[oslo][TC] Dropping lower-constraints testing
Hello,
As you already surely know, we (the openstack project) currently face some issues with our lower-constraints jobs due to pip's latest resolver feature.
By discussing this topic with Thierry Carrez (ttx) from an oslo point of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain needed to maintain them now exceeds the benefits provided by this mechanismes.
Also we should notice that the number of active maintainers is declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop these tests [1].
But before moving further with that we would appreciate discussing this with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are covered by [1].
Thoughts?
Thanks for reading.
[1] https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%20st...)
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud hberaud@redhat.com wrote ----
Hello, As you already surely know, we (the openstack project) currently face some issues with our lower-constraints jobs due to pip's latest resolver feature. By discussing this topic with Thierry Carrez (ttx) from an oslo point of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain neededto maintain them now exceeds the benefits provided by this mechanismes. Also we should notice that the number of active maintainers is declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop these tests [1]. But before moving further with that we would appreciate discussing this with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are covered by [1].
Thoughts?
+1, I think it's not worth to keep maintaining them which is taking too much effort.
-gmann
Thanks for reading.
[1] https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%20st...)
Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://github.com/4383/https://twitter.com/4383hberaud -----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
+1
On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann gmann@ghanshyammann.com wrote:
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud hberaud@redhat.com wrote ----
Hello, As you already surely know, we (the openstack project) currently face
some issues with our lower-constraints jobs due to pip's latest resolver feature.
By discussing this topic with Thierry Carrez (ttx) from an oslo point
of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain neededto maintain them now exceeds the benefits provided by this mechanismes.
Also we should notice that the number of active maintainers is
declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our
lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop these
tests [1].
But before moving further with that we would appreciate discussing this
with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are covered by [1].
Thoughts?
+1, I think it's not worth to keep maintaining them which is taking too much effort.
-gmann
Thanks for reading.
[1]
https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%20st...)
-- Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://
github.com/4383/https://twitter.com/4383hberaud
-----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
Hello everyone,
Here is an email to officialize our position on Olso concerning our lower-constraints testing policy. Indeed we reached a consensus during our last meeting and all the Oslo cores who spoke agreed with this decision to drop the L.C tests [1].
So we already started to drop lower-constraints jobs on Oslo to unlock our gates.
Thanks to everyone who joined the discussion and thanks for reading!
[1] http://eavesdrop.openstack.org/meetings/oslo/2021/oslo.2021-01-04-16.00.log.... #topic Dropping lower-constraints testing
Le ven. 18 déc. 2020 à 20:34, Moises Guimaraes de Medeiros < moguimar@redhat.com> a écrit :
+1
On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann gmann@ghanshyammann.com wrote:
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud hberaud@redhat.com wrote ----
Hello, As you already surely know, we (the openstack project) currently face
some issues with our lower-constraints jobs due to pip's latest resolver feature.
By discussing this topic with Thierry Carrez (ttx) from an oslo point
of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain neededto maintain them now exceeds the benefits provided by this mechanismes.
Also we should notice that the number of active maintainers is
declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our
lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop these
tests [1].
But before moving further with that we would appreciate discussing
this with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are covered by
[1].
Thoughts?
+1, I think it's not worth to keep maintaining them which is taking too much effort.
-gmann
Thanks for reading.
[1]
https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%20st...)
-- Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://
github.com/4383/https://twitter.com/4383hberaud
-----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
--
Moisés Guimarães
Software Engineer
Red Hat https://www.redhat.com
Hi,
Dnia środa, 6 stycznia 2021 13:00:31 CET Herve Beraud pisze:
Hello everyone,
Here is an email to officialize our position on Olso concerning our lower-constraints testing policy. Indeed we reached a consensus during our last meeting and all the Oslo cores who spoke agreed with this decision to drop the L.C tests [1].
So we already started to drop lower-constraints jobs on Oslo to unlock our gates.
Sorry if my question shouldn't be in that thread but does that mean that other projects should/can drop lower constraints jobs too? Is it some general, OpenStack wide decision or it depends on the project?
Thanks to everyone who joined the discussion and thanks for reading!
[1] http://eavesdrop.openstack.org/meetings/oslo/2021/oslo.2021-01-04-16.00.log. txt #topic Dropping lower-constraints testing
Le ven. 18 déc. 2020 à 20:34, Moises Guimaraes de Medeiros <
moguimar@redhat.com> a écrit :
+1
On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann gmann@ghanshyammann.com
wrote:
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud hberaud@redhat.com
wrote ----
Hello, As you already surely know, we (the openstack project) currently face
some issues with our lower-constraints jobs due to pip's latest resolver feature.
By discussing this topic with Thierry Carrez (ttx) from an oslo point
of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain neededto maintain them now exceeds the benefits provided by this mechanismes.
Also we should notice that the number of active maintainers is
declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our
lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop these
tests [1].
But before moving further with that we would appreciate discussing
this with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are covered by
[1].
Thoughts?
+1, I think it's not worth to keep maintaining them which is taking too much effort.
-gmann
Thanks for reading.
[1]
https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%2 0status:merged)>>
-- Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://
github.com/4383/https://twitter.com/4383hberaud
-----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
--
Moisés Guimarães
Software Engineer
Red Hat https://www.redhat.com
-- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud
On Wed, 2021-01-06 at 13:43 +0100, Slawek Kaplonski wrote:
Hi,
Dnia środa, 6 stycznia 2021 13:00:31 CET Herve Beraud pisze:
Hello everyone,
Here is an email to officialize our position on Olso concerning our lower-constraints testing policy. Indeed we reached a consensus during our last meeting and all the Oslo cores who spoke agreed with this decision to drop the L.C tests [1].
So we already started to drop lower-constraints jobs on Oslo to unlock our gates.
Sorry if my question shouldn't be in that thread but does that mean that other projects should/can drop lower constraints jobs too? Is it some general, OpenStack wide decision or it depends on the project?
lower constratis is not a required part fo the PTI https://governance.openstack.org/tc/reference/pti/python.html https://governance.openstack.org/tc/reference/project-testing-interface.html so its technially a per project desision as far as i am aware.
while project were encouraged to adopt lower constratit testing we did not require all project to provide it and i belive there are some that dont unless im mistaken.
so neutron could drop lc testing too i belive if it desired too. unless im misinterperting things
Thanks to everyone who joined the discussion and thanks for reading!
[1] http://eavesdrop.openstack.org/meetings/oslo/2021/oslo.2021-01-04-16.00.log. txt #topic Dropping lower-constraints testing
Le ven. 18 déc. 2020 à 20:34, Moises Guimaraes de Medeiros <
moguimar@redhat.com> a écrit :
+1
On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann gmann@ghanshyammann.com
wrote:
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud hberaud@redhat.com
wrote ----
> Hello, > As you already surely know, we (the openstack project) currently face
some issues with our lower-constraints jobs due to pip's latest resolver feature.
> By discussing this topic with Thierry Carrez (ttx) from an oslo point
of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain neededto maintain them now exceeds the benefits provided by this mechanismes.
> Also we should notice that the number of active maintainers is
declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
> In a first time I tried to fix our gates by fixing our
lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
> So, in a second time I submitted a series of patches to drop these
tests [1].
> But before moving further with that we would appreciate discussing
this with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
> Normally all the oslo projects that need to be fixed are covered by
[1].
> Thoughts?
+1, I think it's not worth to keep maintaining them which is taking too much effort.
-gmann
> Thanks for reading. > > [1]
https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%2 0status:merged)>> > -- > Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://
github.com/4383/https://twitter.com/4383hberaud
> -----BEGIN PGP SIGNATURE----- > > wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ > Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ > RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP > F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G > 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g > glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw > m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ > hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 > qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y > F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 > B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O > v6rDpkeNksZ9fFSyoY2o > =ECSj > -----END PGP SIGNATURE-----
--
Moisés Guimarães
Software Engineer
Red Hat https://www.redhat.com
-- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud
Concerning Oslo this is a team decision/position and for now this isn't something general, so, you can still continue to test LC if it works for you.
The TC still needs to decide if we should continue with that or not in an official manner.
Sorry if my precedent message is misleading some of you.
Le mer. 6 janv. 2021 à 14:25, Sean Mooney smooney@redhat.com a écrit :
On Wed, 2021-01-06 at 13:43 +0100, Slawek Kaplonski wrote:
Hi,
Dnia środa, 6 stycznia 2021 13:00:31 CET Herve Beraud pisze:
Hello everyone,
Here is an email to officialize our position on Olso concerning our lower-constraints testing policy. Indeed we reached a consensus during our last meeting and all the Oslo cores who spoke agreed with this decision to drop the L.C tests [1].
So we already started to drop lower-constraints jobs on Oslo to unlock
our
gates.
Sorry if my question shouldn't be in that thread but does that mean that
other
projects should/can drop lower constraints jobs too? Is it some general, OpenStack wide decision or it depends on the project?
lower constratis is not a required part fo the PTI https://governance.openstack.org/tc/reference/pti/python.html
https://governance.openstack.org/tc/reference/project-testing-interface.html so its technially a per project desision as far as i am aware.
while project were encouraged to adopt lower constratit testing we did not require all project to provide it and i belive there are some that dont unless im mistaken.
so neutron could drop lc testing too i belive if it desired too. unless im misinterperting things
Thanks to everyone who joined the discussion and thanks for reading!
[1]
http://eavesdrop.openstack.org/meetings/oslo/2021/oslo.2021-01-04-16.00.log .
txt #topic Dropping lower-constraints testing
Le ven. 18 déc. 2020 à 20:34, Moises Guimaraes de Medeiros <
moguimar@redhat.com> a écrit :
+1
On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann <
gmann@ghanshyammann.com>
wrote:
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud <
hberaud@redhat.com>
wrote ----
Hello, As you already surely know, we (the openstack project)
currently face
some issues with our lower-constraints jobs due to pip's latest
resolver
feature.
By discussing this topic with Thierry Carrez (ttx) from an oslo
point
of view, we reached the same conclusion that it is more
appropriate to
drop this kind of tests because the complexity and recurring pain
neededto
maintain them now exceeds the benefits provided by this
mechanismes.
Also we should notice that the number of active maintainers is
declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our
lower-constraints project by project but with around ~36 projects
to
maintain this is a painful task, especially due to nested oslo
layers
inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop
these
tests [1].
But before moving further with that we would appreciate
discussing
this with the TC. For now the patches are ready and we just have
to push
the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are
covered by
[1].
Thoughts?
+1, I think it's not worth to keep maintaining them which is
taking too
much effort.
-gmann
Thanks for reading.
[1]
https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%2
0status:merged)>>
-- Hervé BeraudSenior Software Engineer at Red Hatirc:
hberaudhttps://
github.com/4383/https://twitter.com/4383hberaud
-----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
--
Moisés Guimarães
Software Engineer
Red Hat https://www.redhat.com
-- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud
On 1/6/21 9:56 AM, Herve Beraud wrote:
Concerning Oslo this is a team decision/position and for now this isn't something general, so, you can still continue to test LC if it works for you.
The only issue I see is that based on our discussions, the big problem with lower-constraints testing is that it requires all of your dependencies to have compatible l-c too, so if Oslo drops testing of that it's going to make it extremely difficult for any other OpenStack projects to maintain it.
The TC still needs to decide if we should continue with that or not in an official manner.
Sorry if my precedent message is misleading some of you.
Le mer. 6 janv. 2021 à 14:25, Sean Mooney <smooney@redhat.com mailto:smooney@redhat.com> a écrit :
On Wed, 2021-01-06 at 13:43 +0100, Slawek Kaplonski wrote: > Hi, > > Dnia środa, 6 stycznia 2021 13:00:31 CET Herve Beraud pisze: > > Hello everyone, > > > > Here is an email to officialize our position on Olso concerning our > > lower-constraints testing policy. > > Indeed we reached a consensus during our last meeting and all the Oslo > > cores who spoke agreed with this decision to drop the L.C tests [1]. > > > > So we already started to drop lower-constraints jobs on Oslo to unlock our > > gates. > > Sorry if my question shouldn't be in that thread but does that mean that other > projects should/can drop lower constraints jobs too? Is it some general, > OpenStack wide decision or it depends on the project? lower constratis is not a required part fo the PTI https://governance.openstack.org/tc/reference/pti/python.html https://governance.openstack.org/tc/reference/project-testing-interface.html so its technially a per project desision as far as i am aware. while project were encouraged to adopt lower constratit testing we did not require all project to provide it and i belive there are some that dont unless im mistaken. so neutron could drop lc testing too i belive if it desired too. unless im misinterperting things > > > > > Thanks to everyone who joined the discussion and thanks for reading! > > > > [1] > > http://eavesdrop.openstack.org/meetings/oslo/2021/oslo.2021-01-04-16.00.log. > > txt #topic Dropping lower-constraints testing > > > > Le ven. 18 déc. 2020 à 20:34, Moises Guimaraes de Medeiros < > > > > moguimar@redhat.com <mailto:moguimar@redhat.com>> a écrit : > > > +1 > > > > > > On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann <gmann@ghanshyammann.com <mailto:gmann@ghanshyammann.com>> > > > > > > wrote: > > > > ---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud <hberaud@redhat.com <mailto:hberaud@redhat.com>> > > > > > > > > wrote ---- > > > > > > > > > Hello, > > > > > As you already surely know, we (the openstack project) currently face > > > > > > > > some issues with our lower-constraints jobs due to pip's latest resolver > > > > feature. > > > > > > > > > By discussing this topic with Thierry Carrez (ttx) from an oslo point > > > > > > > > of view, we reached the same conclusion that it is more appropriate to > > > > drop > > > > this kind of tests because the complexity and recurring pain neededto > > > > maintain them now exceeds the benefits provided by this mechanismes. > > > > > > > > > Also we should notice that the number of active maintainers is > > > > > > > > declining, so we think that this is the shortest path to solve this > > > > problem > > > > on oslo for now and for the future too. > > > > > > > > > In a first time I tried to fix our gates by fixing our > > > > > > > > lower-constraints project by project but with around ~36 projects to > > > > maintain this is a painful task, especially due to nested oslo layers > > > > inside oslo himself... I saw the face of the hell of dependencies. > > > > > > > > > So, in a second time I submitted a series of patches to drop these > > > > > > > > tests [1]. > > > > > > > > > But before moving further with that we would appreciate discussing > > > > > > > > this with the TC. For now the patches are ready and we just have to push > > > > the good button accordingly to our choices (+W or abandon). > > > > > > > > > Normally all the oslo projects that need to be fixed are covered by > > > > > > > > [1]. > > > > > > > > > Thoughts? > > > > > > > > +1, I think it's not worth to keep maintaining them which is taking too > > > > much effort. > > > > > > > > -gmann > > > > > > > > > Thanks for reading. > > > > > > > > > > [1] > > > > > > > > https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%2 > > > > 0status:merged)>> > > > > > -- > > > > > Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps:// > > > > > > > > github.com/4383/https://twitter.com/4383hberaud <http://github.com/4383/https://twitter.com/4383hberaud> > > > > > > > > > -----BEGIN PGP SIGNATURE----- > > > > > > > > > > wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ > > > > > Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ > > > > > RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP > > > > > F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G > > > > > 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g > > > > > glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw > > > > > m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ > > > > > hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 > > > > > qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y > > > > > F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 > > > > > B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O > > > > > v6rDpkeNksZ9fFSyoY2o > > > > > =ECSj > > > > > -----END PGP SIGNATURE----- > > > > > > -- > > > > > > Moisés Guimarães > > > > > > Software Engineer > > > > > > Red Hat <https://www.redhat.com> > > > > > > <https://red.ht/sig> > > > > -- > > Hervé Beraud > > Senior Software Engineer at Red Hat > > irc: hberaud > > https://github.com/4383/ > > https://twitter.com/4383hberaud > >-- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud -----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
On 2021-01-06 10:30:52 -0600 (-0600), Ben Nemec wrote: [...]
The only issue I see is that based on our discussions, the big problem with lower-constraints testing is that it requires all of your dependencies to have compatible l-c too, so if Oslo drops testing of that it's going to make it extremely difficult for any other OpenStack projects to maintain it.
[...]
I hardly see how projects could ever expect this to be the case anyway, as the vast majority of OpenStack dependencies are maintained outside the OpenStack community and don't maintain any lower-bounds testing of their own.
---- On Wed, 06 Jan 2021 06:43:18 -0600 Slawek Kaplonski skaplons@redhat.com wrote ----
Hi,
Dnia środa, 6 stycznia 2021 13:00:31 CET Herve Beraud pisze:
Hello everyone,
Here is an email to officialize our position on Olso concerning our lower-constraints testing policy. Indeed we reached a consensus during our last meeting and all the Oslo cores who spoke agreed with this decision to drop the L.C tests [1].
So we already started to drop lower-constraints jobs on Oslo to unlock our gates.
Sorry if my question shouldn't be in that thread but does that mean that other projects should/can drop lower constraints jobs too? Is it some general, OpenStack wide decision or it depends on the project?
Yeah this is a good question till now we have been discussing specific project wise dropping like oslo in this thread and Ironic and other projects in another discussion.
I am starting a separate thread to discuss it for all the projects as a general recommendation.
-http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019672.h...
-gmann
Thanks to everyone who joined the discussion and thanks for reading!
[1] http://eavesdrop.openstack.org/meetings/oslo/2021/oslo.2021-01-04-16.00.log. txt #topic Dropping lower-constraints testing
Le ven. 18 déc. 2020 à 20:34, Moises Guimaraes de Medeiros <
moguimar@redhat.com> a écrit :
+1
On Fri, Dec 18, 2020 at 4:46 PM Ghanshyam Mann gmann@ghanshyammann.com
wrote:
---- On Fri, 18 Dec 2020 08:54:26 -0600 hberaud hberaud@redhat.com
wrote ----
Hello, As you already surely know, we (the openstack project) currently face
some issues with our lower-constraints jobs due to pip's latest resolver feature.
By discussing this topic with Thierry Carrez (ttx) from an oslo point
of view, we reached the same conclusion that it is more appropriate to drop this kind of tests because the complexity and recurring pain neededto maintain them now exceeds the benefits provided by this mechanismes.
Also we should notice that the number of active maintainers is
declining, so we think that this is the shortest path to solve this problem on oslo for now and for the future too.
In a first time I tried to fix our gates by fixing our
lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
So, in a second time I submitted a series of patches to drop these
tests [1].
But before moving further with that we would appreciate discussing
this with the TC. For now the patches are ready and we just have to push the good button accordingly to our choices (+W or abandon).
Normally all the oslo projects that need to be fixed are covered by
[1].
Thoughts?
+1, I think it's not worth to keep maintaining them which is taking too much effort.
-gmann
Thanks for reading.
[1]
https://review.opendev.org/q/topic:%22oslo_lc_drop%22+(status:open%20OR%2 0status:merged)>>
-- Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://
github.com/4383/https://twitter.com/4383hberaud
-----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
--
Moisés Guimarães
Software Engineer
Red Hat https://www.redhat.com
-- Hervé Beraud Senior Software Engineer at Red Hat irc: hberaud https://github.com/4383/ https://twitter.com/4383hberaud
-- Slawek Kaplonski Principal Software Engineer Red Hat
On 12/18/20 3:54 PM, hberaud wrote:
In a first time I tried to fix our gates by fixing our lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
Welcome to my world!
Thoughts?
Couldn't someone address the dependency loops in Oslo? It's IMO anyway needed.
Just my 2 cents, not sure if that helps... Cheers,
Thomas Goirand (zigo)
On 1/6/21 2:58 PM, Thomas Goirand wrote:
On 12/18/20 3:54 PM, hberaud wrote:
In a first time I tried to fix our gates by fixing our lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
Welcome to my world!
Thoughts?
Couldn't someone address the dependency loops in Oslo? It's IMO anyway needed.
Which libraries have circular dependencies? It's something we've intentionally avoided, so if they exist I would consider it a bug.
Just my 2 cents, not sure if that helps... Cheers,
Thomas Goirand (zigo)
Le lun. 11 janv. 2021 à 18:11, Ben Nemec openstack@nemebean.com a écrit :
On 1/6/21 2:58 PM, Thomas Goirand wrote:
On 12/18/20 3:54 PM, hberaud wrote:
In a first time I tried to fix our gates by fixing our lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
Welcome to my world!
Thoughts?
Couldn't someone address the dependency loops in Oslo? It's IMO anyway needed.
AFAIK nobody is volunteering on our side, feel free to send patches to fix them....
Which libraries have circular dependencies? It's something we've intentionally avoided, so if they exist I would consider it a bug.
Concerning myself, I meant that by example to fix oslo.messaging its LC you should first address those of oslo.log and etc...
This isn't a circular dependency but an order to follow in our patches to fix all our L.C.
Also I didn't notice "real" circular dependencies, in other words I only noticed an order to apply to fix all the L.C, but over 36 projects that order is a bit annoying, that's all.
Just my 2 cents, not sure if that helps... Cheers,
Thomas Goirand (zigo)
---- On Mon, 11 Jan 2021 11:57:37 -0600 Herve Beraud hberaud@redhat.com wrote ----
Le lun. 11 janv. 2021 à 18:11, Ben Nemec openstack@nemebean.com a écrit :
On 1/6/21 2:58 PM, Thomas Goirand wrote:
On 12/18/20 3:54 PM, hberaud wrote:
In a first time I tried to fix our gates by fixing our lower-constraints project by project but with around ~36 projects to maintain this is a painful task, especially due to nested oslo layers inside oslo himself... I saw the face of the hell of dependencies.
Welcome to my world!
Thoughts?
Couldn't someone address the dependency loops in Oslo? It's IMO anyway needed.
AFAIK nobody is volunteering on our side, feel free to send patches to fix them....
Which libraries have circular dependencies? It's something we've intentionally avoided, so if they exist I would consider it a bug.
Concerning myself, I meant that by example to fix oslo.messaging its LC you should first address those of oslo.log and etc... This isn't a circular dependency but an order to follow in our patches to fix all our L.C. Also I didn't notice "real" circular dependencies, in other words I only noticed an order to apply to fix all the L.C, but over 36 projects that order is a bit annoying, that's all.
I agree.
I have added this topic to the next TC meeting on the 14th Jan agenda to get consensus for all the projects.
- https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Agenda_Suggestio...
-gmann
Just my 2 cents, not sure if that helps... Cheers,
Thomas Goirand (zigo)
-- Hervé BeraudSenior Software Engineer at Red Hatirc: hberaudhttps://github.com/4383/https://twitter.com/4383hberaud -----BEGIN PGP SIGNATURE-----
wsFcBAABCAAQBQJb4AwCCRAHwXRBNkGNegAALSkQAHrotwCiL3VMwDR0vcja10Q+ Kf31yCutl5bAlS7tOKpPQ9XN4oC0ZSThyNNFVrg8ail0SczHXsC4rOrsPblgGRN+ RQLoCm2eO1AkB0ubCYLaq0XqSaO+Uk81QxAPkyPCEGT6SRxXr2lhADK0T86kBnMP F8RvGolu3EFjlqCVgeOZaR51PqwUlEhZXZuuNKrWZXg/oRiY4811GmnvzmUhgK5G 5+f8mUg74hfjDbR2VhjTeaLKp0PhskjOIKY3vqHXofLuaqFDD+WrAy/NgDGvN22g glGfj472T3xyHnUzM8ILgAGSghfzZF5Skj2qEeci9cB6K3Hm3osj+PbvfsXE/7Kw m/xtm+FjnaywZEv54uCmVIzQsRIm1qJscu20Qw6Q0UiPpDFqD7O6tWSRKdX11UTZ hwVQTMh9AKQDBEh2W9nnFi9kzSSNu4OQ1dRMcYHWfd9BEkccezxHwUM4Xyov5Fe0 qnbfzTB1tYkjU78loMWFaLa00ftSxP/DtQ//iYVyfVNfcCwfDszXLOqlkvGmY1/Y F1ON0ONekDZkGJsDoS6QdiUSn8RZ2mHArGEWMV00EV5DCIbCXRvywXV43ckx8Z+3 B8qUJhBqJ8RS2F+vTs3DTaXqcktgJ4UkhYC2c1gImcPRyGrK9VY0sCT+1iA+wp/O v6rDpkeNksZ9fFSyoY2o =ECSj -----END PGP SIGNATURE-----
participants (9)
-
Ben Nemec
-
Ghanshyam Mann
-
hberaud
-
Herve Beraud
-
Jeremy Stanley
-
Moises Guimaraes de Medeiros
-
Sean Mooney
-
Slawek Kaplonski
-
Thomas Goirand