEL is not great place to be for Rails or Node components. You should
not bump versions in EPEL7 (major relase should go into EPEL8).
···
On Tue, Oct 17, 2017 at 12:01 AM, Eric D Helms wrote:
>
>
> On Oct 16, 2017 5:17 PM, "Sean O'Keeffe" wrote:
>
> Why dont we ask the maintainer to pkg a new version or someone offer to
> become a co-maintainer and get a new version into EPEL ?
>
>
> While I think this is the right open source path, I'd weigh:
>
> 1) how long will nighties be broken waiting on a new package?
> 2) 2.0 to 4.1 is a large jump and as a base dependency other EPEL packages
> may not work.
>
>
>
> On Mon, Oct 16, 2017 at 9:31 PM, Ewoud Kohl van Wijngaarden > wrote:
>>
>> On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
>>>
>>> Nightly RPM and tests have been broken for around 2 weeks now. This
>>> morning
>>> a bit of a regression was merged to foreman core to fix the breaking RPM
>>> aspect and I can report that nightly RPMs are now building. However, this
>>> leads to a breakage in plugin asset usage with the newer React
>>> components.
>>> To potentially address this I have opened [1] for testing and input. As
>>> part of the original breakage, I added to test_develop running npm
>>> install
>>> and webpack compile the same way our RPMs do in order to catch these sort
>>> of issues earlier.
>>
>>
>> Thanks for this!
>>
>>> The second half is that after RPMs were built, repoclosure on the test
>>> pipeline is currently failing [2]. The highlight being:
>>>
>>> *20:10:35* package: nodejs-react-dom-15.6.2-1.el7.noarch from
>>> undertest-yum_el7-4203183943-68*20:10:35* unresolved deps:
>>> *20:10:35* npm(object-assign) >= 0:4.1.0*20:10:35*
>>> npm(loose-envify) < 0:2*20:10:35* npm(loose-envify) >= 0:1.1.0
>>>
>>>
>>>
>>> nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
>>> need to add both of these as top level packages carried in
>>> foreman-packaging. Can anyone confirm or deny that is how this should be
>>> working? Or should another package we build be providing these?
>>
>>
>> I do believe we need to package this ourselves since we need a newer
>> version than in EPEL.
>>
>>> [1] https://github.com/theforeman/foreman/pull/4924
>>> [2] http://ci.theforeman.org/job/packaging_repoclosure/37110/console
>>
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "foreman-dev" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to foreman-dev+unsubscribe@googlegroups.com.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "foreman-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to foreman-dev+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "foreman-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to foreman-dev+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
···
On Tue, Oct 17, 2017 at 09:16:52AM +0200, Lukas Zapletal wrote:
>EPEL is not great place to be for Rails or Node components. You should
>not bump versions in EPEL7 (major relase should go into EPEL8).
>
>On Tue, Oct 17, 2017 at 12:01 AM, Eric D Helms wrote:
>>
>>
>> On Oct 16, 2017 5:17 PM, "Sean O'Keeffe" wrote:
>>
>> Why dont we ask the maintainer to pkg a new version or someone offer to
>> become a co-maintainer and get a new version into EPEL ?
>>
>>
>> While I think this is the right open source path, I'd weigh:
>>
>> 1) how long will nighties be broken waiting on a new package?
>> 2) 2.0 to 4.1 is a large jump and as a base dependency other EPEL packages
>> may not work.
>>
>>
>>
>> On Mon, Oct 16, 2017 at 9:31 PM, Ewoud Kohl van Wijngaarden >> wrote:
>>>
>>> On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
>>>>
>>>> Nightly RPM and tests have been broken for around 2 weeks now. This
>>>> morning
>>>> a bit of a regression was merged to foreman core to fix the breaking RPM
>>>> aspect and I can report that nightly RPMs are now building. However, this
>>>> leads to a breakage in plugin asset usage with the newer React
>>>> components.
>>>> To potentially address this I have opened [1] for testing and input. As
>>>> part of the original breakage, I added to test_develop running npm
>>>> install
>>>> and webpack compile the same way our RPMs do in order to catch these sort
>>>> of issues earlier.
>>>
>>>
>>> Thanks for this!
>>>
>>>> The second half is that after RPMs were built, repoclosure on the test
>>>> pipeline is currently failing [2]. The highlight being:
>>>>
>>>> *20:10:35* package: nodejs-react-dom-15.6.2-1.el7.noarch from
>>>> undertest-yum_el7-4203183943-68*20:10:35* unresolved deps:
>>>> *20:10:35* npm(object-assign) >= 0:4.1.0*20:10:35*
>>>> npm(loose-envify) < 0:2*20:10:35* npm(loose-envify) >= 0:1.1.0
>>>>
>>>>
>>>>
>>>> nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
>>>> need to add both of these as top level packages carried in
>>>> foreman-packaging. Can anyone confirm or deny that is how this should be
>>>> working? Or should another package we build be providing these?
>>>
>>>
>>> I do believe we need to package this ourselves since we need a newer
>>> version than in EPEL.
>>>
>>>> [1] https://github.com/theforeman/foreman/pull/4924
>>>> [2] http://ci.theforeman.org/job/packaging_repoclosure/37110/console
···
On Wed, Oct 18, 2017 at 02:24:19PM +0200, Ewoud Kohl van Wijngaarden wrote:
>Agreed. https://github.com/theforeman/foreman-packaging/pull/1866
>
>On Tue, Oct 17, 2017 at 09:16:52AM +0200, Lukas Zapletal wrote:
>>EPEL is not great place to be for Rails or Node components. You should
>>not bump versions in EPEL7 (major relase should go into EPEL8).
>>
>>On Tue, Oct 17, 2017 at 12:01 AM, Eric D Helms wrote:
>>>
>>>
>>>On Oct 16, 2017 5:17 PM, "Sean O'Keeffe" wrote:
>>>
>>>Why dont we ask the maintainer to pkg a new version or someone offer to
>>>become a co-maintainer and get a new version into EPEL ?
>>>
>>>
>>>While I think this is the right open source path, I'd weigh:
>>>
>>> 1) how long will nighties be broken waiting on a new package?
>>> 2) 2.0 to 4.1 is a large jump and as a base dependency other EPEL packages
>>>may not work.
>>>
>>>
>>>
>>>On Mon, Oct 16, 2017 at 9:31 PM, Ewoud Kohl van Wijngaarden >>> wrote:
>>>>
>>>>On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
>>>>>
>>>>>Nightly RPM and tests have been broken for around 2 weeks now. This
>>>>>morning
>>>>>a bit of a regression was merged to foreman core to fix the breaking RPM
>>>>>aspect and I can report that nightly RPMs are now building. However, this
>>>>>leads to a breakage in plugin asset usage with the newer React
>>>>>components.
>>>>>To potentially address this I have opened [1] for testing and input. As
>>>>>part of the original breakage, I added to test_develop running npm
>>>>>install
>>>>>and webpack compile the same way our RPMs do in order to catch these sort
>>>>>of issues earlier.
>>>>
>>>>
>>>>Thanks for this!
>>>>
>>>>>The second half is that after RPMs were built, repoclosure on the test
>>>>>pipeline is currently failing [2]. The highlight being:
>>>>>
>>>>>*20:10:35* package: nodejs-react-dom-15.6.2-1.el7.noarch from
>>>>>undertest-yum_el7-4203183943-68*20:10:35* unresolved deps:
>>>>>*20:10:35* npm(object-assign) >= 0:4.1.0*20:10:35*
>>>>>npm(loose-envify) < 0:2*20:10:35* npm(loose-envify) >= 0:1.1.0
>>>>>
>>>>>
>>>>>
>>>>>nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we will
>>>>>need to add both of these as top level packages carried in
>>>>>foreman-packaging. Can anyone confirm or deny that is how this should be
>>>>>working? Or should another package we build be providing these?
>>>>
>>>>
>>>>I do believe we need to package this ourselves since we need a newer
>>>>version than in EPEL.
>>>>
>>>>>[1] https://github.com/theforeman/foreman/pull/4924
>>>>>[2] http://ci.theforeman.org/job/packaging_repoclosure/37110/console
On Mon, Oct 16, 2017 at 04:18:53PM -0400, Eric D Helms wrote:
Nightly RPM and tests have been broken for around 2 weeks now. This
morning
a bit of a regression was merged to foreman core to fix the breaking
RPM
aspect and I can report that nightly RPMs are now building. However,
this
leads to a breakage in plugin asset usage with the newer React
components.
To potentially address this I have opened [1] for testing and input.
As
part of the original breakage, I added to test_develop running npm
install
and webpack compile the same way our RPMs do in order to catch these
sort
of issues earlier.
Thanks for this!
The second half is that after RPMs were built, repoclosure on the test
pipeline is currently failing [2]. The highlight being:
nodejs-object-assign-2.0.0 exists in EPEL leading me to believe we
will
need to add both of these as top level packages carried in
foreman-packaging. Can anyone confirm or deny that is how this should
be
working? Or should another package we build be providing these?
I do believe we need to package this ourselves since we need a newer
version than in EPEL.
