[OpenDaylight TSC] CSIT status in proposed topic:neon-mri ("Weather Item" TSC-132)


Michael Vorburger <vorburger@...>
 

-SFC but +aaa & Tom & Robert,

On Tue, Oct 9, 2018 at 6:35 PM Michael Vorburger <vorburger@...> wrote:
On Tue, Oct 9, 2018 at 6:18 PM Jamo Luhrsen <jluhrsen@...> wrote:
+netconf, +sfc

On 10/9/18 8:45 AM, Michael Vorburger wrote:
> [starting a new thread as we're now moving from netvirt to general, Cc TSC]
>
> With the build problem below in netvirt having been fixed a few ago today (by Stephen; thank you!), we have now finally
> reached the stage where we have a passing multipatch quick build for Neon-MRI in
> https://jira.opendaylight.org/browse/TSC-132, and can run CSIT on it. (We do not however have a passing full non-quick
> build just yet.)
>
> Jamo, https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-neon/62/ has a lot of red
> though, want to and add people/lists of any affected projects which should look at this now?

yeah, something basic is broken. from the CSIT level it looks like restconf
is broken, where there is no ability to open a connection on port 8181.

but, looking at a karaf.log, I think something even more basic is broken,
maybe in SFC?

2018-10-09T12:38:24,657 | ERROR | pool-1-thread-2  | BootFeaturesInstaller            | 8 -
org.apache.karaf.features.core - 4.1.5 | Error installing boot features
org.apache.karaf.features.internal.util.MultiException: Error:
        Unable to create resource for bundle mvn:org.opendaylight.sfc/sfc-sb-rest/0.9.0-SNAPSHOT

Caused by: org.osgi.framework.BundleException: Unable to build resource for mvn:org.opendaylight.sfc/sfc-sb-rest/0.9.0-SNAPSHOT: Duplicate import: org.opendaylight.yangtools.util.concurrent


This should have been seen by SFT when locally running (without -Pq) to prepare this... perhaps running CSIT on the quick build is actually premature, and what we really should do is focus on getting a full build to pass, and only then run CSIT? I'll kick a new one off shortly, and keep updating https://jira.opendaylight.org/browse/TSC-132.

Jamo and Ijust had a look at https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-neon/63/, and he from there finds https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/aaa-csit-1node-authn-all-neon/59/odl_1/odl1_karaf.log.gz with this new beauty, looks like some problem in AAA re. BouncyCastle version?! The SFT in AAA should have caught this, but apparently didn't... perhaps it only appears in an integration/distribution Karaf? Tom or Robert, is this something you would be willing to dig more into?
2018-10-09T19:48:59,694 | ERROR | pool-1-thread-2  | BootFeaturesInstaller            | 8 - org.apache.karaf.features.core - 4.1.5 | Error installing boot features
org.osgi.service.resolver.ResolutionException: Unable to resolve root: missing requirement [root] osgi.identity; osgi.identity=odl-integration-compatible-with-all; type=karaf.feature; version="[0.10.0.SNAPSHOT,0.10.0.SNAPSHOT]"; filter:="(&(osgi.identity=odl-integration-compatible-with-all)(type=karaf.feature)(version>=0.10.0.SNAPSHOT)(version<=0.10.0.SNAPSHOT))" [caused by: Unable to resolve odl-integration-compatible-with-all/0.10.0.SNAPSHOT: missing requirement [odl-integration-compatible-with-all/0.10.0.SNAPSHOT] osgi.identity; osgi.identity=odl-sfc-ovs; type=karaf.feature; version="[0.9.0.SNAPSHOT,0.9.0.SNAPSHOT]" [caused by: Unable to resolve odl-sfc-ovs/0.9.0.SNAPSHOT: missing requirement [odl-sfc-ovs/0.9.0.SNAPSHOT] osgi.identity; osgi.identity=org.opendaylight.sfc.ovs; type=osgi.bundle; version="[0.9.0.SNAPSHOT,0.9.0.SNAPSHOT]"; resolution:=mandatory [caused by: Unable to resolve org.opendaylight.sfc.ovs/0.9.0.SNAPSHOT: missing requirement [org.opendaylight.sfc.ovs/0.9.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.opendaylight.ovsdb.southbound)(version>=1.8.0)(!(version>=2.0.0)))" [caused by: Unable to resolve org.opendaylight.ovsdb.southbound-impl/1.8.0.SNAPSHOT: missing requirement [org.opendaylight.ovsdb.southbound-impl/1.8.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.opendaylight.ovsdb.lib)(version>=1.8.0)(!(version>=2.0.0)))" [caused by: Unable to resolve org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT: missing requirement [org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.opendaylight.aaa.cert.api)(version>=0.9.0)(!(version>=1.0.0)))" [caused by: Unable to resolve org.opendaylight.aaa.cert/0.9.0.SNAPSHOT: missing requirement [org.opendaylight.aaa.cert/0.9.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.bouncycastle.asn1)(version>=1.60.0)(!(version>=2.0.0)))"]]]]]]
	at org.apache.felix.resolver.ResolutionError.toException(ResolutionError.java:42) ~[?:?]
	at org.apache.felix.resolver.ResolverImpl.doResolve(ResolverImpl.java:391) ~[?:?]
	at org.apache.felix.resolver.ResolverImpl.resolve(ResolverImpl.java:377) ~[?:?]
	at org.apache.felix.resolver.ResolverImpl.resolve(ResolverImpl.java:349) ~[?:?]
	at org.apache.karaf.features.internal.region.SubsystemResolver.resolve(SubsystemResolver.java:218) ~[?:?]
	at org.apache.karaf.features.internal.service.Deployer.deploy(Deployer.java:291) ~[?:?]
	at org.apache.karaf.features.internal.service.FeaturesServiceImpl.doProvision(FeaturesServiceImpl.java:1248) ~[?:?]
	at org.apache.karaf.features.internal.service.FeaturesServiceImpl.lambda$doProvisionInThread$1(FeaturesServiceImpl.java:1147) ~[?:?]
	at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:?]
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) [?:?]
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) [?:?] 

from this log:
https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/aaa-csit-1node-authn-all-neon/58/odl_1/odl1_karaf.log.gz

yeah, seeing the same thing in the netconf job.

JamO

> Tx,
> M.
> --
> Michael Vorburger, Red Hat
> vorburger@... <mailto:vorburger@...> | IRC: vorburger @freenode | ~ = http://vorburger.ch
> <http://vorburger.ch/>
>
>
> ---------- Forwarded message ---------
> From: *Vishal Thapar* <vthapar@... <mailto:vthapar@...>>
> Date: Tue, Oct 9, 2018 at 7:52 AM
> Subject: Re: [netvirt-dev] NETVIRT-1425 netvirt build failure in proposed topic:neon-mri ("Weather Item" TSC-132)
> To: Michael Vorburger <vorburger@... <mailto:vorburger@...>>
> Cc: Sam Hague <shague@... <mailto:shague@...>>, Stephen Kitt <skitt@... <mailto:skitt@...>>,
> odl netvirt dev <netvirt-dev@... <mailto:netvirt-dev@...>>
>
>
>
>
> On Tue, Oct 9, 2018 at 7:00 AM Michael Vorburger <vorburger@... <mailto:vorburger@...>> wrote:
>
>     On Mon, Oct 8, 2018 at 10:25 PM Sam Hague <shague@... <mailto:shague@...>> wrote:
>
>         thanks Vishal! Do we want to remove the wip now?
>
>         Where do we stand on the full multipatch and not the fast? I know the OSGI patch never passed SFT since there
>         was always a missing injection somewhere.
>
>     Latest status now tracked on https://jira.opendaylight.org/browse/TSC-132 - summary: Until 1/2h ago, Neutron was
>     broken on Neon-MRI; that is now fixed (full details in NEUTRON-197
>     <https://jira.opendaylight.org/browse/NEUTRON-197> & NEUTRON-200 <https://jira.opendaylight.org/browse/NEUTRON-200>;
>     what a PITA), and if we could get a passing quick or even better full build, we should start running CSITs for
>     Neon-MRI to see how things are looking.
>
>         On Mon, Oct 8, 2018 at 6:50 AM Vishal Thapar <vthapar@... <mailto:vthapar@...>> wrote:
>
>             Netvirt passing now:
>
>             https://git.opendaylight.org/gerrit/#/c/76476/
>
>
>     but https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-neon/60/ failed, see
>     https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/integration-multipatch-test-neon/60/console.log.gz, it's
>     the blueprint-maven-plugin problem from https://jira.opendaylight.org/browse/ODLPARENT-167 hitting
>     neutronvpn-impl... odlparent 4.0.2 instead of 4.0.0 should have fixed that - huh?
>
>     I had a bit of a look, and noticed that it ran blueprint-maven-plugin:1.4.0 - that is wrong, it should be 1.10.0 ...
>     I can't quite see why though, because neutronvpn-impl uses org.opendaylight.mdsal:binding-parent, via
>     netvirt:binding-parent, which at 3.0.0-SNAPSHOT that
>     https://git.opendaylight.org/gerrit/#/c/76476/7/commons/binding-parent/pom.xml is on already has odlparent 4.0.2...
>     hm. BTW: You should switch mdsal 3.0.0-SNAPSHOT, but I don't think it will help for this problem (but it will avoid
>     others after we're past this).
>
> Answer to this question at least is in the very file yo mentioned which has following entry further down:
>            <!-- Support Blueprint XML construction using annotations -->
>            <groupId>org.apache.aries.blueprint</groupId>
>            <artifactId>blueprint-maven-plugin</artifactId>
>            <version>1.4.0</version>
>            <configuration>
> I tried with version removed and set to 1.10.0 both, neither helped with this particular failure.
>
>
>     Very curious though that the previous
>     https://jenkins.opendaylight.org/releng/job/integration-multipatch-test-neon/59/ passes - I can't quite explain that.
>
>             On Mon, Oct 8, 2018 at 2:03 PM Vishal Thapar <vthapar@... <mailto:vthapar@...>> wrote:
>
>                 Working on it. Revision needs to change. I'll also pull in Sam's OSGI-INF changes into this so we have
>                 single patch for all neon-mri changes.
>
>                 On Sun, Oct 7, 2018 at 5:18 AM Michael Vorburger <vorburger@... <mailto:vorburger@...>> wrote:
>
>                     Stephen, Sam,
>
>                     https://logs.opendaylight.org/releng/vex-yul-odl-jenkins-1/integration-multipatch-test-neon/56/console.log.gz
>                     now fails in netvirt - will you look into and update https://git.opendaylight.org/gerrit/#/c/76476/ ?
>
>                     [ERROR] /w/workspace/integration-multipatch-test-neon/patch_tester/netvirt/aclservice/impl/src/main/java/org/opendaylight/netvirt/aclservice/listeners/AclInterfaceStateListener.java:[29,88] package org.opendaylight.yang.gen.v1.urn.ietf.params.xml.ns.yang.iana._if.type.rev140508 does not exist
>                     [ERROR] /w/workspace/integration-multipatch-test-neon/patch_tester/netvirt/aclservice/impl/src/main/java/org/opendaylight/netvirt/aclservice/listeners/AclInterfaceStateListener.java:[91,14] cannot find symbol
>                        symbol:   class L2vlan
>                        location: class org.opendaylight.netvirt.aclservice.listeners.AclInterfaceStateListener
>                     [ERROR] /w/workspace/integration-multipatch-test-neon/patch_tester/netvirt/aclservice/impl/src/main/java/org/opendaylight/netvirt/aclservice/listeners/AclInterfaceStateListener.java:[121,41] cannot find symbol
>                        symbol:   class L2vlan
>                        location: class org.opendaylight.netvirt.aclservice.listeners.AclInterfaceStateListener
>                     [ERROR] /w/workspace/integration-multipatch-test-neon/patch_tester/netvirt/aclservice/impl/src/main/java/org/opendaylight/netvirt/aclservice/listeners/AclInterfaceStateListener.java:[130,14] cannot find symbol
>                        symbol:   class L2vlan
>                        location: class org.opendaylight.netvirt.aclservice.listeners.AclInterfaceStateListener
>                     [INFO] 4 errors
>
>                     Tx,
>                     M.
>                     --
>                     Michael Vorburger, Red Hat
>                     vorburger@... <mailto:vorburger@...> | IRC: vorburger @freenode | ~ =
>                     http://vorburger.ch <http://vorburger.ch/>
>                     _______________________________________________
>                     netvirt-dev mailing list
>                     netvirt-dev@... <mailto:netvirt-dev@...>
>                     https://lists.opendaylight.org/mailman/listinfo/netvirt-dev
>
>
> _______________________________________________
> TSC mailing list
> TSC@...
> https://lists.opendaylight.org/mailman/listinfo/tsc


Robert Varga
 

On 10/10/2018 00:42, Michael Vorburger wrote:
The SFT in AAA should have caught this, but apparently didn't... perhaps
it only appears in an integration/distribution Karaf? Tom or Robert, is
this something you would be willing to dig more into?
int/dist is just another project, hence it obviously needs a neon-mri
patch: https://git.opendaylight.org/gerrit/76817.

Bye,
Robert


Michael Vorburger <vorburger@...>
 

On Wed, Oct 10, 2018 at 12:30 PM Robert Varga <nite@...> wrote:
On 10/10/2018 00:42, Michael Vorburger wrote:
> The SFT in AAA should have caught this, but apparently didn't... perhaps
> it only appears in an integration/distribution Karaf? Tom or Robert, is
> this something you would be willing to dig more into?

int/dist is just another project, hence it obviously needs a neon-mri
patch: https://git.opendaylight.org/gerrit/76817.

Oh! Right. - Thanks a lot for puttting that together.

Luis, I want to run a new multipatch job incl. c/76817 ... have just kicked off new https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-neon/66, where I manually added "integration/distribution" to "BUILD_ORDER" ... is that correct? If yes, do we need to add that in the default (similar to https://git.opendaylight.org/gerrit/#/c/76724) as well now? If not, confirm if this is automatic/implicit anyway?

PS: RELENG-135 makes me a little sceptical how well this actually work, but let's discuss that further on the other thread "Build failures, aren't we making a big mistake by not isolating version bumping multipatch jobs from new SNAPSHOTs?" instead.


Stephen Kitt <skitt@...>
 

On Wed, 10 Oct 2018 13:35:29 +0200
Michael Vorburger <vorburger@...> wrote:
On Wed, Oct 10, 2018 at 12:30 PM Robert Varga <nite@...> wrote:
int/dist is just another project, hence it obviously needs a
neon-mri patch: https://git.opendaylight.org/gerrit/76817.
Oh! Right. - Thanks a lot for puttting that together.

Luis, I want to run a new multipatch job incl. c/76817 ... have just
kicked off new
https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-neon/66,
where I manually added "integration/distribution" to
"BUILD_ORDER" ... is that correct? If yes, do we need to add that in
the default (similar to
https://git.opendaylight.org/gerrit/#/c/76724) as well now? If not,
confirm if this is automatic/implicit anyway?
Since 76817 is in the neon-mri topic, it should be picked up by
multipatch automatically.

Regards,

Stephen


Luis Gomez
 

On Oct 10, 2018, at 6:36 AM, Stephen Kitt <skitt@...> wrote:

On Wed, 10 Oct 2018 13:35:29 +0200
Michael Vorburger <vorburger@...> wrote:
On Wed, Oct 10, 2018 at 12:30 PM Robert Varga <nite@...> wrote:
int/dist is just another project, hence it obviously needs a
neon-mri patch: https://git.opendaylight.org/gerrit/76817.
Oh! Right. - Thanks a lot for puttting that together.

Luis, I want to run a new multipatch job incl. c/76817 ... have just
kicked off new
https://jenkins.opendaylight.org/releng/view/integration/job/integration-multipatch-test-neon/66,
where I manually added "integration/distribution" to
"BUILD_ORDER" ... is that correct? If yes, do we need to add that in
the default (similar to
https://git.opendaylight.org/gerrit/#/c/76724) as well now? If not,
confirm if this is automatic/implicit anyway?
Since 76817 is in the neon-mri topic, it should be picked up by
multipatch automatically.
+1, distribution is always added at the end so no need to specify in the BUILD_ORDER. If there is an int/dist patch in the patch list or the topic it will grab that, otherwise it will grab the job stream code (e.g. neon=master).


Regards,

Stephen


Michael Vorburger <vorburger@...>
 

- netconf-dev, as it's no longer their problem:

On Wed, Oct 10, 2018 at 1:35 PM Michael Vorburger <vorburger@...> wrote:
On Wed, Oct 10, 2018 at 12:30 PM Robert Varga <nite@...> wrote:
On 10/10/2018 00:42, Michael Vorburger wrote:
> The SFT in AAA should have caught this, but apparently didn't... perhaps
> it only appears in an integration/distribution Karaf? Tom or Robert, is
> this something you would be willing to dig more into?

int/dist is just another project, hence it obviously needs a neon-mri
patch: https://git.opendaylight.org/gerrit/76817.

Oh! Right. - Thanks a lot for puttting that together.

Luis, I want to run a new multipatch job incl. c/76817 ... have just kicked off new

after having sorted out a few other problems, we're unfortunately still back to this issue:

2018-10-10T19:21:48,049 | ERROR | pool-1-thread-2  | BootFeaturesInstaller            | 8 - org.apache.karaf.features.core - 4.1.5 | Error installing boot features
org.osgi.service.resolver.ResolutionException: Unable to resolve root: missing requirement [root] osgi.identity; osgi.identity=odl-integration-compatible-with-all; type=karaf.feature; version="[0.10.0.SNAPSHOT,0.10.0.SNAPSHOT]"; filter:="(&(osgi.identity=odl-integration-compatible-with-all)(type=karaf.feature)(version>=0.10.0.SNAPSHOT)(version<=0.10.0.SNAPSHOT))" [caused by: Unable to resolve odl-integration-compatible-with-all/0.10.0.SNAPSHOT: missing requirement [odl-integration-compatible-with-all/0.10.0.SNAPSHOT] osgi.identity; osgi.identity=odl-ovsdb-southbound-impl; type=karaf.feature; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]" [caused by: Unable to resolve odl-ovsdb-southbound-impl/1.8.0.SNAPSHOT: missing requirement [odl-ovsdb-southbound-impl/1.8.0.SNAPSHOT] osgi.identity; osgi.identity=odl-ovsdb-library; type=karaf.feature; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]" [caused by: Unable to resolve odl-ovsdb-library/1.8.0.SNAPSHOT: missing requirement [odl-ovsdb-library/1.8.0.SNAPSHOT] osgi.identity; osgi.identity=org.opendaylight.ovsdb.library; type=osgi.bundle; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]"; resolution:=mandatory [caused by: Unable to resolve org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT: missing requirement [org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.opendaylight.aaa.cert.api)(version>=0.9.0)(!(version>=1.0.0)))" [caused by: Unable to resolve org.opendaylight.aaa.cert/0.9.0.SNAPSHOT: missing requirement [org.opendaylight.aaa.cert/0.9.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.bouncycastle.asn1)(version>=1.60.0)(!(version>=2.0.0)))"]]]]]


Luis Gomez
 

Just to be clear, this is not a multi-patch job issue but a controller version integration issue, right?

On Oct 10, 2018, at 2:47 PM, Michael Vorburger <vorburger@...> wrote:

- netconf-dev, as it's no longer their problem:

On Wed, Oct 10, 2018 at 1:35 PM Michael Vorburger <vorburger@...> wrote:
On Wed, Oct 10, 2018 at 12:30 PM Robert Varga <nite@...> wrote:
On 10/10/2018 00:42, Michael Vorburger wrote:
> The SFT in AAA should have caught this, but apparently didn't... perhaps
> it only appears in an integration/distribution Karaf? Tom or Robert, is
> this something you would be willing to dig more into?

int/dist is just another project, hence it obviously needs a neon-mri
patch: https://git.opendaylight.org/gerrit/76817.

Oh! Right. - Thanks a lot for puttting that together.

Luis, I want to run a new multipatch job incl. c/76817 ... have just kicked off new

after having sorted out a few other problems, we're unfortunately still back to this issue:

2018-10-10T19:21:48,049 | ERROR | pool-1-thread-2  | BootFeaturesInstaller            | 8 - org.apache.karaf.features.core - 4.1.5 | Error installing boot features
org.osgi.service.resolver.ResolutionException: Unable to resolve root: missing requirement [root] osgi.identity; osgi.identity=odl-integration-compatible-with-all; type=karaf.feature; version="[0.10.0.SNAPSHOT,0.10.0.SNAPSHOT]"; filter:="(&(osgi.identity=odl-integration-compatible-with-all)(type=karaf.feature)(version>=0.10.0.SNAPSHOT)(version<=0.10.0.SNAPSHOT))" [caused by: Unable to resolve odl-integration-compatible-with-all/0.10.0.SNAPSHOT: missing requirement [odl-integration-compatible-with-all/0.10.0.SNAPSHOT] osgi.identity; osgi.identity=odl-ovsdb-southbound-impl; type=karaf.feature; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]" [caused by: Unable to resolve odl-ovsdb-southbound-impl/1.8.0.SNAPSHOT: missing requirement [odl-ovsdb-southbound-impl/1.8.0.SNAPSHOT] osgi.identity; osgi.identity=odl-ovsdb-library; type=karaf.feature; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]" [caused by: Unable to resolve odl-ovsdb-library/1.8.0.SNAPSHOT: missing requirement [odl-ovsdb-library/1.8.0.SNAPSHOT] osgi.identity; osgi.identity=org.opendaylight.ovsdb.library; type=osgi.bundle; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]"; resolution:=mandatory [caused by: Unable to resolve org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT: missing requirement [org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.opendaylight.aaa.cert.api)(version>=0.9.0)(!(version>=1.0.0)))" [caused by: Unable to resolve org.opendaylight.aaa.cert/0.9.0.SNAPSHOT: missing requirement [org.opendaylight.aaa.cert/0.9.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.bouncycastle.asn1)(version>=1.60.0)(!(version>=2.0.0)))"]]]]]



Robert Varga
 

features-core-4.1.5 ... wrong base karaf distro, pulled from odlparent-3.1.0-3, not 4.0.2.

Sent from my BlackBerry - the most secure mobile device - via the Orange Network
From: ecelgp@...

Sent: October 10, 2018 23:59
To: vorburger@...
Cc: nite@...; aaa-dev@...; tompantelis@...; netvirt-dev@...; tsc@...; netconf-dev@...; skitt@...; jluhrsen@...; jluhrsen@...
Subject: Re: [OpenDaylight TSC] CSIT status in proposed topic:neon-mri ("Weather Item" TSC-132)

Just to be clear, this is not a multi-patch job issue but a controller version integration issue, right?

On Oct 10, 2018, at 2:47 PM, Michael Vorburger <vorburger@...> wrote:

- netconf-dev, as it's no longer their problem:

On Wed, Oct 10, 2018 at 1:35 PM Michael Vorburger <vorburger@...> wrote:
On Wed, Oct 10, 2018 at 12:30 PM Robert Varga <nite@...> wrote:
On 10/10/2018 00:42, Michael Vorburger wrote:
> The SFT in AAA should have caught this, but apparently didn't... perhaps
> it only appears in an integration/distribution Karaf? Tom or Robert, is
> this something you would be willing to dig more into?

int/dist is just another project, hence it obviously needs a neon-mri
patch: https://git.opendaylight.org/gerrit/76817.

Oh! Right. - Thanks a lot for puttting that together.

Luis, I want to run a new multipatch job incl. c/76817 ... have just kicked off new

after having sorted out a few other problems, we're unfortunately still back to this issue:

2018-10-10T19:21:48,049 | ERROR | pool-1-thread-2  | BootFeaturesInstaller            | 8 - org.apache.karaf.features.core - 4.1.5 | Error installing boot features
org.osgi.service.resolver.ResolutionException: Unable to resolve root: missing requirement [root] osgi.identity; osgi.identity=odl-integration-compatible-with-all; type=karaf.feature; version="[0.10.0.SNAPSHOT,0.10.0.SNAPSHOT]"; filter:="(&(osgi.identity=odl-integration-compatible-with-all)(type=karaf.feature)(version>=0.10.0.SNAPSHOT)(version<=0.10.0.SNAPSHOT))" [caused by: Unable to resolve odl-integration-compatible-with-all/0.10.0.SNAPSHOT: missing requirement [odl-integration-compatible-with-all/0.10.0.SNAPSHOT] osgi.identity; osgi.identity=odl-ovsdb-southbound-impl; type=karaf.feature; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]" [caused by: Unable to resolve odl-ovsdb-southbound-impl/1.8.0.SNAPSHOT: missing requirement [odl-ovsdb-southbound-impl/1.8.0.SNAPSHOT] osgi.identity; osgi.identity=odl-ovsdb-library; type=karaf.feature; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]" [caused by: Unable to resolve odl-ovsdb-library/1.8.0.SNAPSHOT: missing requirement [odl-ovsdb-library/1.8.0.SNAPSHOT] osgi.identity; osgi.identity=org.opendaylight.ovsdb.library; type=osgi.bundle; version="[1.8.0.SNAPSHOT,1.8.0.SNAPSHOT]"; resolution:=mandatory [caused by: Unable to resolve org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT: missing requirement [org.opendaylight.ovsdb.library/1.8.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.opendaylight.aaa.cert.api)(version>=0.9.0)(!(version>=1.0.0)))" [caused by: Unable to resolve org.opendaylight.aaa.cert/0.9.0.SNAPSHOT: missing requirement [org.opendaylight.aaa.cert/0.9.0.SNAPSHOT] osgi.wiring.package; filter:="(&(osgi.wiring.package=org.bouncycastle.asn1)(version>=1.60.0)(!(version>=2.0.0)))"]]]]]



Robert Varga
 

On 11/10/2018 00:02, Robert Varga wrote:
features-core-4.1.5 ... wrong base karaf distro, pulled from
odlparent-3.1.0-3, not 4.0.2.
perhaps
https://git.opendaylight.org/gerrit/#/c/76817/4..5/artifacts/upstream/properties/pom.xml

Bye,
Robert


Robert Varga
 

On 11/10/2018 00:13, Robert Varga wrote:
On 11/10/2018 00:02, Robert Varga wrote:
features-core-4.1.5 ... wrong base karaf distro, pulled from
odlparent-3.1.0-3, not 4.0.2.
perhaps
https://git.opendaylight.org/gerrit/#/c/76817/4..5/artifacts/upstream/properties/pom.xml
And if it's not that, neon-mri patches need to be audited. I am pretty
sure controller-netconf are okay.

Bye,
Robert


Robert Varga
 

On 11/10/2018 00:18, Robert Varga wrote:
On 11/10/2018 00:13, Robert Varga wrote:
On 11/10/2018 00:02, Robert Varga wrote:
features-core-4.1.5 ... wrong base karaf distro, pulled from
odlparent-3.1.0-3, not 4.0.2.
perhaps
https://git.opendaylight.org/gerrit/#/c/76817/4..5/artifacts/upstream/properties/pom.xml
And if it's not that, neon-mri patches need to be audited. I am pretty
sure controller-netconf are okay.
RTFC:

00:02:10.037 adding integration/distribution
00:02:10.039 Cloning into 'distribution'...
00:02:12.545 Already on 'master'
00:02:12.545 Your branch is up to date with 'origin/master'.
00:02:12.546
/w/workspace/integration-multipatch-test-neon/patch_tester/infrautils
/w/workspace/integration-multipatch-test-neon/patch_tester
00:02:12.856 Apache Maven 3.5.3
(3383c37e1f9e9b3bc3df5050c29c8aff9f295297; 2018-02-24T19:49:05Z)

i.e. int/dist is not patched. Luis, back to you :)

Bye,
Robert


Luis Gomez
 

OK, it seems like int/dist project is not included in the logic resolving a topic, let me fix that quickly. It is also weird this miss was not detected before...

On Oct 10, 2018, at 3:21 PM, Robert Varga <nite@...> wrote:

On 11/10/2018 00:18, Robert Varga wrote:
On 11/10/2018 00:13, Robert Varga wrote:
On 11/10/2018 00:02, Robert Varga wrote:
features-core-4.1.5 ... wrong base karaf distro, pulled from
odlparent-3.1.0-3, not 4.0.2.
perhaps
https://git.opendaylight.org/gerrit/#/c/76817/4..5/artifacts/upstream/properties/pom.xml
And if it's not that, neon-mri patches need to be audited. I am pretty
sure controller-netconf are okay.
RTFC:

00:02:10.037 adding integration/distribution
00:02:10.039 Cloning into 'distribution'...
00:02:12.545 Already on 'master'
00:02:12.545 Your branch is up to date with 'origin/master'.
00:02:12.546
/w/workspace/integration-multipatch-test-neon/patch_tester/infrautils
/w/workspace/integration-multipatch-test-neon/patch_tester
00:02:12.856 Apache Maven 3.5.3
(3383c37e1f9e9b3bc3df5050c29c8aff9f295297; 2018-02-24T19:49:05Z)

i.e. int/dist is not patched. Luis, back to you :)

Bye,
Robert


Robert Varga
 

On 11/10/2018 00:49, Luis Gomez wrote:
It is also weird this miss was not detected before...
I believe we weren't upgrading karaf using this process -- both Fluorine
and Oxygen are shipping odlparent-3.1.x.

Regards,
Robert


Luis Gomez
 

On Oct 10, 2018, at 3:53 PM, Robert Varga <nite@...> wrote:

On 11/10/2018 00:49, Luis Gomez wrote:
It is also weird this miss was not detected before...
I believe we weren't upgrading karaf using this process -- both Fluorine
and Oxygen are shipping odlparent-3.1.x.

Regards,
Robert


Michael Vorburger <vorburger@...>
 

Hello,

On Thu, Oct 11, 2018 at 1:12 AM Luis Gomez <ecelgp@...> wrote:
This should fix:

https://git.opendaylight.org/gerrit/#/c/76839/

 
I am also running this job in sandbox to verify:

https://jenkins.opendaylight.org/sandbox/job/integration-multipatch-test-neon/1

this failed with a new error, is this just because a ran on sandbox or a real new problem:
[ERROR] Failed to execute goal org.apache.maven.plugins:maven-antrun-plugin:1.8:run (default) on project karaf: An Ant BuildException has occured: Directory does not exist: /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/assembly/system/mysql
[ERROR] around Ant part ...<delete includeemptydirs="true">... @ 4:35 in /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/antrun/build-main.xml
[ERROR] -> [Help 1]
org.apache.maven.lifecycle.LifecycleExecutionException: Failed to execute goal org.apache.maven.plugins:maven-antrun-plugin:1.8:run (default) on project karaf: An Ant BuildException has occured: Directory does not exist: /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/assembly/system/mysql
around Ant part ...<delete includeemptydirs="true">... @ 4:35 in /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/antrun/build-main.xml
This mysql business sounds vaguely familiar - is this the thing we did to yank out (remove) the mysql JDBC client because of license concerns? MAYBE that is no longer there now (check please), because the newer version of Karaf in the new odlparent has picked up a new PAX JDBC where "someone" (me!) had done https://ops4j1.jira.com/browse/PAXJDBC-128 to remove that at source upstream? If that has trickled down and finally reached us now, we can probably remove the Ant <delete> hack we had to do at the time. Will someone have a closer look at this and confirm?
 
BR/Luis


> On Oct 10, 2018, at 3:53 PM, Robert Varga <nite@...> wrote:
>
> On 11/10/2018 00:49, Luis Gomez wrote:
>> It is also weird this miss was not detected before...
>
> I believe we weren't upgrading karaf using this process -- both Fluorine
> and Oxygen are shipping odlparent-3.1.x.
>
> Regards,
> Robert


Luis Gomez
 

Yeah, I also saw the error, which BTW I have never seen before but on the other hand it is very unlike that this is introduced by my patch that only thing that does is to add int/dist to the topic patch list.

On Oct 11, 2018, at 3:46 AM, Michael Vorburger <vorburger@...> wrote:

Hello,

On Thu, Oct 11, 2018 at 1:12 AM Luis Gomez <ecelgp@...> wrote:
This should fix:

https://git.opendaylight.org/gerrit/#/c/76839/

 
I am also running this job in sandbox to verify:

https://jenkins.opendaylight.org/sandbox/job/integration-multipatch-test-neon/1

this failed with a new error, is this just because a ran on sandbox or a real new problem:
[ERROR] Failed to execute goal org.apache.maven.plugins:maven-antrun-plugin:1.8:run (default) on project karaf: An Ant BuildException has occured: Directory does not exist: /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/assembly/system/mysql
[ERROR] around Ant part ...<delete includeemptydirs="true">... @ 4:35 in /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/antrun/build-main.xml
[ERROR] -> [Help 1]
org.apache.maven.lifecycle.LifecycleExecutionException: Failed to execute goal org.apache.maven.plugins:maven-antrun-plugin:1.8:run (default) on project karaf: An Ant BuildException has occured: Directory does not exist: /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/assembly/system/mysql
around Ant part ...<delete includeemptydirs="true">... @ 4:35 in /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/antrun/build-main.xml
This mysql business sounds vaguely familiar - is this the thing we did to yank out (remove) the mysql JDBC client because of license concerns? MAYBE that is no longer there now (check please), because the newer version of Karaf in the new odlparent has picked up a new PAX JDBC where "someone" (me!) had done https://ops4j1.jira.com/browse/PAXJDBC-128 to remove that at source upstream? If that has trickled down and finally reached us now, we can probably remove the Ant <delete> hack we had to do at the time. Will someone have a closer look at this and confirm?
 
BR/Luis


> On Oct 10, 2018, at 3:53 PM, Robert Varga <nite@...> wrote:
>
> On 11/10/2018 00:49, Luis Gomez wrote:
>> It is also weird this miss was not detected before...
>
> I believe we weren't upgrading karaf using this process -- both Fluorine
> and Oxygen are shipping odlparent-3.1.x.
>
> Regards,
> Robert


Michael Vorburger <vorburger@...>
 

On Thu, Oct 11, 2018 at 7:08 PM Luis Gomez <ecelgp@...> wrote:
Yeah, I also saw the error, which BTW I have never seen before

Luis, FTR Stephen fixed the problem below in https://git.opendaylight.org/gerrit/#/c/76817/5..6/karaf/pom.xml (TX skitt!)
 
but on the other hand it is very unlike that this is introduced by my patch that only thing that does is to add int/dist to the topic patch list.

nobody said your patch introduced it; it's an impact of the new Karaf in the new odlparent. 

The first, but may be not the last one? ;-)

On Oct 11, 2018, at 3:46 AM, Michael Vorburger <vorburger@...> wrote:

Hello,

On Thu, Oct 11, 2018 at 1:12 AM Luis Gomez <ecelgp@...> wrote:
This should fix:

https://git.opendaylight.org/gerrit/#/c/76839/

 
I am also running this job in sandbox to verify:

https://jenkins.opendaylight.org/sandbox/job/integration-multipatch-test-neon/1

this failed with a new error, is this just because a ran on sandbox or a real new problem:
[ERROR] Failed to execute goal org.apache.maven.plugins:maven-antrun-plugin:1.8:run (default) on project karaf: An Ant BuildException has occured: Directory does not exist: /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/assembly/system/mysql
[ERROR] around Ant part ...<delete includeemptydirs="true">... @ 4:35 in /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/antrun/build-main.xml
[ERROR] -> [Help 1]
org.apache.maven.lifecycle.LifecycleExecutionException: Failed to execute goal org.apache.maven.plugins:maven-antrun-plugin:1.8:run (default) on project karaf: An Ant BuildException has occured: Directory does not exist: /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/assembly/system/mysql
around Ant part ...<delete includeemptydirs="true">... @ 4:35 in /w/workspace/integration-multipatch-test-neon/patch_tester/distribution/karaf/target/antrun/build-main.xml
This mysql business sounds vaguely familiar - is this the thing we did to yank out (remove) the mysql JDBC client because of license concerns? MAYBE that is no longer there now (check please), because the newer version of Karaf in the new odlparent has picked up a new PAX JDBC where "someone" (me!) had done https://ops4j1.jira.com/browse/PAXJDBC-128 to remove that at source upstream? If that has trickled down and finally reached us now, we can probably remove the Ant <delete> hack we had to do at the time. Will someone have a closer look at this and confirm?
 
BR/Luis


> On Oct 10, 2018, at 3:53 PM, Robert Varga <nite@...> wrote:
>
> On 11/10/2018 00:49, Luis Gomez wrote:
>> It is also weird this miss was not detected before...
>
> I believe we weren't upgrading karaf using this process -- both Fluorine
> and Oxygen are shipping odlparent-3.1.x.
>
> Regards,
> Robert