CI: replace JDK 15 with 16 #16195

famod · 2021-04-01T22:05:42Z

quarkus-bot · 2021-04-01T22:05:44Z

famod · 2021-04-01T22:06:51Z

I'm currently testing some fixes/workarounds for the known problems:
https://github.com/famod/quarkus/actions/workflows/jdk-early-access-build.yml?query=branch%3Ajdk16-fixes

famod · 2021-04-02T14:36:05Z

After skipping devtools/gradle and integration-tests/gradle and adding an --add-opens workaround for the kotlin-maven-plugin, the build failed in my fork with 100 test failures. I'll try to group the failures...

famod · 2021-04-02T14:58:41Z

Affected test classes with number of failing test methods:

grep -P '<<< (ERROR|FAILURE)' quarkus-jdk16-failures.log | grep -Po '[^ ]+Test(?=\.)' | sort | uniq -c
      2 io.quarkus.cli.CliTest
     11 io.quarkus.elytron.security.ldap.CustomRoleDecoderTest
     11 io.quarkus.elytron.security.ldap.MinimalConfigurationTest
      1 io.quarkus.elytron.security.ldap.SearchRecursiveTest
      1 io.quarkus.it.kubernetes.BasicKubernetesTest
      1 io.quarkus.it.kubernetes.BasicOpenshiftTest
      1 io.quarkus.it.kubernetes.KnativeClusterLocalTest
      1 io.quarkus.it.kubernetes.KnativeContainerImageTest
      1 io.quarkus.it.kubernetes.KnativeScaleBoundsTest
      1 io.quarkus.it.kubernetes.KnativeScaleToZeroTest
      1 io.quarkus.it.kubernetes.KnativeTest
      1 io.quarkus.it.kubernetes.KnativeWithApplicationPropertiesTest
      1 io.quarkus.it.kubernetes.KnativeWithHealthTest
      1 io.quarkus.it.kubernetes.KnativeWithSecretConfigTest
      1 io.quarkus.it.kubernetes.KnativeWithVolumesTest
      1 io.quarkus.it.kubernetes.KubernetesAndMinikubeWithApplicationPropertiesTest
      1 io.quarkus.it.kubernetes.KubernetesConfigWithSecretsTest
      1 io.quarkus.it.kubernetes.KubernetesInitContainersTest
      1 io.quarkus.it.kubernetes.KubernetesServiceMappingTest
      1 io.quarkus.it.kubernetes.KubernetesWithApplicationPropertiesTest
      1 io.quarkus.it.kubernetes.KubernetesWithConfigMapCustomModeTest
      1 io.quarkus.it.kubernetes.KubernetesWithConfigMapTest
      1 io.quarkus.it.kubernetes.KubernetesWithDefaultsTest
      1 io.quarkus.it.kubernetes.KubernetesWithDeprecatedPropertiesTest
      1 io.quarkus.it.kubernetes.KubernetesWithEnvFromConfigMapTest
      1 io.quarkus.it.kubernetes.KubernetesWithEnvFromFieldTest
      1 io.quarkus.it.kubernetes.KubernetesWithEnvFromSecretTest
      1 io.quarkus.it.kubernetes.KubernetesWithGitlabLikeImageTest
      1 io.quarkus.it.kubernetes.KubernetesWithHealthAndJibTest
      1 io.quarkus.it.kubernetes.KubernetesWithHealthTest
      1 io.quarkus.it.kubernetes.KubernetesWithHostAliasesTest
      1 io.quarkus.it.kubernetes.KubernetesWithMetricsCustomAbsoluteTest
      1 io.quarkus.it.kubernetes.KubernetesWithMetricsCustomRelativeTest
      1 io.quarkus.it.kubernetes.KubernetesWithMetricsNoAnnotationsTest
      1 io.quarkus.it.kubernetes.KubernetesWithMetricsTest
      1 io.quarkus.it.kubernetes.KubernetesWithMicrometerTest
      1 io.quarkus.it.kubernetes.KubernetesWithMixedStyleEnvTest
      1 io.quarkus.it.kubernetes.KubernetesWithNewStyleEnvTest
      1 io.quarkus.it.kubernetes.KubernetesWithOldStyleEnvTest
      1 io.quarkus.it.kubernetes.KubernetesWithQuarkusAppNameTest
      1 io.quarkus.it.kubernetes.KubernetesWithRbacAndNamespaceTest
      1 io.quarkus.it.kubernetes.KubernetesWithResourcesRequirementsTest
      1 io.quarkus.it.kubernetes.KubernetesWithRootAndHealthTest
      1 io.quarkus.it.kubernetes.KubernetesWithSidecarAndJibTest
      1 io.quarkus.it.kubernetes.KubernetesWithSidecarAndProbesTest
      1 io.quarkus.it.kubernetes.KubernetesWithSysGroupTest
      1 io.quarkus.it.kubernetes.KubernetesWithWarningsEnvTest
      1 io.quarkus.it.kubernetes.MinikubeWithApplicationPropertiesTest
      1 io.quarkus.it.kubernetes.MinikubeWithDefaultsTest
      1 io.quarkus.it.kubernetes.MinikubeWithMixedStyleEnvTest
      1 io.quarkus.it.kubernetes.OpenshiftV3Test
      1 io.quarkus.it.kubernetes.OpenshiftV4Test
      1 io.quarkus.it.kubernetes.OpenshiftWithAppConfigMapTest
      1 io.quarkus.it.kubernetes.OpenshiftWithAppSecretTest
      1 io.quarkus.it.kubernetes.OpenshiftWithApplicationPropertiesTest
      1 io.quarkus.it.kubernetes.OpenshiftWithHealthTest
      1 io.quarkus.it.kubernetes.OpenshiftWithLegacySidecarAndS2iTest
      1 io.quarkus.it.kubernetes.OpenshiftWithRoutePropertiesTest
      1 io.quarkus.it.kubernetes.OpenshiftWithS2iTest
      1 io.quarkus.it.kubernetes.OpenshiftWithSidecarAndS2iTest
      1 io.quarkus.it.kubernetes.OpenshiftWithUberJarTest
      1 io.quarkus.it.kubernetes.WithKubernetesClientTest
      2 io.quarkus.it.main.MethodSourceTest
      2 io.quarkus.it.main.ParameterResolverTest
      1 org.eclipse.microprofile.fault.tolerance.tck.fallbackmethod.FallbackMethodDefaultMethodTest
     14 org.eclipse.microprofile.rest.client.tck.ClientHeaderParamTest
      1 org.eclipse.microprofile.rest.client.tck.InvalidInterfaceTest
      1 org.eclipse.microprofile.rest.client.tck.timeout.TimeoutBuilderIndependentOfMPConfigTest
      1 org.eclipse.microprofile.rest.client.tck.timeout.TimeoutTest
      2 org.eclipse.microprofile.rest.client.tck.timeout.TimeoutViaMPConfigTest
      2 org.eclipse.microprofile.rest.client.tck.timeout.TimeoutViaMPConfigWithConfigKeyTest

updated on 02.05.2021

famod · 2021-05-02T19:39:40Z

All those io.quarkus.it.kubernetes.* failures are (AFAICT) caused by this change:
openjdk/jdk@6bb7e45#diff-64e57a650df91c988910dd64b53c82b0521869c5f4afd1024f03c6a97f3a8344
Surefire is still setting java.io.tmpdir for the forked JVM, but "too late" to be picked up by that JDK StaticProperty class.
This results in a mismatch of expected and actual working directories for those tests (that expect a kubernetes directory in a specific path), because System.getProperty("java.io.tmpdir") does yield the expected result (the target dir) but Files.createTempDirectory(...) picks up the default system temp dir. 🤦

I think I'll create an issue for Surefire. But we probably can't wait for a release with a fix (if any such fix is even possible in the first place).
Workaround ideas:

pass System.getProperty("java.io.tmpdir") to Files.createTempDirectory(...)
major downside: we'd need to catch all usages!
set java.io.tmpdir via surefire config (yet to be tested, but should be set early enough...hopefully)

FYI @Tibor17. What's interesting is that Surefire is also setting user.dir (at least it's in this surefire_<many numbers>tmp file that is passed to the forked JVM) and user.dir has the same problem...but already since JDK 11!
openjdk/jdk@4098f25#diff-bce1cc542611fe2285622bbb95442ebc5ba4836fa8ef12438b2738a11cb375b7R66

famod · 2021-05-02T21:09:27Z

Two tests in integration-tests/main (MethodSourceTest & ParameterResolverTest) fail due to:

com.thoughtworks.xstream.converters.ConversionException:
No converter available
---- Debugging information ----
message             : No converter available
type                : java.util.Arrays$ArrayList
converter           : com.thoughtworks.xstream.converters.reflection.ReflectionConverter
message[1]          : Unable to make field protected transient int java.util.AbstractList.modCount accessible: module java.base does not "opens java.util" to unnamed module @2a79781f
-------------------------------
        at com.thoughtworks.xstream.core.DefaultConverterLookup.lookupConverterForType(DefaultConverterLookup.java:88)

Will have to set --add-opens here as well.

famod · 2021-05-02T21:27:03Z

Various tests fail in extensions/elytron-security-ldap/deployment with:

Caused by: java.lang.IllegalAccessException: class io.quarkus.elytron.security.ldap.QuarkusDirContextFactory$QuarkusInitialContextFactoryBuilder cannot access class com.sun.jndi.ldap.LdapCtxFactory (in module java.naming) because module java.naming does not export com.sun.jndi.ldap to unnamed module @2894f984
        at java.base/jdk.internal.reflect.Reflection.newIllegalAccessException(Reflection.java:385)
        at java.base/java.lang.reflect.AccessibleObject.checkAccess(AccessibleObject.java:687)
        at java.base/java.lang.reflect.Constructor.newInstanceWithCaller(Constructor.java:489)
        at java.base/java.lang.reflect.ReflectAccess.newInstance(ReflectAccess.java:128)
        at java.base/jdk.internal.reflect.ReflectionFactory.newInstance(ReflectionFactory.java:350)
        at java.base/java.lang.Class.newInstance(Class.java:642)
        at io.quarkus.elytron.security.ldap.QuarkusDirContextFactory$QuarkusInitialContextFactoryBuilder.createInitialContextFactory(QuarkusDirContextFactory.java:166)
        ... 23 more

Will have to set --add-opens here as well.

gastaldi · 2021-05-02T21:35:22Z

set java.io.tmpdir via surefire config (yet to be tested, but should be set early enough...hopefully)

We already do that in https://github.com/jboss/jboss-parent-pom/blob/main/pom.xml#L587.

Maybe that's not being used correctly?

famod · 2021-05-02T21:36:58Z

@gastaldi

Maybe that's not being used correctly?

Thanks for checking, but turns out it has to be set via -D in <argLine>... (for JDK16+).

famod · 2021-05-02T21:37:53Z

Ah! So it's not actually Surefire setting it on its own! That's good to know. So, sorry for bothering you @Tibor17!

gastaldi · 2021-05-02T21:46:53Z

Thanks for checking, but turns out it has to be set via -D in <argLine>... (for JDK16+).

Is that because of JDK 16+ only? I remember this setting working in other projects just fine.

famod · 2021-05-02T21:48:13Z

Yes, JDK16+ only as far I could trace it back: openjdk/jdk@6bb7e45#diff-64e57a650df91c988910dd64b53c82b0521869c5f4afd1024f03c6a97f3a8344

famod · 2021-05-02T22:21:34Z

I've updated the TODO list to reflect the current state of affairs. Two builds are running currently in my fork, will check them in the next few days.
This is what I have added so far: main...famod:jdk16-fixes (I'll remove [skip ci] once I think I'm done).

Tibor17 · 2021-05-02T22:34:20Z

@famod
We are not overriding java.io.tmpdir in the forked surefire JVM. I checked in the code. And we do not need to do it, no reason.
The CWD should be set to the path where the POM resides.

Tibor17 · 2021-05-02T22:52:00Z

@famod
We are overriding user.dir at the line setSystemProperties( new File( tmpDir, effectiveSystemPropertiesFileName ) );. It is an old code and I do not understand why we do it like this even if the CWD is set to the process cli.setWorkingDirectory( getWorkingDirectory( forkNumber ).getAbsolutePath() );. I think this can be checked out and the line result.setProperty( "user.dir", getWorkingDirectory().getAbsolutePath() ); can be removed in the AbstractSurefireMojo.

famod · 2021-05-03T20:47:42Z

@Tibor17 WDYT about extending the systemPropertyVariables logic in surefire to detect "special" system properties like java.io.tmpdir and automatically set those via -D instead of setSystemProperties(...)?

Btw @gastaldi I'm now getting:

[WARNING] The system property java.io.tmpdir is configured twice! The property appears in <argLine/> and any of <systemPropertyVariables/>, <systemProperties/> or user property.

So I'll have to override the property from jboss-parent completely.

PS: Meh...I guess IntelliJ IDEA will then also use the default temp dir for testing.

famod · 2021-05-03T21:20:15Z

@Ladicek @jmartisk @michalszynkiewicz
RestClientTestCase.testFaultTolerance (in integration-tests/main) fails on JDK 16 with:

Caused by: java.lang.reflect.InaccessibleObjectException: Unable to make java.lang.invoke.MethodHandles$Lookup(java.lang.Class) accessible: module java.base does not "opens java.lang.invoke" to unnamed module @486ecb69
        at java.base/java.lang.reflect.AccessibleObject.checkCanSetAccessible(AccessibleObject.java:357)
        at java.base/java.lang.reflect.AccessibleObject.checkCanSetAccessible(AccessibleObject.java:297)
        at java.base/java.lang.reflect.Constructor.checkCanSetAccessible(Constructor.java:188)
        at java.base/java.lang.reflect.Constructor.setAccessible(Constructor.java:181)
        at io.smallrye.faulttolerance.DefaultMethodFallbackProvider.getFallback(DefaultMethodFallbackProvider.java:19)
        at io.smallrye.faulttolerance.FaultToleranceInterceptor.lambda$prepareFallbackFunction$7(FaultToleranceInterceptor.java:481)
        at io.smallrye.faulttolerance.core.fallback.Fallback.doApply(Fallback.java:59)
        at io.smallrye.faulttolerance.core.fallback.Fallback.apply(Fallback.java:32)
        at io.smallrye.faulttolerance.core.metrics.MetricsCollector.doApply(MetricsCollector.java:80)
        at io.smallrye.faulttolerance.core.metrics.MetricsCollector.apply(MetricsCollector.java:70)
        at io.smallrye.faulttolerance.FaultToleranceInterceptor.syncFlow(FaultToleranceInterceptor.java:174)
        at io.smallrye.faulttolerance.FaultToleranceInterceptor.interceptCommand(FaultToleranceInterceptor.java:151)
        at io.smallrye.faulttolerance.FaultToleranceInterceptor_Bean.intercept(FaultToleranceInterceptor_Bean.zig:517)
        at org.jboss.resteasy.microprofile.client.InvocationContextImpl$InterceptorInvocation.invoke(InvocationContextImpl.java:154)
        at org.jboss.resteasy.microprofile.client.InvocationContextImpl.invokeNext(InvocationContextImpl.java:53)
        at org.jboss.resteasy.microprofile.client.InvocationContextImpl.proceed(InvocationContextImpl.java:89)

Is this a known problem? Is there a fix?

Tibor17 · 2021-05-04T08:21:15Z

@famod
Regarding this warning

[WARNING] The system property java.io.tmpdir is configured twice! The property appears in and any of , or user property.

it is a parameterized message and java.io.tmpdir in the blacklist. Not sure if the property is really a must there. After making a fast lookup in the code, I would say it is not but of course I have to double check it, remove it and run all ITs.

Regarding systemPropertyVariables I have no answer yet. Pls let me think about it a bit.

Ladicek · 2021-05-04T09:03:22Z

@famod It is a known problem now that you pointed it to me :-) I'll take a look, thanks.

famod · 2021-05-04T10:03:10Z

Current state (rest is ok):

[INFO] Quarkus - Integration Tests - Devtools ............. FAILURE [04:34 min]
[INFO] Quarkus - Integration Tests - Kotlin ............... FAILURE [20:14 min]
[INFO] Quarkus - TCK - MicroProfile Fault Tolerance ....... FAILURE [11:34 min]
[INFO] Quarkus - TCK - MicroProfile REST Client ........... FAILURE [11:05 min]

Tibor17 · 2021-05-04T20:05:15Z

@famod
I have removed user.dir. Let's see the build result.
https://github.com/apache/maven-surefire/tree/userDir
https://ci-builds.apache.org/job/Maven/job/maven-box/job/maven-surefire/job/userDir/
The subprocess CWD and the property should have the same value. So, removing the setter for the property should not change anything but it's safe for new versions of Java.

Tibor17 · 2021-05-04T20:13:12Z

Regarding java.io.tmpdir we do not override it. This warning means that your project set this twice. Once in the config <argLine> and some system property.

[WARNING] The system property java.io.tmpdir is configured twice! The property appears in and any of , or user property.

Quoting from Surefire code, the system properties are the Maven system properties and config parameters systemProperties, systemPropertyVariables and User Properties. Check also settings.xml, folder for .mvn or the environment JAVA_OPTS. Pls check everything!

famod · 2021-05-04T21:09:26Z

@Tibor17

Regarding java.io.tmpdir we do not override it. This warning means that your project set this twice. Once in the config <argLine> and some system property.

It's clear where it's coming from: jboss-parent is setting it via systemPropertyVariables and since it is set too late for JDK 16, I had to add it via argLine here in Quarkus (not yet merged, still only in my fork).
Until it is changed in jboss-parent (in whichever way), I'm using this workaround to avoid inheritance:

<systemPropertyVariables combine.self="override">

Btw, I think I'll create a SUREFIRE issue in the next few days for the suggestion to handle certain systemPropertyVariables in a special way.

famod · 2021-05-04T21:11:17Z

@Ladicek

@famod It is a known problem now that you pointed it to me :-) I'll take a look, thanks.

Thanks! Btw, the same happens in org.eclipse.microprofile.fault.tolerance.tck.fallbackmethod.FallbackMethodDefaultMethodTest.fallbackMethodDefaultMethod (tcks/microprofile-fault-tolerance).

Tibor17 · 2021-05-04T21:26:59Z

@famod
http://maven.apache.org/surefire/maven-surefire-plugin/test-mojo.html#systemProperties should be set eagerly.

famod · 2021-05-04T21:30:53Z

@Tibor17 Interesting! But that's kind of funny because those are deprecated and not long ago I changed it in jboss-parent to systemPropertyVariables (because of the deprecation).

Tibor17 · 2021-05-04T21:38:05Z

@famod
I checked it again. It is not set eagerly. We can make it eager or a new parameter similar to ...Values.

Ladicek · 2021-05-05T07:22:36Z

@famod It is a known problem now that you pointed it to me :-) I'll take a look, thanks.

Thanks! Btw, the same happens in org.eclipse.microprofile.fault.tolerance.tck.fallbackmethod.FallbackMethodDefaultMethodTest.fallbackMethodDefaultMethod (tcks/microprofile-fault-tolerance).

Funny thing is that once I updated Weld in SmallRye Fault Tolerance (to the latest release that supports JDK 16), all tests pass, including the entire TCK. Which suggests there may be something wrong in Quarkus. Hmm.

Ladicek · 2021-05-05T10:32:05Z

Ah, I know what's wrong. smallrye-fault-tolerance-5.0.0.jar is supposed to be a MR JAR, but the java9 variant of DefaultMethodFallbackProvider is lost somewhere during the build process!

EDIT: the META-INF/versions directory was lost in the 4.3.1 release, so this commit must be guilty: smallrye/smallrye-fault-tolerance@4b5081b

Ladicek · 2021-05-05T11:46:13Z

Should be fixed in next SmallRye Fault Tolerance release that will contain smallrye/smallrye-fault-tolerance#409

famod · 2021-05-05T12:45:35Z

@Ladicek Great, thanks! Will exclude the affected tests until then.

Tibor17 · 2021-05-06T08:14:17Z

@famod
The CI build passed successfully. It removed the code overriding user.dir in forked surefire JVM. Pls create a Jira ticket for this, thx.

famod added the area/housekeeping Issue type for generalized tasks not related to bugs or enhancements label Apr 1, 2021

quarkus-bot bot added the area/kotlin label Apr 1, 2021

famod removed the area/kotlin label Apr 1, 2021

famod self-assigned this Apr 1, 2021

MartinX3 mentioned this issue Apr 26, 2021

Kotlin 1.5.10 #16806

Closed

famod mentioned this issue Apr 28, 2021

Error during build: Unrecognized option: --add-opens #16862

Closed

asfgit pushed a commit to apache/maven-surefire that referenced this issue May 4, 2021

https://github.com/quarkusio/quarkus/issues/16195

7df99f3

asfgit pushed a commit to apache/maven-surefire that referenced this issue May 5, 2021

https://github.com/quarkusio/quarkus/issues/16195

4d17174

famod mentioned this issue May 6, 2021

JDK16 fixes and workarounds that should make the EA job pass #17053

Merged

famod linked a pull request May 7, 2021 that will close this issue

JDK16 fixes and workarounds that should make the EA job pass #17053

Merged

This was referenced May 7, 2021

Enable fault-tolerance tests on JDK 16+ #17080

Closed

Enable Gradle modules and tests on JDK 16+ #17081

Closed

CI: Bump 'JVM Tests - JDK 15' to JDK 16 and EA job to 17-ea #17082

Merged

gsmet closed this as completed in #17082 May 8, 2021

quarkus-bot bot added this to the 2.0 - main milestone May 8, 2021

This was referenced May 8, 2021

Enable elytron-security-ldap tests on JDK 16+ #17092

Closed

Enable xstream based tests on JDK 16+ #17091

Closed

Enable TCKS rest-client tests on JDK 16+ #17093

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

CI: replace JDK 15 with 16 #16195

CI: replace JDK 15 with 16 #16195

famod commented Apr 1, 2021 •

edited

quarkus-bot bot commented Apr 1, 2021

famod commented Apr 1, 2021 •

edited

famod commented Apr 2, 2021 •

edited

famod commented Apr 2, 2021 •

edited

famod commented May 2, 2021 •

edited

famod commented May 2, 2021

famod commented May 2, 2021 •

edited

gastaldi commented May 2, 2021

famod commented May 2, 2021 •

edited

famod commented May 2, 2021 •

edited

gastaldi commented May 2, 2021

famod commented May 2, 2021

famod commented May 2, 2021

Tibor17 commented May 2, 2021 •

edited

Tibor17 commented May 2, 2021 •

edited

famod commented May 3, 2021 •

edited

famod commented May 3, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

Ladicek commented May 4, 2021

famod commented May 4, 2021

Tibor17 commented May 4, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

famod commented May 4, 2021 •

edited

famod commented May 4, 2021

Tibor17 commented May 4, 2021

famod commented May 4, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

Ladicek commented May 5, 2021

Ladicek commented May 5, 2021 •

edited

Ladicek commented May 5, 2021

famod commented May 5, 2021

Tibor17 commented May 6, 2021 •

edited

CI: replace JDK 15 with 16 #16195

CI: replace JDK 15 with 16 #16195

Comments

famod commented Apr 1, 2021 • edited

Description

Currently known TODOs (potentially incomplete)

quarkus-bot bot commented Apr 1, 2021

famod commented Apr 1, 2021 • edited

famod commented Apr 2, 2021 • edited

famod commented Apr 2, 2021 • edited

famod commented May 2, 2021 • edited

famod commented May 2, 2021

famod commented May 2, 2021 • edited

gastaldi commented May 2, 2021

famod commented May 2, 2021 • edited

famod commented May 2, 2021 • edited

gastaldi commented May 2, 2021

famod commented May 2, 2021

famod commented May 2, 2021

Tibor17 commented May 2, 2021 • edited

Tibor17 commented May 2, 2021 • edited

famod commented May 3, 2021 • edited

famod commented May 3, 2021 • edited

Tibor17 commented May 4, 2021 • edited

Ladicek commented May 4, 2021

famod commented May 4, 2021

Tibor17 commented May 4, 2021 • edited

Tibor17 commented May 4, 2021 • edited

famod commented May 4, 2021 • edited

famod commented May 4, 2021

Tibor17 commented May 4, 2021

famod commented May 4, 2021 • edited

Tibor17 commented May 4, 2021 • edited

Ladicek commented May 5, 2021

Ladicek commented May 5, 2021 • edited

Ladicek commented May 5, 2021

famod commented May 5, 2021

Tibor17 commented May 6, 2021 • edited

famod commented Apr 1, 2021 •

edited

famod commented Apr 1, 2021 •

edited

famod commented Apr 2, 2021 •

edited

famod commented Apr 2, 2021 •

edited

famod commented May 2, 2021 •

edited

famod commented May 2, 2021 •

edited

famod commented May 2, 2021 •

edited

famod commented May 2, 2021 •

edited

Tibor17 commented May 2, 2021 •

edited

Tibor17 commented May 2, 2021 •

edited

famod commented May 3, 2021 •

edited

famod commented May 3, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

famod commented May 4, 2021 •

edited

famod commented May 4, 2021 •

edited

Tibor17 commented May 4, 2021 •

edited

Ladicek commented May 5, 2021 •

edited

Tibor17 commented May 6, 2021 •

edited