Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
and here Pagure:
https://src.fedoraproject.org/rpms/cockpit/pull-request/3
The Cockpit Git repository has a "test-canary" test that checks whether CI gates on test failure. Here's what you see when you run the tests in dist-git locally:
TASK [standard-test-scripts : Check the results] ******************************* fatal: [localhost]: FAILED! => {"changed": true, "cmd": "grep "^FAIL" /var/tmp/artifacts/test.log", "delta": "0:00:00.002181", "end": "2018-03-26 14:04:10.414274", "failed_when_result": true, "rc": 0, "start": "2018-03-26 14:04:10.412093", "stderr": "", "stderr_lines": [], "stdout": "FAIL ./test-canary", "stdout_lines": ["FAIL ./test-canary"]} /tests.retry
Cheers,
Stef
On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote:
Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
Will look into this one
and here Pagure:
I don't believe there is anything triggering off PR or reporting to PR at the moment.
Pierre
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote:
Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
Will look into this one
And the answer is easy: the CI pipeline does not run for F28: https://jenkins-continuous-infra.apps.ci.centos.org/
Pierre
On 26.03.2018 14:47, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote:
Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
Will look into this one
And the answer is easy: the CI pipeline does not run for F28:
What about these? I guess they're in a staging environment for now?
https://jenkins-continuous-infra.apps.ci.centos.org/view/Fedora%20All%20Pack...
But that's no the reason. I see the same behavior for F27 though:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-ecee722a34
Cheers,
Stef
On Mon, Mar 26, 2018 at 02:50:39PM +0200, Stef Walter wrote:
On 26.03.2018 14:47, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote:
Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
Will look into this one
And the answer is easy: the CI pipeline does not run for F28:
What about these? I guess they're in a staging environment for now?
https://jenkins-continuous-infra.apps.ci.centos.org/view/Fedora%20All%20Pack...
Unfortunately nope, they have been pushed to prod directly...
But I don't know the status of that pipeline, so I'll let Dominik comment more on this. Looking at the greenwave policy, it does not seem to take it into account: https://infrastructure.fedoraproject.org/cgit/ansible.git/tree/roles/openshi...
Pierre
On Mon, Mar 26, 2018 at 02:50:39PM +0200, Stef Walter wrote:
But that's no the reason. I see the same behavior for F27 though:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-ecee722a34
It should have been gated at least because of the missing results, but looks like greenwave has been told to only care about Atomic CI on F26 and not on F27.
I'll see if I can adjust that.
Pierre
On Mon, Mar 26, 2018 at 02:47:33PM +0200, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote:
Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
Will look into this one
And the answer is easy: the CI pipeline does not run for F28: https://jenkins-continuous-infra.apps.ci.centos.org/
The commit on the f27 branch is date of March 21st: https://src.fedoraproject.org/rpms/cockpit/c/e5892baf1dd0eab5154cbc4bd868757... But looking through all the builds in the Atomic CI pipeline for f27: https://jenkins-continuous-infra.apps.ci.centos.org/job/continuous-infra-ci-... I'm not seeing cockpit anywhere.
And this can be explained by the fact that cockpit is ignored by the Atomic CI pipeline apparently: https://apps.fedoraproject.org/datagrepper/id?id=2018-7947c931-b5ac-476c-a23...
Pierre
-== @ri ==-
On Mon, Mar 26, 2018 at 8:47 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote:
On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote:
Hi there,
Is the CI pipeline running on Fedora and gating? It's really hard to find test results.
I'm looking around both in Bodhi:
https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18
Will look into this one
And the answer is easy: the CI pipeline does not run for F28: https://jenkins-continuous-infra.apps.ci.centos.org/
That is incorrect and Johnny reported the status last week to multiple
lists and people:
https://jenkins-continuous-infra.apps.ci.centos.org/job/upstream-fedora-f28-...
I will make it more visible on the default view
Pierre
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
On Mon, Mar 26, 2018 at 10:05:06AM -0400, Ari LiVigni wrote:
-== @ri ==- On Mon, Mar 26, 2018 at 8:47 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote: > On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote: > > Hi there, > > > > Is the CI pipeline running on Fedora and gating? It's really hard to > > find test results. > > > > I'm looking around both in Bodhi: > > > https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18 > > Will look into this one And the answer is easy: the CI pipeline does not run for F28: https://jenkins-continuous-infra.apps.ci.centos.org/
That is incorrect and Johnny reported the status last week to multiple lists and people: https://jenkins-continuous-infra.apps.ci.centos.org/job/upstream-fedora-f28-... I will make it more visible on the default viewÂ
This is the allpackages pipeline not the Atomic CI one, no?
Pierre
-== @ri ==-
On Mon, Mar 26, 2018 at 10:08 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
On Mon, Mar 26, 2018 at 10:05:06AM -0400, Ari LiVigni wrote:
-== @ri ==- On Mon, Mar 26, 2018 at 8:47 AM, Pierre-Yves Chibon <
pingou@pingoured.fr>
wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote: > On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote: > > Hi there, > > > > Is the CI pipeline running on Fedora and gating? It's really
hard to
> > find test results. > > > > I'm looking around both in Bodhi: > > > https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18 > > Will look into this one And the answer is easy: the CI pipeline does not run for F28: https://jenkins-continuous-infra.apps.ci.centos.org/
That is incorrect and Johnny reported the status last week to multiple lists and people: https://jenkins-continuous-infra.apps.ci.centos.org/job/
upstream-fedora-f28-pipeline/
I will make it more visible on the default viewÂ
This is the allpackages pipeline not the Atomic CI one, no?
But this will get the Atomic ones and the non-atomic ones so not seeing
your point. Before we were only focused on Atomic in a compose ostree, now we do all packages with regular compose and image for f27 and f28.
Pierre _______________________________________________ CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
On Mon, Mar 26, 2018 at 10:11:13AM -0400, Ari LiVigni wrote:
-== @ri ==- On Mon, Mar 26, 2018 at 10:08 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
On Mon, Mar 26, 2018 at 10:05:06AM -0400, Ari LiVigni wrote: >Â Â -== @ri ==- >Â Â On Mon, Mar 26, 2018 at 8:47 AM, Pierre-Yves Chibon <pingou@pingoured.fr> >Â Â wrote: > >Â Â Â On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote: >Â Â Â > On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote: >Â Â Â > > Hi there, >Â Â Â > > >Â Â Â > > Is the CI pipeline running on Fedora and gating? It's really hard to >Â Â Â > > find test results. >Â Â Â > > >Â Â Â > > I'm looking around both in Bodhi: >Â Â Â > >Â Â Â > > https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18 >Â Â Â > >Â Â Â > Will look into this one > >Â Â Â And the answer is easy: the CI pipeline does not run for F28: >Â Â Â https://jenkins-continuous-infra.apps.ci.centos.org/ > >Â Â That is incorrect and Johnny reported the status last week to multiple >Â Â lists and people: >Â Â https://jenkins-continuous-infra.apps.ci.centos.org/job/upstream-fedora-f28-pipeline/ >Â Â I will make it more visible on the default viewÃ*Â This is the allpackages pipeline not the Atomic CI one, no?
But this will get the Atomic ones and the non-atomic ones so not seeing your point. Before we were only focused on Atomic in a compose ostree, now we do all packages with regular compose and image for f27 and f28.
My point is that they are two different pipeline sending different topics and thus on the consuming side, they are treated differently. In this case, the ci.pipeline is used for gating, the allpackages one not yet.
Pierre
-== @ri ==-
On Mon, Mar 26, 2018 at 10:15 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
On Mon, Mar 26, 2018 at 10:11:13AM -0400, Ari LiVigni wrote:
-== @ri ==- On Mon, Mar 26, 2018 at 10:08 AM, Pierre-Yves Chibon <
pingou@pingoured.fr>
wrote:
On Mon, Mar 26, 2018 at 10:05:06AM -0400, Ari LiVigni wrote: >Â Â -== @ri ==- >Â Â On Mon, Mar 26, 2018 at 8:47 AM, Pierre-Yves Chibon <pingou@pingoured.fr> >Â Â wrote: > >Â Â Â On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves
Chibon
wrote: >Â Â Â > On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter
wrote:
>Â Â Â > > Hi there, >Â Â Â > > >Â Â Â > > Is the CI pipeline running on Fedora and gating? It's really hard to >Â Â Â > > find test results. >Â Â Â > > >Â Â Â > > I'm looking around both in Bodhi: >Â Â Â > >Â Â Â > > https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18 >Â Â Â > >Â Â Â > Will look into this one > >Â Â Â And the answer is easy: the CI pipeline does not run for
F28:
>Â Â Â https://jenkins-continuous-infra.apps.ci.centos.org/ > >Â Â That is incorrect and Johnny reported the status last week to multiple >Â Â lists and people: >Â Â https://jenkins-continuous-infra.apps.ci.centos.org/job/
upstream-fedora-f28-pipeline/
>Â Â I will make it more visible on the default viewÃ*Â This is the allpackages pipeline not the Atomic CI one, no?
But this will get the Atomic ones and the non-atomic ones so not
seeing
your point. Before we were only focused on Atomic in a compose
ostree,
now we do all packages with regular compose and image for f27 and f28.
My point is that they are two different pipeline sending different topics and thus on the consuming side, they are treated differently. In this case, the ci.pipeline is used for gating, the allpackages one not yet.
Right but that wasn't your statement you said we didn't have this yet for f28 and we do it just isn't plumbed on the Fedora side. There is more work on the Fedora side that is out of our hands to update. Are you working on that piece or is someone else in Fedora atm? If you want we can discontinue the atomic ci-pipeline ones if that makes things easier for you.
Pierre _______________________________________________ CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
On Mon, Mar 26, 2018 at 10:22:34AM -0400, Ari LiVigni wrote:
Right but that wasn't your statement you said we didn't have this yet for f28
Well, the two pipelines are doing different things right? One runs the tests, the other ensures the rpm-ostree is still buildable and runs the tests. So while overlapping they are complementary and distinct, no?
So I stand with what I said, the Atomic CI pipeline isn't running for f28.
and we do it just isn't plumbed on the Fedora side. There is more work on the Fedora side that is out of our hands to update. Are you working on that piece or is someone else in Fedora atm?
For one, I'd need a green light that the allpackages pipeline is running as expected and ready for production use. Then I'll see to adjust the tools for this new pipeline. This does mean we will announce it broadly and that we can expect its load to increase as more people opt-in.
If you want we can discontinue the atomic ci-pipeline ones if that makes things easier for you.
Do we want to discontinue the Atomic CI one? I thought the idea was to have both running (see what I said above about the two pipelines being complementary).
Pierre
-== @ri ==-
On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
On Mon, Mar 26, 2018 at 10:22:34AM -0400, Ari LiVigni wrote:
Right but that wasn't your statement you said we didn't have this yet
for
f28
Well, the two pipelines are doing different things right? One runs the tests, the other ensures the rpm-ostree is still buildable and runs the tests. So while overlapping they are complementary and distinct, no?
So I stand with what I said, the Atomic CI pipeline isn't running for f28.
and we do it just isn't plumbed on the Fedora side. There is more work on the Fedora side that is out of our hands to update. Are you working on that piece or is someone else in Fedora atm?
For one, I'd need a green light that the allpackages pipeline is running as expected and ready for production use. Then I'll see to adjust the tools for this new pipeline. This does mean we will announce it broadly and that we can expect its load to increase as more people opt-in.
Cool makes sense
If you
want we can discontinue the atomic ci-pipeline ones if that makes things
easier for you.
Do we want to discontinue the Atomic CI one? I thought the idea was to have both running (see what I said above about the two pipelines being complementary).
Let's have that discussion when Johnny is back from PTO. I am fine having both if it makes sense and there is no duplication, which I don't think there is since in the atomic one
Pierre _______________________________________________ CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
On 03/26/2018 05:34 PM, Ari LiVigni wrote:
-== @ri ==-
On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon <pingou@pingoured.fr mailto:pingou@pingoured.fr> wrote:
For one, I'd need a green light that the allpackages pipeline is running as expected and ready for production use. Then I'll see to adjust the tools for this new pipeline. This does mean we will announce it broadly and that we can expect its load to increase as more people opt-in.
I agree. Let's make sure it's stable and production ready, then announce it and help people when the inevitable issues and questions arise. :)
Do we want to discontinue the Atomic CI one? I thought the idea was to have both running (see what I said above about the two pipelines being complementary).
Let's have that discussion when Johnny is back from PTO. I am fine having both if it makes sense and there is no duplication, which I don't think there is since in the atomic one
I think both pipelines have merit, since they test different deliverables. I would prefer to keep the Atomic one around, but also have the "regular" one once that is stable.
-Dominik
Hey all,
To clear up some of the questions on this thread:
- Why no cockpit testing? "cockpit" as a package name is not in https://pagure.io/fedora-atomic/blob/f27/f/fedora-atomic-host-base.json for example, so if the distgit repo name is "cockpit" the commit by design would not have been picked up by the Atomic Pipeline.
- Atomic Pipeline doesn't test f28? The Atomic pipeline is currently set up for f26 and f27. I have been working on getting it working for f28, but I am hitting a bug and haven't resolved it yet, so it is not ready yet.
- What is the All Packages Pipeline ( https://jenkins-continuous-infra.apps.ci.centos.org/view/Fedora%20All%20Pack...) ? As alluded to, this is different than the Atomic pipelines. The Atomic pipeline does the following: trigger on distgit commit to a repo of interest (see question 1 for package lists), build rpm, compose ostree, compose qcow2 atomic image with ostree, boot sanity tests, package tests, atomic host integration tests, openshift e2e tests, report all results from all stages on fedmsg under org.centos.prod.ci.pipeline.* topics. The All Packages Pipelines do the following: trigger on distgit commit to any package to branch [fXX | master] AND the repo contains standard-test-roles defined tests in that branch, build the rpm in koji, construct cloud qcow2 image while injecting the rpm at build time, run package tests, report all results from all stages on fedmsg under org.centos.prod.allpackages.pipeline.* topics.
- Is the All Packages Pipeline production ready? I think it is. If you look here https://jenkins-continuous-infra.apps.ci.centos.org/view/all/job/upstream-fe... it is passing when it should. There even seem to be some legitimate package test failures found. If you go back in the build history, you will likely find some bugs that have since been fixed, but most of the latest builds are legit. We still need the linkage setup between these fedmsgs and pagure, though, to the best of my knowledge.
If I can clear anything else up, feel free to ask. Best, Johnny Bieren
On Mon, Mar 26, 2018 at 4:21 PM, Dominik Perpeet dperpeet@redhat.com wrote:
On 03/26/2018 05:34 PM, Ari LiVigni wrote:
-== @ri ==-
On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon pingou@pingoured.fr wrote:
For one, I'd need a green light that the allpackages pipeline is running as expected and ready for production use. Then I'll see to adjust the tools for this new pipeline. This does mean we will announce it broadly and that we can expect its load to increase as more people opt-in.
I agree. Let's make sure it's stable and production ready, then announce it and help people when the inevitable issues and questions arise. :)
Do we want to discontinue the Atomic CI one? I thought the idea was to
have both running (see what I said above about the two pipelines being complementary).
Let's have that discussion when Johnny is back from PTO. I am fine having both if it makes sense and there is no duplication, which I don't think there is since in the atomic one
I think both pipelines have merit, since they test different
deliverables. I would prefer to keep the Atomic one around, but also have the "regular" one once that is stable.
-Dominik
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
On 03.04.2018 15:57, Johnny Bieren wrote:
Hey all,
To clear up some of the questions on this thread:
- Why no cockpit testing?
"cockpit" as a package name is not in https://pagure.io/fedora-atomic/blob/f27/f/fedora-atomic-host-base.json for example, so if the distgit repo name is "cockpit" the commit by design would not have been picked up by the Atomic Pipeline.
How do we fix that?
I guess this gap is because "Discovery" section of
https://fedoraproject.org/wiki/Changes/InvokingTests
Does not address how to find the dist-git repo for a given subpackage?
Stef
- Atomic Pipeline doesn't test f28?
The Atomic pipeline is currently set up for f26 and f27. I have been working on getting it working for f28, but I am hitting a bug and haven't resolved it yet, so it is not ready yet.
- What is the All Packages Pipeline
(https://jenkins-continuous-infra.apps.ci.centos.org/view/Fedora%20All%20Pack...) ? As alluded to, this is different than the Atomic pipelines. The Atomic pipeline does the following: trigger on distgit commit to a repo of interest (see question 1 for package lists), build rpm, compose ostree, compose qcow2 atomic image with ostree, boot sanity tests, package tests, atomic host integration tests, openshift e2e tests, report all results from all stages on fedmsg under org.centos.prod.ci.pipeline.* topics. The All Packages Pipelines do the following: trigger on distgit commit to any package to branch [fXX | master] AND the repo contains standard-test-roles defined tests in that branch, build the rpm in koji, construct cloud qcow2 image while injecting the rpm at build time, run package tests, report all results from all stages on fedmsg under org.centos.prod.allpackages.pipeline.* topics.
- Is the All Packages Pipeline production ready?
I think it is. If you look here https://jenkins-continuous-infra.apps.ci.centos.org/view/all/job/upstream-fe... it is passing when it should. There even seem to be some legitimate package test failures found. If you go back in the build history, you will likely find some bugs that have since been fixed, but most of the latest builds are legit. We still need the linkage setup between these fedmsgs and pagure, though, to the best of my knowledge.
If I can clear anything else up, feel free to ask. Best, Johnny Bieren
On Mon, Mar 26, 2018 at 4:21 PM, Dominik Perpeet <dperpeet@redhat.com mailto:dperpeet@redhat.com> wrote:
On 03/26/2018 05:34 PM, Ari LiVigni wrote:
-== @ri ==- On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon <pingou@pingoured.fr <mailto:pingou@pingoured.fr>> wrote: For one, I'd need a green light that the allpackages pipeline is running as expected and ready for production use. Then I'll see to adjust the tools for this new pipeline. This does mean we will announce it broadly and that we can expect its load to increase as more people opt-in.
I agree. Let's make sure it's stable and production ready, then announce it and help people when the inevitable issues and questions arise. :)
Do we want to discontinue the Atomic CI one? I thought the idea was to have both running (see what I said above about the two pipelines being complementary). Let's have that discussion when Johnny is back from PTO. I am fine having both if it makes sense and there is no duplication, which I don't think there is since in the atomic one
I think both pipelines have merit, since they test different deliverables. I would prefer to keep the Atomic one around, but also have the "regular" one once that is stable. -Dominik _______________________________________________ CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> To unsubscribe send an email to ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org>
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
I'm not sure what you mean by fix it. If it isn't part of atomic host, shouldn't it not be tested by the atomic host pipeline?
Cockpit *is* picked up by the All Package Pipeline. You can see one commit from cockpit here https://jenkins-continuous-infra.apps.ci.centos.org/blue/organizations/jenki...
Unfortunately, that build is many builds ago so the artifacts have been discarded so you cannot see the package test logs anymore. But, wouldn't this be the solution, not having the atomic pipeline test it?
The integration with pagure for the All Packages Pipeline is still WIP I believe
On Fri, Apr 6, 2018 at 7:35 AM, Stef Walter stefw@redhat.com wrote:
On 03.04.2018 15:57, Johnny Bieren wrote:
Hey all,
To clear up some of the questions on this thread:
- Why no cockpit testing? "cockpit" as a package name is not
atomic-host-base.json
for example, so if the distgit repo name is "cockpit" the commit by design would not have been picked up by the Atomic Pipeline.
How do we fix that?
I guess this gap is because "Discovery" section of
https://fedoraproject.org/wiki/Changes/InvokingTests
Does not address how to find the dist-git repo for a given subpackage?
Stef
- Atomic Pipeline doesn't test f28? The Atomic pipeline is currently set up for f26 and f27. I have been
working on getting it working for f28, but I am hitting a bug and haven't resolved it yet, so it is not ready yet.
- What is the All Packages Pipeline
Fedora%20All%20Packages%20Pipeline/)
? As alluded to, this is different than the Atomic pipelines. The Atomic pipeline does the following: trigger on distgit commit to a repo of interest (see question 1 for package lists), build rpm, compose ostree, compose qcow2 atomic image with ostree, boot sanity tests, package tests, atomic host integration tests, openshift e2e tests, report all results from all stages on fedmsg under org.centos.prod.ci.pipeline.* topics. The All Packages Pipelines do the following: trigger on distgit commit to any package to branch [fXX | master] AND the repo contains standard-test-roles defined tests in that branch, build the rpm in koji, construct cloud qcow2 image while injecting the rpm at build time, run package tests, report all results from all stages on fedmsg under org.centos.prod.allpackages.pipeline.* topics.
- Is the All Packages Pipeline production ready? I think it is. If you look
here https://jenkins-continuous-infra.apps.ci.centos.org/view/all/job/
upstream-fedora-f28-pipeline/
it is passing when it should. There even seem to be some legitimate package test failures found. If you go back in the build history, you will likely find some bugs that have since been fixed, but most of the latest builds are legit. We still need the linkage setup between these fedmsgs and pagure, though, to the best of my knowledge.
If I can clear anything else up, feel free to ask. Best, Johnny Bieren
On Mon, Mar 26, 2018 at 4:21 PM, Dominik Perpeet <dperpeet@redhat.com mailto:dperpeet@redhat.com> wrote:
On 03/26/2018 05:34 PM, Ari LiVigni wrote:
-== @ri ==- On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon <pingou@pingoured.fr <mailto:pingou@pingoured.fr>> wrote: For one, I'd need a green light that the allpackages pipeline is running as expected and ready for production use. Then I'll see to adjust the tools for this new pipeline. This does mean we will announce it broadly and that we can expect its load to increase as more people opt-in.
I agree. Let's make sure it's stable and production ready, then announce it and help people when the inevitable issues and questions arise. :)
Do we want to discontinue the Atomic CI one? I thought the idea was to have both running (see what I said above about the two pipelines being complementary). Let's have that discussion when Johnny is back from PTO. I am fine having both if it makes sense and there is no duplication, which I don't think there is since in the atomic one
I think both pipelines have merit, since they test different deliverables. I would prefer to keep the Atomic one around, but also have the "regular" one once that is stable. -Dominik _______________________________________________ CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> To unsubscribe send an email to ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org>
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
Three reasons:
* dist-git repositories do not necessarily produce a package of the same name. They usually do.
* dist-git repositories can produce multiple sub-packages, as you can see in the case of cockpit. Any and all of those sub-packages should be tested and/or gated by tests that live in that dist-git repo.
In this case there are 5 sub-packages from the cockpit dist-git repo in the Atomic Host that should trigger the pipeline.
Although I understand current effort is on the All Package Pipeline, I believe this might still be a problem going forward:
* There is no defined way to discover which dist-git repos to look for tests related to a certain package in the standard test specification.
Does that make sense? Dominik, did you have any ideas on how we might fix this?
Cheers,
Stef
On 06.04.2018 15:22, Johnny Bieren wrote:
I'm not sure what you mean by fix it. If it isn't part of atomic host, shouldn't it not be tested by the atomic host pipeline?
Cockpit *is* picked up by the All Package Pipeline. You can see one commit from cockpit here https://jenkins-continuous-infra.apps.ci.centos.org/blue/organizations/jenki...
Unfortunately, that build is many builds ago so the artifacts have been discarded so you cannot see the package test logs anymore. But, wouldn't this be the solution, not having the atomic pipeline test it?
The integration with pagure for the All Packages Pipeline is still WIP I believe
On Fri, Apr 6, 2018 at 7:35 AM, Stef Walter <stefw@redhat.com mailto:stefw@redhat.com> wrote:
On 03.04.2018 15:57, Johnny Bieren wrote: > Hey all, > > To clear up some of the questions on this thread: > > - Why no cockpit testing? > "cockpit" as a package name is not > in https://pagure.io/fedora-atomic/blob/f27/f/fedora-atomic-host-base.json <https://pagure.io/fedora-atomic/blob/f27/f/fedora-atomic-host-base.json> > for example, so if the distgit repo name is "cockpit" the commit by > design would not have been picked up by the Atomic Pipeline. How do we fix that? I guess this gap is because "Discovery" section of https://fedoraproject.org/wiki/Changes/InvokingTests <https://fedoraproject.org/wiki/Changes/InvokingTests> Does not address how to find the dist-git repo for a given subpackage? Stef > > - Atomic Pipeline doesn't test f28? > The Atomic pipeline is currently set up for f26 and f27. I have been > working on getting it working for f28, but I am hitting a bug and > haven't resolved it yet, so it is not ready yet. > > - What is the All Packages Pipeline > (https://jenkins-continuous-infra.apps.ci.centos.org/view/Fedora%20All%20Packages%20Pipeline/ <https://jenkins-continuous-infra.apps.ci.centos.org/view/Fedora%20All%20Packages%20Pipeline/>) > ? > As alluded to, this is different than the Atomic pipelines. The Atomic > pipeline does the following: trigger on distgit commit to a repo of > interest (see question 1 for package lists), build rpm, compose ostree, > compose qcow2 atomic image with ostree, boot sanity tests, package > tests, atomic host integration tests, openshift e2e tests, report all > results from all stages on fedmsg under org.centos.prod.ci.pipeline.* > topics. The All Packages Pipelines do the following: trigger on distgit > commit to any package to branch [fXX | master] AND the repo contains > standard-test-roles defined tests in that branch, build the rpm in koji, > construct cloud qcow2 image while injecting the rpm at build time, run > package tests, report all results from all stages on fedmsg under > org.centos.prod.allpackages.pipeline.* topics. > > - Is the All Packages Pipeline production ready? > I think it is. If you look > here https://jenkins-continuous-infra.apps.ci.centos.org/view/all/job/upstream-fedora-f28-pipeline/ <https://jenkins-continuous-infra.apps.ci.centos.org/view/all/job/upstream-fedora-f28-pipeline/> > it is passing when it should. There even seem to be some legitimate > package test failures found. If you go back in the build history, you > will likely find some bugs that have since been fixed, but most of the > latest builds are legit. We still need the linkage setup between these > fedmsgs and pagure, though, to the best of my knowledge. > > If I can clear anything else up, feel free to ask. > Best, > Johnny Bieren > > On Mon, Mar 26, 2018 at 4:21 PM, Dominik Perpeet <dperpeet@redhat.com <mailto:dperpeet@redhat.com> > <mailto:dperpeet@redhat.com <mailto:dperpeet@redhat.com>>> wrote: > > On 03/26/2018 05:34 PM, Ari LiVigni wrote: >> >> >> -== @ri ==- >> >> On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon >> <pingou@pingoured.fr <mailto:pingou@pingoured.fr> <mailto:pingou@pingoured.fr <mailto:pingou@pingoured.fr>>> wrote: >> >> >> >> For one, I'd need a green light that the allpackages pipeline >> is running as >> expected and ready for production use. Then I'll see to adjust >> the tools for >> this new pipeline. >> This does mean we will announce it broadly and that we can >> expect its load to >> increase as more people opt-in. >> > > I agree. Let's make sure it's stable and production ready, then > announce it and help people when the inevitable issues and questions > arise. :) > >> >> Do we want to discontinue the Atomic CI one? I thought the >> idea was to have both >> running (see what I said above about the two pipelines being >> complementary). >> >> >> Let's have that discussion when Johnny is back from PTO. I am >> fine having both if it makes sense and there is no duplication, >> which I don't think there is since in the atomic one >> >> >> > I think both pipelines have merit, since they test different > deliverables. I would prefer to keep the Atomic one around, but also > have the "regular" one once that is stable. > > -Dominik > > > _______________________________________________ > CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> > <mailto:ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org>> > To unsubscribe send an email to ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org> > <mailto:ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org>> > > > > > _______________________________________________ > CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> > To unsubscribe send an email to ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org> >
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
Ah, I wasn't personally aware of the distgit repos not mapping 1:1 with package name. I agree that there is something to fix here then. Thanks for clarifying
On Fri, Apr 6, 2018 at 11:00 AM, Stef Walter stefw@redhat.com wrote:
Three reasons:
dist-git repositories do not necessarily produce a package of the same name. They usually do.
dist-git repositories can produce multiple sub-packages, as you can see in the case of cockpit. Any and all of those sub-packages should be tested and/or gated by tests that live in that dist-git repo.
In this case there are 5 sub-packages from the cockpit dist-git repo in the Atomic Host that should trigger the pipeline.
Although I understand current effort is on the All Package Pipeline, I believe this might still be a problem going forward:
- There is no defined way to discover which dist-git repos to look for tests related to a certain package in the standard test specification.
Does that make sense? Dominik, did you have any ideas on how we might fix this?
Cheers,
Stef
On 06.04.2018 15:22, Johnny Bieren wrote:
I'm not sure what you mean by fix it. If it isn't part of atomic host, shouldn't it not be tested by the atomic host pipeline?
Cockpit *is* picked up by the All Package Pipeline. You can see one commit from cockpit here https://jenkins-continuous-infra.apps.ci.centos.org/blue/
organizations/jenkins/upstream-fedora-f28-pipeline/ detail/upstream-fedora-f28-pipeline/375/pipeline
Unfortunately, that build is many builds ago so the artifacts have been discarded so you cannot see the package test logs anymore. But, wouldn't this be the solution, not having the atomic pipeline test it?
The integration with pagure for the All Packages Pipeline is still WIP I believe
On Fri, Apr 6, 2018 at 7:35 AM, Stef Walter <stefw@redhat.com mailto:stefw@redhat.com> wrote:
On 03.04.2018 15:57, Johnny Bieren wrote: > Hey all, > > To clear up some of the questions on this thread: > > - Why no cockpit testing? > "cockpit" as a package name is not > in https://pagure.io/fedora-atomic/blob/f27/f/fedora-
atomic-host-base.json
<https://pagure.io/fedora-atomic/blob/f27/f/fedora-
atomic-host-base.json>
> for example, so if the distgit repo name is "cockpit" the commit by > design would not have been picked up by the Atomic Pipeline. How do we fix that? I guess this gap is because "Discovery" section of https://fedoraproject.org/wiki/Changes/InvokingTests <https://fedoraproject.org/wiki/Changes/InvokingTests> Does not address how to find the dist-git repo for a given
subpackage?
Stef > > - Atomic Pipeline doesn't test f28? > The Atomic pipeline is currently set up for f26 and f27. I have
been
> working on getting it working for f28, but I am hitting a bug and > haven't resolved it yet, so it is not ready yet. > > - What is the All Packages Pipeline > (https://jenkins-continuous-infra.apps.ci.centos.org/view/
Fedora%20All%20Packages%20Pipeline/
<https://jenkins-continuous-infra.apps.ci.centos.org/view/
Fedora%20All%20Packages%20Pipeline/>)
> ? > As alluded to, this is different than the Atomic pipelines. The Atomic > pipeline does the following: trigger on distgit commit to a repo of > interest (see question 1 for package lists), build rpm, compose ostree, > compose qcow2 atomic image with ostree, boot sanity tests, package > tests, atomic host integration tests, openshift e2e tests, report
all
> results from all stages on fedmsg under
org.centos.prod.ci.pipeline.*
> topics. The All Packages Pipelines do the following: trigger on distgit > commit to any package to branch [fXX | master] AND the repo
contains
> standard-test-roles defined tests in that branch, build the rpm in koji, > construct cloud qcow2 image while injecting the rpm at build time,
run
> package tests, report all results from all stages on fedmsg under > org.centos.prod.allpackages.pipeline.* topics. > > - Is the All Packages Pipeline production ready? > I think it is. If you look > here https://jenkins-continuous-infra.apps.ci.
centos.org/view/all/job/upstream-fedora-f28-pipeline/
<https://jenkins-continuous-infra.apps.ci.centos.org/view/
all/job/upstream-fedora-f28-pipeline/>
> it is passing when it should. There even seem to be some legitimate > package test failures found. If you go back in the build history,
you
> will likely find some bugs that have since been fixed, but most of
the
> latest builds are legit. We still need the linkage setup between
these
> fedmsgs and pagure, though, to the best of my knowledge. > > If I can clear anything else up, feel free to ask. > Best, > Johnny Bieren > > On Mon, Mar 26, 2018 at 4:21 PM, Dominik Perpeet <dperpeet@redhat.com <mailto:dperpeet@redhat.com> > <mailto:dperpeet@redhat.com <mailto:dperpeet@redhat.com>>> wrote: > > On 03/26/2018 05:34 PM, Ari LiVigni wrote: >> >> >> -== @ri ==- >> >> On Mon, Mar 26, 2018 at 10:40 AM, Pierre-Yves Chibon >> <pingou@pingoured.fr <mailto:pingou@pingoured.fr> <mailto:pingou@pingoured.fr <mailto:pingou@pingoured.fr>>> wrote: >> >> >> >> For one, I'd need a green light that the allpackages
pipeline
>> is running as >> expected and ready for production use. Then I'll see to adjust >> the tools for >> this new pipeline. >> This does mean we will announce it broadly and that we can >> expect its load to >> increase as more people opt-in. >> > > I agree. Let's make sure it's stable and production ready, then > announce it and help people when the inevitable issues and questions > arise. :) > >> >> Do we want to discontinue the Atomic CI one? I thought the >> idea was to have both >> running (see what I said above about the two pipelines
being
>> complementary). >> >> >> Let's have that discussion when Johnny is back from PTO. I am >> fine having both if it makes sense and there is no
duplication,
>> which I don't think there is since in the atomic one >> >> >> > I think both pipelines have merit, since they test different > deliverables. I would prefer to keep the Atomic one around, but also > have the "regular" one once that is stable. > > -Dominik > > > _______________________________________________ > CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> > <mailto:ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org>> > To unsubscribe send an email to ci-leave@lists.fedoraproject.
org
<mailto:ci-leave@lists.fedoraproject.org> > <mailto:ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org>> > > > > > _______________________________________________ > CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> > To unsubscribe send an email to ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org> >
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org
On Fri, Apr 6, 2018 at 11:00 AM, Stef Walter stefw@redhat.com wrote:
Three reasons:
dist-git repositories do not necessarily produce a package of the same name. They usually do.
dist-git repositories can produce multiple sub-packages, as you can see in the case of cockpit. Any and all of those sub-packages should be tested and/or gated by tests that live in that dist-git repo.
In this case there are 5 sub-packages from the cockpit dist-git repo in the Atomic Host that should trigger the pipeline.
Although I understand current effort is on the All Package Pipeline, I believe this might still be a problem going forward:
- There is no defined way to discover which dist-git repos to look for tests related to a certain package in the standard test specification.
Does that make sense? Dominik, did you have any ideas on how we might fix this?
To solve this problem for rpm, we can use the Source RPM header to map to a dist-git instead of using the rpm name.
Fred
On 06.04.2018 17:38, Frederic Lepied wrote:
On Fri, Apr 6, 2018 at 11:00 AM, Stef Walter <stefw@redhat.com mailto:stefw@redhat.com> wrote:
Three reasons: * dist-git repositories do not necessarily produce a package of the same name. They usually do. * dist-git repositories can produce multiple sub-packages, as you can see in the case of cockpit. Any and all of those sub-packages should be tested and/or gated by tests that live in that dist-git repo. In this case there are 5 sub-packages from the cockpit dist-git repo in the Atomic Host that should trigger the pipeline. Although I understand current effort is on the All Package Pipeline, I believe this might still be a problem going forward: * There is no defined way to discover which dist-git repos to look for tests related to a certain package in the standard test specification. Does that make sense? Dominik, did you have any ideas on how we might fix this?
To solve this problem for rpm, we can use the Source RPM header to map to a dist-git instead of using the rpm name.
Nice and clean. Anyone up for proposing the wording to add this to the spec in the "Discovery" section?
Stef
On Fri, Apr 06, 2018 at 11:38:26AM -0400, Frederic Lepied wrote:
On Fri, Apr 6, 2018 at 11:00 AM, Stef Walter stefw@redhat.com wrote:
Three reasons:  * dist-git repositories do not necessarily produce a package of the   same name. They usually do.  * dist-git repositories can produce multiple sub-packages, as you can   see in the case of cockpit. Any and all of those sub-packages should   be tested and/or gated by tests that live in that dist-git repo. In this case there are 5 sub-packages from the cockpit dist-git repo in the Atomic Host that should trigger the pipeline. Although I understand current effort is on the All Package Pipeline, I believe this might still be a problem going forward:  * There is no defined way to discover which dist-git repos to look for   tests related to a certain package in the standard test   specification. Does that make sense? Dominik, did you have any ideas on how we might fix this?
To solve this problem for rpm, we can use the Source RPM header to map to a dist-git instead of using the rpm name.
That info being present in the repo metadata, mdapi may be a potential provide of the info: https://apps.fedoraproject.org/mdapi/
Pierre
On 04/06/2018 11:00 AM, Stef Walter wrote:
- There is no defined way to discover which dist-git repos to look for tests related to a certain package in the standard test specification.
dnf can tell you which source package created an RPM:
$ dnf repoquery -s bodhi-client bodhi-3.5.2-1.fc28.src.rpm
On 03/26/2018 07:05 AM, Ari LiVigni wrote:
-== @ri ==-
On Mon, Mar 26, 2018 at 8:47 AM, Pierre-Yves Chibon <pingou@pingoured.fr mailto:pingou@pingoured.fr> wrote:
On Mon, Mar 26, 2018 at 02:41:41PM +0200, Pierre-Yves Chibon wrote: > On Mon, Mar 26, 2018 at 02:06:32PM +0200, Stef Walter wrote: > > Hi there, > > > > Is the CI pipeline running on Fedora and gating? It's really hard to > > find test results. > > > > I'm looking around both in Bodhi: > > > https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18 <https://bodhi.fedoraproject.org/updates/FEDORA-2018-5c64b23a18> > > Will look into this one And the answer is easy: the CI pipeline does not run for F28: https://jenkins-continuous-infra.apps.ci.centos.org/ <https://jenkins-continuous-infra.apps.ci.centos.org/>
That is incorrect and Johnny reported the status last week to multiple lists and people:
https://jenkins-continuous-infra.apps.ci.centos.org/job/upstream-fedora-f28-...
I will make it more visible on the default view
How can we get our package's tests added to this pipeline?
-Tom
Pierre _______________________________________________ CI mailing list -- ci@lists.fedoraproject.org <mailto:ci@lists.fedoraproject.org> To unsubscribe send an email to ci-leave@lists.fedoraproject.org <mailto:ci-leave@lists.fedoraproject.org>
CI mailing list -- ci@lists.fedoraproject.org To unsubscribe send an email to ci-leave@lists.fedoraproject.org