allow re-use of existing pods for restarts #244

scoheb · 2017-11-03T14:40:09Z

[FIXES JENKINS-47476]

iocanel

Aside from the one remark about final field converted to non-final everything else looks great to me.

I'd like to better understand the reason behind that change, but still I am ok with merging.

iocanel · 2017-11-07T13:55:02Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/ContainerExecDecorator.java


    @SuppressFBWarnings(value = "SE_TRANSIENT_FIELD_NOT_RESTORED", justification = "not needed on deserialization")
-    private final transient List<Closeable> closables = new ArrayList<>();


When the pipeline is resumed, since it is transient, it will not be deserialized...therefore when we come to add to array, it will NPE...therefore, we check before we add to it and create if necessary. Therefore it cannot be final.

Makes sense to me.

carlossg

There's an error with a dependency, so tests haven't run yet

carlossg · 2017-11-07T16:55:41Z

pom.xml

@@ -43,7 +43,7 @@
    <!-- jenkins plugins versions -->
    <jenkins-basic-steps.version>2.3</jenkins-basic-steps.version>
    <jenkins-credentials.version>2.1.7</jenkins-credentials.version>
-    <jenkins-durable-task.version>1.13</jenkins-durable-task.version>
+    <jenkins-durable-task.version>1.16</jenkins-durable-task.version>


1.16 does not exist

Yes. I am waiting for the new release that will include jenkinsci/durable-task-plugin#49

Use 1.16-20171108.031054-1 in the meantime. Based on jenkinsci/durable-task-plugin@9b0b723.

Thanks @jglick

carlossg · 2017-11-07T16:58:53Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/KubernetesLauncher.java

-            pod = client.pods().inNamespace(namespace).create(pod);
-            LOGGER.log(Level.INFO, "Created Pod: {0} in namespace {1}", new Object[]{podId, namespace});
-            listener.getLogger().printf("Created Pod: %s in namespace %s", podId, namespace);
+            Pod checkPod = client.pods().inNamespace(namespace).withName(podId).get();


IIUC when the master restarts, agents reconnect and this method would not be called and the agents already reconnect.
There may be multiple pods with same podId running different builds

A simple test (https://paste.fedoraproject.org/paste/rwMXdUC670i5PIZFAokjlQ) using the current master gives this in the logs when a restart happens:

https://paste.fedoraproject.org/paste/BRTBZEMvyPxsOo2BJJf6uw

With or without idleMinutes

interesting, this was changed in abb2b63, can you take a look @Vlatombe

I guess a simple check of launched value at the beginning of the method would be enough. If the pod has been deleted in the mean time we may risk launching another one, leading to an inconsistent state anyway.

@Vlatombe so you are ok with the approach in this PR?

@Vlatombe but how is launched being persisted across restarts? on a restart, the attempt to reconnect fails with a handshake error:

https://paste.fedoraproject.org/paste/yHv3FoJ35r7vG8J3hWwYEQ

@scoheb Should be in the slave's config.xml > launcher > launched.

Dunno about the handshake for now since the real error is likely in the master's logs.

I checked the slave config.xml once launched and while running a build:

https://paste.fedoraproject.org/paste/cp~cnrkWdoDkoZS~OLkWLg

Looks like the launched property (when true) is not saved. Do we need a slave.save() after at the end of the launch() ?

Sounds like the right move.

@Vlatombe

With master and ensuring 'launched == true' in config.xml (via direct save() in launch()), here is Jenkins log after restart:
https://paste.fedoraproject.org/paste/UhqADZdp010vXMAk2HI8nw

No trace of why the slave is not connecting properly...

carlossg · 2017-11-07T17:00:25Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/PodTemplateStepExecution.java

@@ -40,7 +46,9 @@
    @Override
    public boolean start() throws Exception {

-        Cloud cloud = Jenkins.getInstance().getCloud(step.getCloud());
+        LOGGER.info("starting");


can you make this more descriptive and probably just FINE level ?

I will remove this.

carlossg · 2017-11-07T18:17:30Z

...ces/org/csanchez/jenkins/plugins/kubernetes/pipeline/runInPodWithRestartWithLongSleep.groovy

+                    semaphore 'wait'
+                    sh '''
+                    #!/bin/bash
+                    for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20


sh 'for i in `seq 1 20`; do echo $i; sleep 5; done'

I was having issues with 'seq' at one time...I thought it was due to the busybox image. I will update my sh call.

jglick · 2017-11-08T03:12:51Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/ContainerLogStepExecution.java

+            KubernetesNodeContext nodeContext = new KubernetesNodeContext(getContext());
+            client = nodeContext.connectToCloud();
+        } catch (Exception e) {
+            ContainerLogStepExecution.this.getContext().onFailure(e);


Include a TODO comment mentioning JENKINS-40161.

jglick · 2017-11-08T03:13:07Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/ContainerStepExecution.java

+            client = nodeContext.connectToCloud();
+            decorator.setKubernetesClient(client);
+        } catch (Exception e) {
+            ContainerStepExecution.this.getContext().onFailure(e);


jglick · 2017-11-08T03:15:51Z

src/test/java/org/jvnet/hudson/test/RestartableJenkinsNonLocalhostRule.java

+
+        // run each step inside its own JenkinsRule
+        for (Statement step : steps) {
+            j = new JenkinsRuleNonLocalhost().with(loader);


You could probably use jenkinsci/jenkins-test-harness#75 and avoid a lot of code duplication, no?

jglick · 2017-11-08T03:16:27Z

src/test/java/org/jvnet/hudson/test/JenkinsRuleNonLocalhost.java

@@ -42,8 +40,6 @@
 import org.eclipse.jetty.webapp.Configuration;
 import org.eclipse.jetty.webapp.WebAppContext;
 import org.eclipse.jetty.webapp.WebXmlConfiguration;
-import org.jvnet.hudson.test.JenkinsRule;
-import org.jvnet.hudson.test.ThreadPoolImpl;


Why these changes?

IntelliJ showed them as unused.

scoheb · 2017-11-08T21:32:00Z

@Vlatombe Since we were not checking launched, the provision would fail (since pod already existed) and it would fall into catch block where we would removeNode(), therefore causing the subsequent handshake error on the agent side. This become more evident when I switched to using JNLP-4 and it clearly said it was rejecting the client since the node name was not found.
Thanks for your help!

scoheb · 2017-11-09T00:36:12Z

@carlossg Would you be able to provide some of the test logs from https://jenkins.gke.csanchez.org/job/jenkinsci/job/kubernetes-plugin/job/PR-244/10/ ?

I can run the RestartPipelineTest locally using minikube and -Djenkins.host.address=192.168.99.1 but not in the test environment. Surprisingly enough, KubernetesPipelineTest runs fine above, so I must not be setting up 'RestartPipelineTest` correctly.

carlossg · 2017-11-13T11:41:31Z

...ces/org/csanchez/jenkins/plugins/kubernetes/pipeline/runInPodWithRestartWithLongSleep.groovy

+package org.csanchez.jenkins.plugins.kubernetes.pipeline
+
+podTemplate(label: 'mypod', containers: [
+        containerTemplate(name: 'busybox', image: 'startx/fedora', ttyEnabled: true, command: '/bin/cat'),


why not busybox image ?

carlossg · 2017-11-13T11:41:48Z

...hez/jenkins/plugins/kubernetes/pipeline/runInPodWithRestartWithMultipleContainerCalls.groovy

+package org.csanchez.jenkins.plugins.kubernetes.pipeline
+
+podTemplate(label: 'mypod', namespace: 'default', idleMinutes: 0, containers: [
+        containerTemplate(name: 'busybox', image: 'startx/fedora', ttyEnabled: true, command: '/bin/cat'),


why not busybox image ?

carlossg · 2017-11-13T12:05:11Z

for reference, the stack when agents try to reconnect is

WARNING: Error in provisioning; slave=KubernetesSlave name: jenkins-slave-v1sqv-fzmk0, template=org.csanchez.jenkins.plugins.kubernetes.PodTemplate@562287df
io.fabric8.kubernetes.client.KubernetesClientException: Failure executing: POST at: https://kubernetes.default.svc/api/v1/namespaces/default/pods. Message: pods "jenkins-slave-v1sqv-fzmk0" already exists. Received status: Status(apiVersion=v1, code=409, details=StatusDetails(causes=[], group=null, kind=pods, name=jenkins-slave-v1sqv-fzmk0, retryAfterSeconds=null, additionalProperties={}), kind=Status, message=pods "jenkins-slave-v1sqv-fzmk0" already exists, metadata=ListMeta(resourceVersion=null, selfLink=null, additionalProperties={}), reason=AlreadyExists, status=Failure, additionalProperties={}).
	at io.fabric8.kubernetes.client.dsl.base.OperationSupport.requestFailure(OperationSupport.java:470)
	at io.fabric8.kubernetes.client.dsl.base.OperationSupport.assertResponseCode(OperationSupport.java:409)
	at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse(OperationSupport.java:379)
	at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleResponse(OperationSupport.java:343)
	at io.fabric8.kubernetes.client.dsl.base.OperationSupport.handleCreate(OperationSupport.java:226)
	at io.fabric8.kubernetes.client.dsl.base.BaseOperation.handleCreate(BaseOperation.java:769)
	at io.fabric8.kubernetes.client.dsl.base.BaseOperation.create(BaseOperation.java:356)
	at org.csanchez.jenkins.plugins.kubernetes.KubernetesLauncher.launch(KubernetesLauncher.java:133)
	at hudson.slaves.SlaveComputer$1.call(SlaveComputer.java:285)
	at jenkins.util.ContextResettingExecutorService$2.call(ContextResettingExecutorService.java:46)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)

scoheb · 2017-11-13T20:32:46Z

@carlossg updated podTemplates in *.groovy to use busybox. At one time, I thought the ps output from the busybox images was causing issues...

If the tests fail again, can you help me and provide any logs from k8s side?
Or maybe update the Jenkinsfile that is used to ALWAYS capture the junit results and logs...it is not the case now.

carlossg · 2017-11-15T13:27:40Z

now test logs are captured and removed the explicit namespace definition, that was the reason for failure, let's see

scoheb · 2017-11-15T14:17:15Z

@carlossg Thanks

Seems like tests are failing due to leftover pods...can you take a look please?

scoheb · 2017-11-15T17:41:16Z

@carlossg added to java files only. Do we need to add to .groovy files too?

jglick · 2017-11-16T14:43:19Z

BTW as a matter of courtesy to reviewers, it is nice to hyperlink to JIRA JENKINS-47476 in the PR description, and then also mark the JIRA In Review.

jglick · 2017-11-16T14:43:37Z

(assuming this PR purports to fix the defect)

scoheb · 2017-11-16T14:45:47Z

@jglick Thanks. I will properly link to the JIRA and mark it as In Review in the future.

jglick

Do not understand the K8S aspects, but there are some minor mistakes in generic Pipeline usage (not related to jenkinsci/durable-task-plugin#49).

jglick · 2017-11-16T14:46:01Z

src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/RestartPipelineTest.java

+            }
+        });
+        story.addStep(new Statement() {
+            @SuppressWarnings("SleepWhileInLoop")


No such sleep here AFAICT?

jglick · 2017-11-16T14:47:46Z

src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/RestartPipelineTest.java

+
+    @Test
+    public void getContainerLogWithRestart() throws Exception {
+        story.addStep(new Statement() {


BTW you are on java.level=8 (even though the supposed jenkins.version=2.32.1 runs on Java 7 and thus people actually using 2.32.x or 2.46.x on Java 7 and installing this would see Jenkins crash and burn if they installed this plugin), so you may as well use

story.then(r -> { configureCloud(); r.jenkins.addNode(…); // … });

jglick · 2017-11-16T14:48:19Z

pom.xml

@@ -3,7 +3,7 @@
  <parent>
    <groupId>org.jenkins-ci.plugins</groupId>
    <artifactId>plugin</artifactId>
-    <version>2.32</version>
+    <version>2.37</version>


jglick · 2017-11-16T14:49:27Z

pom.xml

@@ -46,12 +46,12 @@
    <jenkins.version>2.32.1</jenkins.version>

    <kubernetes-client.version>2.6.1</kubernetes-client.version>
-    <slf4j.version>1.7.13</slf4j.version>
+    <slf4j.version>1.7.25</slf4j.version>


Hmm, this override should not be here at all; cf. jenkinsci/plugin-pom#73.

jglick · 2017-11-16T14:50:24Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/KubernetesLauncher.java

@@ -120,6 +120,12 @@ public void launch(SlaveComputer computer, TaskListener listener) {
        if (slave == null) {
            throw new IllegalStateException("Node has been removed, cannot launch " + computer.getName());
        }
+        if (launched) {
+            LOGGER.info("Agent: " + slave.getNodeName() + " has already been launched. Activating...");


BTW it does not really matter for INFO and above, but it is good to get in the habit of using either message formats or lambdas when logging nonconstant messages.

yep, I'd also prefer

LOGGER.log(INFO, "Agent has already been launched, activating: {}", slave.getNodeName());

jglick · 2017-11-16T15:13:11Z

src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/RestartPipelineTest.java

+                SemaphoreStep.success("wait/1", false);
+                // we need to wait until we are sure that the sh
+                // step has started...
+                Thread.sleep(10000);


jglick · 2017-11-16T15:13:34Z

src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/RestartPipelineTest.java

+                        , true));
+                WorkflowRun b = p.scheduleBuild2(0).waitForStart();
+                SemaphoreStep.waitForStart("wait/1", b);
+                SemaphoreStep.success("wait/1", false);


jglick · 2017-11-16T15:13:38Z

src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/RestartPipelineTest.java

+                WorkflowRun b = p.scheduleBuild2(0).waitForStart();
+                SemaphoreStep.waitForStart("wait/1", b);
+                SemaphoreStep.success("wait/1", false);
+                Thread.sleep(1000);


jglick · 2017-11-16T15:14:23Z

src/test/java/org/jvnet/hudson/test/JenkinsRuleNonLocalhost.java

@@ -24,8 +24,6 @@

 package org.jvnet.hudson.test;


Eek! Cf. jenkinsci/jenkins-test-harness#84

jglick · 2017-11-16T15:15:03Z

src/test/java/org/jvnet/hudson/test/RestartableJenkinsNonLocalhostRule.java

+    }
+
+    @Override
+    protected JenkinsRule createJenkinsRule(Description description) {


Ah good to see jenkinsci/jenkins-test-harness#75 getting used at least.

scoheb · 2017-11-16T18:26:19Z

@carlossg I believe I have addressed all of Jesse's comments.

scoheb · 2017-11-17T14:11:48Z

Latest test run shows this in logs:

Failed to count the # of live instances on Kubernetes io.fabric8.kubernetes.client.KubernetesClientException: Failure executing: GET at: https://kubernetes.default.svc/api/v1/namespaces/default/pods?labelSelector=jenkins%3Dslave. Message: Forbidden!Configured service account does

and

org.jvnet.hudson.reactor.ReactorException: java.lang.Error: java.lang.reflect.InvocationTargetException at org.jvnet.hudson.reactor.Reactor.execute(Reactor.java:269) at jenkins.InitReactorRunner.run(InitReactorRunner.java:47) at jenkins.model.Jenkins.executeReactor(Jenkins.java:1110) at jenkins.model.Jenkins.<init>(Jenkins.java:926) at hudson.model.Hudson.<init>(Hudson.java:85) at org.jvnet.hudson.test.JenkinsRule.newHudson(JenkinsRule.java:611) at org.jvnet.hudson.test.JenkinsRule.before(JenkinsRule.java:384) at org.jvnet.hudson.test.JenkinsRule$1.evaluate(JenkinsRule.java:537) at org.junit.internal.runners.statements.FailOnTimeout$CallableStatement.call(FailOnTimeout.java:298) at org.junit.internal.runners.statements.FailOnTimeout$CallableStatement.call(FailOnTimeout.java:292) at java.util.concurrent.FutureTask.run(FutureTask.java:266) at java.lang.Thread.run(Thread.java:748) Caused by: java.lang.Error: java.lang.reflect.InvocationTargetException at hudson.init.TaskMethodFinder.invoke(TaskMethodFinder.java:110) at hudson.init.TaskMethodFinder$TaskImpl.run(TaskMethodFinder.java:175) at org.jvnet.hudson.reactor.Reactor.runTask(Reactor.java:282) at jenkins.model.Jenkins$7.runTask(Jenkins.java:1086) at org.jvnet.hudson.reactor.Reactor$2.run(Reactor.java:210) at org.jvnet.hudson.reactor.Reactor$Node.run(Reactor.java:117) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) ... 1 more Caused by: java.lang.reflect.InvocationTargetException at sun.reflect.GeneratedMethodAccessor323.invoke(Unknown Source) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at hudson.init.TaskMethodFinder.invoke(TaskMethodFinder.java:104) ... 8 more Caused by: java.lang.IndexOutOfBoundsException: Index: 0 at java.util.Collections$EmptyList.get(Collections.java:4454) at hudson.ExtensionList.get(ExtensionList.java:172) at org.jenkinsci.plugins.pipeline.modeldefinition.DescriptorLookupCache.getPublicCache(DescriptorLookupCache.java:53) at org.jenkinsci.plugins.pipeline.modeldefinition.DescriptorLookupCache.invalidateGlobalCache(DescriptorLookupCache.java:58)

scoheb · 2017-11-17T15:02:29Z

@carlossg Still getting this warning which I think is causing tests to fail...:

WARNING o.c.j.p.k.KubernetesCloud#provision: Failed to count the # of live instances on Kubernetes
io.fabric8.kubernetes.client.KubernetesClientException: Failure executing: GET at: https://kubernetes.default.svc/api/v1/namespaces/default/pods?labelSelector=jenkins%3Dslave. Message: Forbidden!Configured service account does....

scoheb · 2017-11-17T16:09:00Z

@carlossg Sorry. my bad. Some of the test groovy files had a namespace defined in them...I removed them and now tests are passing.

Vlatombe · 2017-11-21T13:28:59Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/ContainerLogStepExecution.java

    private transient KubernetesClient client;

+    @Override
+    // TODO Revisit for JENKINS-40161
+    public void onResume() {


I'm not convinced this is the proper implementation for this step. The parent class SynchronousNonBlockingStepExecution makes it clear that resuming such a step is unsupported, the bulk of the execution being run in a different thread, which is not resumed on restart.

As I understand it, the current onResume is just a no-op, so you may as well remove injection of 'client' as it won't be used in later calls.

@jglick you didn't raise this point before, do you agree with @Vlatombe ?

scoheb · 2017-11-21T14:05:47Z

I agree with you @Vlatombe ... I was just going for completeness (i.e. to have all pipeline steps resume) but as I look at it, the step is not a long running one like container()

I will amend my PR

- container step Includes: - update to parent pom to get latest jenkins-test-harness - update to durable-task plugin v1.16 [FIXES JENKINS-47476]

scoheb · 2017-11-21T14:29:41Z

@Vlatombe @carlossg reverted the change to ContainerLogStepExecution. Please take a look.

Vlatombe · 2017-11-21T14:30:14Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/ContainerLogStepExecution.java

@@ -16,6 +16,8 @@

 package org.csanchez.jenkins.plugins.kubernetes.pipeline;

+import com.google.inject.Inject;


Some unused imports to clean up here.

Vlatombe · 2017-11-21T14:30:28Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/ContainerStepExecution.java

@@ -4,30 +4,47 @@
 import java.util.logging.Level;
 import java.util.logging.Logger;

+import com.google.inject.Inject;


Vlatombe · 2017-11-21T14:30:37Z

src/main/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/PodTemplateStepExecution.java

@@ -5,12 +5,15 @@
 import java.util.logging.Level;
 import java.util.logging.Logger;

+import com.google.inject.Inject;


Vlatombe

lgtm. CI is failing due to rebase I believe.

carlossg · 2017-11-21T15:28:14Z

tests passing so 🎉

thanks for the hard work!

chancez · 2017-12-15T01:50:21Z

This doesn't seem to have fixed anything for me. Jenkins is still rejecting old agents because it's an unknown client name.

chancez · 2017-12-18T18:36:20Z

It turns out my plugin upgrade got reverted while I was testing by someone. This is working great!

scoheb force-pushed the restart-2 branch from d0b2e61 to e664bc8 Compare November 7, 2017 01:48

scoheb mentioned this pull request Nov 7, 2017

[JENKINS-47791] Remove ProcessLiveness and PID tracking in favor of a simple timestamp check on the log file jenkinsci/durable-task-plugin#49

Merged

23 tasks

scoheb force-pushed the restart-2 branch from e664bc8 to 0dd2998 Compare November 7, 2017 01:55

iocanel approved these changes Nov 7, 2017

View reviewed changes

scoheb force-pushed the restart-2 branch from 0dd2998 to e2b660d Compare November 7, 2017 14:19

carlossg requested changes Nov 7, 2017

View reviewed changes

carlossg reviewed Nov 7, 2017

View reviewed changes

scoheb force-pushed the restart-2 branch from e2b660d to e60e8d7 Compare November 7, 2017 18:44

jglick reviewed Nov 8, 2017

View reviewed changes

scoheb force-pushed the restart-2 branch 4 times, most recently from 15f629a to dd33690 Compare November 8, 2017 21:24

scoheb force-pushed the restart-2 branch from dd33690 to 2a573a6 Compare November 8, 2017 22:20

scoheb force-pushed the restart-2 branch 4 times, most recently from 776dafa to 94f8d71 Compare November 9, 2017 03:06

carlossg reviewed Nov 13, 2017

View reviewed changes

scoheb force-pushed the restart-2 branch from 94f8d71 to 3151b5a Compare November 13, 2017 20:29

scoheb force-pushed the restart-2 branch from ed0ca26 to 76e409b Compare November 15, 2017 17:40

jglick requested changes Nov 16, 2017

View reviewed changes

carlossg mentioned this pull request Nov 16, 2017

Add LoggerRule.recordPackage jenkinsci/jenkins-test-harness#87

Merged

scoheb force-pushed the restart-2 branch from 37aeda8 to 029f027 Compare November 17, 2017 14:11

scoheb force-pushed the restart-2 branch from 2d2e2e8 to 8f274b4 Compare November 17, 2017 15:50

Vlatombe reviewed Nov 21, 2017

View reviewed changes

scoheb added 4 commits November 21, 2017 09:19

allow re-use of existing pods for restarts

ec59d1c

- container step Includes: - update to parent pom to get latest jenkins-test-harness - update to durable-task plugin v1.16 [FIXES JENKINS-47476]

Update tests and code as per PR review comments

f463e07

fix findbugs

2d02a3b

remove namespace

4e10ca4

scoheb force-pushed the restart-2 branch from 8f274b4 to 4e10ca4 Compare November 21, 2017 14:28

Vlatombe reviewed Nov 21, 2017

View reviewed changes

Fix unused imports

2fc30fa

Vlatombe approved these changes Nov 21, 2017

View reviewed changes

carlossg merged commit 3ee086b into jenkinsci:master Nov 21, 2017


		@SuppressFBWarnings(value = "SE_TRANSIENT_FIELD_NOT_RESTORED", justification = "not needed on deserialization")
		private final transient List<Closeable> closables = new ArrayList<>();

		@@ -16,6 +16,8 @@

		package org.csanchez.jenkins.plugins.kubernetes.pipeline;

		import com.google.inject.Inject;

allow re-use of existing pods for restarts #244

allow re-use of existing pods for restarts #244

Conversation

scoheb commented Nov 3, 2017 • edited Loading

iocanel left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

carlossg left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

scoheb commented Nov 8, 2017 • edited Loading

scoheb commented Nov 9, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

carlossg commented Nov 13, 2017

scoheb commented Nov 13, 2017

carlossg commented Nov 15, 2017

scoheb commented Nov 15, 2017

scoheb commented Nov 15, 2017

jglick commented Nov 16, 2017

jglick commented Nov 16, 2017

scoheb commented Nov 16, 2017

jglick left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

scoheb commented Nov 16, 2017

scoheb commented Nov 17, 2017

scoheb commented Nov 17, 2017

scoheb commented Nov 17, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

scoheb commented Nov 21, 2017

scoheb commented Nov 21, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Vlatombe left a comment

Choose a reason for hiding this comment

carlossg commented Nov 21, 2017

chancez commented Dec 15, 2017

chancez commented Dec 18, 2017

scoheb commented Nov 3, 2017 •

edited

Loading

scoheb commented Nov 8, 2017 •

edited

Loading

scoheb commented Nov 9, 2017 •

edited

Loading